PowerModelsITD-Tutorial.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "c60a0936-4fc5-4307-8f63-613e488e2cc9",
   "metadata": {},
   "source": [
    "# Introduction to PowerModels, PowerModelsDistribution, and PowerModelsITD\n",
    "\n",
    "**Author**:\\\n",
    "Juan Ospina,\\\n",
    "Scientist,\\\n",
    "A-1 Information Systems and Modeling Group,\\\n",
    "jjospina@lanl.gov\n",
    "\n",
    "**Presented by**: David M. Fobes\n",
    "\n",
    "LA-UR-25-20073\n",
    "\n",
    "**Los Alamos National Laboratory Grid Science Winter School 2025**\n",
    "\n",
    "Welcome! This tutorial will introduce you to the basics of the [PowerModels.jl](https://github.com/lanl-ansi/PowerModels.jl) (PM), [PowerModelsDistribution.jl](https://github.com/lanl-ansi/PowerModelsDistribution.jl) (PMD) and [PowerModelsITD.jl](https://github.com/lanl-ansi/PowerModelsITD.jl) (PMITD) packages. These packages are developed using the [Julia](https://julialang.org/) programming language and are based on [JuMP](https://jump.dev/JuMP.jl/stable/), so basic knowledge is assumed.\n",
    "\n",
    "## What are these packages?\n",
    "PM, PMD, and PMTID are Julia/JuMP packages for conducting steady-state power network optimization. This means we can run (or solve) power flow (PF) and optimal power flow (OPFs) optimization problems modeled in the steady-state domain. \n",
    "\n",
    "<img src=\"https://camo.githubusercontent.com/e9dbe9703f789d305a3e0f8006fc64e9474bc19f95439dc01fecd1000e0af86c/68747470733a2f2f6c616e6c2d616e73692e6769746875622e696f2f506f7765724d6f64656c732e6a6c2f6465762f6173736574732f6c6f676f2e737667\" style=\"height:100px\" /> **_PowerModels_ (PM)** : Focuses on **Transmission** systems. Single-phase (Positive sequence), balanced systems.\n",
    "\n",
    "<img src=\"https://camo.githubusercontent.com/35dd704d4c5091e212e462fc7929432eb9e4725796f48c8cda1166243ac189d9/68747470733a2f2f6c616e6c2d616e73692e6769746875622e696f2f506f7765724d6f64656c73446973747269627574696f6e2e6a6c2f6465762f6173736574732f6c6f676f2e737667\" style=\"height:100px\" /> **_PowerModelsDistribution_ (PMD)** : Focuses on **Distribution** systems. Multiconductor (multiphase) unbalanced systems.\n",
    "\n",
    "\n",
    "<img src=\"https://camo.githubusercontent.com/d8f39ffb7e1bb8975b41e0417b9cbede693b8bfd3b731b3a5e64da73af585550/68747470733a2f2f6c616e6c2d616e73692e6769746875622e696f2f506f7765724d6f64656c734954442e6a6c2f6465762f6173736574732f6c6f676f2e737667\" style=\"height:100px\" /> **_PowerModelsITD_ (PMITD)** : Focuses on **Transmission & Distribution** systems. _Single-phase-balanced--to--multiconductor-unbalanced_ systems.\n",
    "\n",
    "## How do these packages distinguish from other tools such as OpenDSS and MATPOWER?\n",
    "\n",
    "1. **Problem Formulations**: They allow users to evaluate multiple problem formulations, for example the _AC polar_ (voltage magnitude and angles), _AC rectangular_ (real and imaginary voltage values), _DC approximation_, _SOC relaxation_, _Current-Voltage (IV) rectangular_, etc. This means that users can solve the same problem, i.e., PF or OPF, using any of these formulations. They can even develop their own formulations and compare them with the state-of-the-art without lengthy rewrites.\n",
    "2. **Problem Specifications**: Using any of these formulations, users can solve different problem specifications such as PF, OPF, Optimal Transmission Switching (ots), and Transmission Network Expansion Planning (tnep). Users can create their problem specifications and use any formulation to solve them.\n",
    "3. The **InfrastructureModels/PowerModels Ecosystem** is designed to be a common platform for evaluating emerging computational power network formulations, specifications, and algorithms.\n",
    "\n",
    "\n",
    "### (Some) Specific Differences\n",
    "\n",
    "1. **PM vs MATPOWER**: MATPOWER does not allow using different formulations (AC rectangular, IV rectangular, approximations, relaxations)\n",
    "2. **PMD vs OpenDSS**: OpenDSS only supports PF by default and is based on one formulation. If users want to run OPFs, they must add a mathematical optimization problem through other means. Also, similar to MATPOWER, OpenDSS does not allow using different formulations.\n",
    "3. **PMITD vs ???**: To the author's knowledge, there is no other (open-source) tool that allows the **co-optimization** of T&D systems (i.e., building and solving large-scale T&D system co-optimization) where distribution systems are modeled using unbalanced multiphase models. Note that there is a difference between _co-optimization and co-simulation_."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "74ffd894-c36a-4320-b229-03a2820da2c6",
   "metadata": {},
   "source": [
    "# Installing **PowerModels/PowerModelsDistribution/PowerModelsITD**\n",
    "\n",
    "1) First, you will need to install Julia.\n",
    "\n",
    "- Follow the instructions shown [here](https://julialang.org/downloads/) to install Julia in your respective operating system.\n",
    "\n",
    "2) Install PowerModels, PowerModelsDistribution, PowerModelsITD\n",
    "\n",
    "- PowerModels installation instructions can be found [here](https://lanl-ansi.github.io/PowerModels.jl/stable/).\n",
    "- PowerModelsDistribution installation instructions can be found [here](https://lanl-ansi.github.io/PowerModelsDistribution.jl/stable/installation.html).\n",
    "- PowerModelsITD installation instructions can be found [here](https://lanl-ansi.github.io/PowerModelsITD.jl/stable/installation.html#Installation-Guide).\n",
    "\n",
    "The easiest way to install all of them is to simply, 1) **start Julia**, 2) enter the **Julia REPL**, and 3) type the following command(s):\n",
    "\n",
    "```\n",
    "]add PowerModels\n",
    "```\n",
    "\n",
    "```\n",
    "]add PowerModelsDistribution\n",
    "```\n",
    "\n",
    "```\n",
    "]add PowerModelsITD\n",
    "```\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "82646468",
   "metadata": {},
   "source": [
    "In this notebook, the easiest way is to set the Julia environment to the current folder and install the packages with `Pkg`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "d919ff96",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[32m\u001b[1m  Activating\u001b[22m\u001b[39m project at `~/Library/CloudStorage/OneDrive-LosAlamosNationalLaboratory/local/working/winter-school`\n",
      "\u001b[32m\u001b[1m    Updating\u001b[22m\u001b[39m registry at `~/.julia/registries/General.toml`\n",
      "\u001b[32m\u001b[1m   Resolving\u001b[22m\u001b[39m package versions...\n",
      "\u001b[32m\u001b[1m  No Changes\u001b[22m\u001b[39m to `~/Library/CloudStorage/OneDrive-LosAlamosNationalLaboratory/local/working/winter-school/Project.toml`\n",
      "\u001b[32m\u001b[1m  No Changes\u001b[22m\u001b[39m to `~/Library/CloudStorage/OneDrive-LosAlamosNationalLaboratory/local/working/winter-school/Manifest.toml`\n",
      "\u001b[32m\u001b[1m   Resolving\u001b[22m\u001b[39m package versions...\n",
      "\u001b[32m\u001b[1m  No Changes\u001b[22m\u001b[39m to `~/Library/CloudStorage/OneDrive-LosAlamosNationalLaboratory/local/working/winter-school/Project.toml`\n",
      "\u001b[32m\u001b[1m  No Changes\u001b[22m\u001b[39m to `~/Library/CloudStorage/OneDrive-LosAlamosNationalLaboratory/local/working/winter-school/Manifest.toml`\n",
      "\u001b[32m\u001b[1m   Resolving\u001b[22m\u001b[39m package versions...\n",
      "\u001b[32m\u001b[1m  No Changes\u001b[22m\u001b[39m to `~/Library/CloudStorage/OneDrive-LosAlamosNationalLaboratory/local/working/winter-school/Project.toml`\n",
      "\u001b[32m\u001b[1m  No Changes\u001b[22m\u001b[39m to `~/Library/CloudStorage/OneDrive-LosAlamosNationalLaboratory/local/working/winter-school/Manifest.toml`\n",
      "\u001b[32m\u001b[1m   Resolving\u001b[22m\u001b[39m package versions...\n",
      "\u001b[32m\u001b[1m  No Changes\u001b[22m\u001b[39m to `~/Library/CloudStorage/OneDrive-LosAlamosNationalLaboratory/local/working/winter-school/Project.toml`\n",
      "\u001b[32m\u001b[1m  No Changes\u001b[22m\u001b[39m to `~/Library/CloudStorage/OneDrive-LosAlamosNationalLaboratory/local/working/winter-school/Manifest.toml`\n"
     ]
    }
   ],
   "source": [
    "using Pkg\n",
    "\n",
    "Pkg.activate(\".\")\n",
    "\n",
    "Pkg.add(\"PowerModels\")\n",
    "Pkg.add(\"PowerModelsDistribution\")\n",
    "Pkg.add(\"PowerModelsITD\")\n",
    "Pkg.add(\"Ipopt\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3ad7e225-c727-4993-a4ca-f65f035b7665",
   "metadata": {},
   "source": [
    "# Let's start using **PowerModels**\n",
    "\n",
    "## File Formats Supported\n",
    "\n",
    "- Matpower \".m\"\n",
    "- PTI \".raw\" files (PSS(R)E v33 specification)\n",
    "\n",
    "Let's use the `case 5`: PJM 5 Bus **Transmission System** as example. See here the original [MATPOWER file](https://github.com/lanl-ansi/PowerModels.jl/blob/master/test/data/matpower/case5.m) \n",
    "\n",
    "Using this file, let's solve an **optimal power flow (OPF)**."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7add09bb-b295-49b8-a7ef-341f746e0e94",
   "metadata": {},
   "source": [
    "### Let's import PowerModels and parse the file we want to solve"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "259698ec-cb86-4e8e-95b1-d48f3282a9a3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[35m[warn | PowerModels]: The last 5 generator cost records will be ignored due to too few generator records.\u001b[39m\n",
      "\u001b[35m[warn | PowerModels]: reversing the orientation of branch 6 (4, 3) to be consistent with other parallel branches\u001b[39m\n",
      "\u001b[35m[warn | PowerModels]: bus 3 has an unrecongized bus_type 0, updating to bus_type 2\u001b[39m\n",
      "\u001b[35m[warn | PowerModels]: the voltage setpoint on generator 4 does not match the value at bus 4\u001b[39m\n",
      "\u001b[35m[warn | PowerModels]: the voltage setpoint on generator 1 does not match the value at bus 1\u001b[39m\n",
      "\u001b[35m[warn | PowerModels]: the voltage setpoint on generator 5 does not match the value at bus 10\u001b[39m\n",
      "\u001b[35m[warn | PowerModels]: the voltage setpoint on generator 2 does not match the value at bus 1\u001b[39m\n",
      "\u001b[35m[warn | PowerModels]: the voltage setpoint on generator 3 does not match the value at bus 3\u001b[39m\n",
      "\u001b[32m[info | PowerModels]: removing 1 cost terms from generator 4: [4000.0, 0.0]\u001b[39m\n",
      "\u001b[32m[info | PowerModels]: removing 1 cost terms from generator 1: [1400.0, 0.0]\u001b[39m\n",
      "\u001b[32m[info | PowerModels]: removing 1 cost terms from generator 5: [1000.0, 0.0]\u001b[39m\n",
      "\u001b[32m[info | PowerModels]: removing 1 cost terms from generator 2: [1500.0, 0.0]\u001b[39m\n",
      "\u001b[32m[info | PowerModels]: removing 1 cost terms from generator 3: [3000.0, 0.0]\u001b[39m\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 13 entries:\n",
       "  \"bus\"            => Dict{String, Any}(\"4\"=>Dict{String, Any}(\"zone\"=>1, \"bus_…\n",
       "  \"source_type\"    => \"matpower\"\n",
       "  \"name\"           => \"case5\"\n",
       "  \"dcline\"         => Dict{String, Any}()\n",
       "  \"source_version\" => \"2\"\n",
       "  \"gen\"            => Dict{String, Any}(\"4\"=>Dict{String, Any}(\"pg\"=>0.0, \"mode…\n",
       "  \"branch\"         => Dict{String, Any}(\"4\"=>Dict{String, Any}(\"br_r\"=>0.00108,…\n",
       "  \"storage\"        => Dict{String, Any}()\n",
       "  \"switch\"         => Dict{String, Any}()\n",
       "  \"baseMVA\"        => 100.0\n",
       "  \"per_unit\"       => true\n",
       "  \"shunt\"          => Dict{String, Any}()\n",
       "  \"load\"           => Dict{String, Any}(\"1\"=>Dict{String, Any}(\"source_id\"=>Any…"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# import package\n",
    "using PowerModels\n",
    "\n",
    "# Define the path to be the PowerModels path (optional)\n",
    "files_path = joinpath(dirname(pathof(PowerModels)), \"..\")\n",
    "pm_file = joinpath(files_path, \"test/data/matpower/case5.m\")\n",
    "\n",
    "# parse the file into dictionary\n",
    "data = parse_file(pm_file)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2fa33767-c600-4273-a81e-67b45ae249ab",
   "metadata": {},
   "source": [
    "As you can see, the information gets parsed into a **Julia dictionary**. Also, we get some informational messages (i.e., `info` and `warnings`). If we want to **silence** this output we need to use the following line in the code:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "7af9ac55-a87d-4483-aa21-dcd874eb521f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32m[info | PowerModels]: Suppressing information and warning messages for the rest of this session.  Use the Memento package for more fine-grained control of logging.\u001b[39m\n"
     ]
    }
   ],
   "source": [
    "PowerModels.silence()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "acee4a4a-f836-438e-88b4-50e35ef02c06",
   "metadata": {},
   "source": [
    "Now, let's look at some of the data in the network presented in the dictionary.\n",
    "\n",
    "Let's start with the `buses` in the network. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "8547d288-a387-4f94-8e8d-04d0f62775f4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 5 entries:\n",
       "  \"4\"  => Dict{String, Any}(\"zone\"=>1, \"bus_i\"=>4, \"bus_type\"=>3, \"vmax\"=>1.1, …\n",
       "  \"1\"  => Dict{String, Any}(\"zone\"=>1, \"bus_i\"=>1, \"bus_type\"=>2, \"vmax\"=>1.1, …\n",
       "  \"10\" => Dict{String, Any}(\"zone\"=>1, \"bus_i\"=>10, \"bus_type\"=>2, \"vmax\"=>1.1,…\n",
       "  \"2\"  => Dict{String, Any}(\"zone\"=>1, \"bus_i\"=>2, \"bus_type\"=>1, \"vmax\"=>1.1, …\n",
       "  \"3\"  => Dict{String, Any}(\"zone\"=>1, \"bus_i\"=>3, \"bus_type\"=>2, \"vmax\"=>1.1, …"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[\"bus\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2911bb91-3202-4902-88e6-61ae39fcb53f",
   "metadata": {},
   "source": [
    "Let's look specifically for bus `2` data. Here, we can see the bus type, voltage max in per unit (vmax), voltage min (vmin) in per unit, and other data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "6bb11618-fb73-4970-9c9f-66c717fc3299",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 11 entries:\n",
       "  \"zone\"      => 1\n",
       "  \"bus_i\"     => 2\n",
       "  \"bus_type\"  => 1\n",
       "  \"vmax\"      => 1.1\n",
       "  \"source_id\" => Any[\"bus\", 2]\n",
       "  \"area\"      => 1\n",
       "  \"vmin\"      => 0.9\n",
       "  \"index\"     => 2\n",
       "  \"va\"        => -0.0128221\n",
       "  \"vm\"        => 1.08407\n",
       "  \"base_kv\"   => 230.0"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[\"bus\"][\"2\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "68e5df7d-5cdf-4aef-980b-59eb90038d16",
   "metadata": {},
   "source": [
    "Now, let's take a look at branches (i.e., lines) and loads"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "39549134-60af-4acc-858c-8b935f6bc947",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 7 entries:\n",
       "  \"4\" => Dict{String, Any}(\"br_r\"=>0.00108, \"rate_a\"=>4.26, \"shift\"=>0.0, \"rate…\n",
       "  \"1\" => Dict{String, Any}(\"br_r\"=>0.00281, \"rate_a\"=>4.0, \"shift\"=>0.0, \"rate_…\n",
       "  \"5\" => Dict{String, Any}(\"br_r\"=>0.00297, \"rate_a\"=>4.26, \"shift\"=>0.0174533,…\n",
       "  \"2\" => Dict{String, Any}(\"br_r\"=>0.00304, \"rate_a\"=>4.26, \"shift\"=>0.0, \"rate…\n",
       "  \"6\" => Dict{String, Any}(\"br_r\"=>0.00327443, \"rate_a\"=>4.26, \"shift\"=>0.01745…\n",
       "  \"7\" => Dict{String, Any}(\"br_r\"=>0.00297, \"rate_a\"=>2.4, \"shift\"=>0.0, \"rate_…\n",
       "  \"3\" => Dict{String, Any}(\"br_r\"=>0.00064, \"rate_a\"=>4.26, \"shift\"=>0.0, \"rate…"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[\"branch\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "09f01b7a-e496-4ee4-b762-1768b252e9ab",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 19 entries:\n",
       "  \"br_r\"        => 0.00297\n",
       "  \"rate_a\"      => 4.26\n",
       "  \"shift\"       => 0.0174533\n",
       "  \"rate_b\"      => 4.26\n",
       "  \"br_x\"        => 0.0297\n",
       "  \"rate_c\"      => 4.26\n",
       "  \"g_to\"        => 0.0\n",
       "  \"g_fr\"        => 0.0\n",
       "  \"source_id\"   => Any[\"branch\", 5]\n",
       "  \"b_fr\"        => 0.00337\n",
       "  \"f_bus\"       => 3\n",
       "  \"br_status\"   => 1\n",
       "  \"t_bus\"       => 4\n",
       "  \"b_to\"        => 0.00337\n",
       "  \"index\"       => 5\n",
       "  \"angmin\"      => -0.523599\n",
       "  \"angmax\"      => 0.523599\n",
       "  \"transformer\" => true\n",
       "  \"tap\"         => 1.05"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[\"branch\"][\"5\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "295c7f6f-b7f3-4e1e-85cd-f18ceaab9341",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 3 entries:\n",
       "  \"1\" => Dict{String, Any}(\"source_id\"=>Any[\"bus\", 2], \"load_bus\"=>2, \"status\"=…\n",
       "  \"2\" => Dict{String, Any}(\"source_id\"=>Any[\"bus\", 3], \"load_bus\"=>3, \"status\"=…\n",
       "  \"3\" => Dict{String, Any}(\"source_id\"=>Any[\"bus\", 4], \"load_bus\"=>4, \"status\"=…"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[\"load\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "aae1b519-fe72-46c9-a32e-99879f9f63e9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 6 entries:\n",
       "  \"source_id\" => Any[\"bus\", 4]\n",
       "  \"load_bus\"  => 4\n",
       "  \"status\"    => 1\n",
       "  \"qd\"        => 1.3147\n",
       "  \"pd\"        => 4.0\n",
       "  \"index\"     => 3"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[\"load\"][\"3\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4fad3cb2-7269-48de-be56-5b593c335132",
   "metadata": {},
   "source": [
    "We can also use the `print_summary` function (_only available in PowerModels_) to print a table-like summary of the network data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "293a8105-7932-4d93-8f31-6a4e7ff9c15a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1mMetadata\u001b[0m\n",
      "  baseMVA: 100.000\n",
      "  name: case5\n",
      "  per_unit: true\n",
      "  source_type: matpower\n",
      "  source_version: 2\n",
      "\n",
      "\u001b[1mTable Counts\u001b[0m\n",
      "  bus: 5\n",
      "  load: 3\n",
      "  shunt: 0\n",
      "  gen: 5\n",
      "  storage: 0\n",
      "  switch: 0\n",
      "  branch: 7\n",
      "  dcline: 0\n",
      "\n",
      "\n",
      "\u001b[1mTable: bus\u001b[0m\n",
      "         bus_i, bus_type,    vm,     va\n",
      "      1:     1,        2, 1.000,  0.049\n",
      "      2:     2,        1, 1.084, -0.013\n",
      "      3:     3,        2, 1.000, -0.010\n",
      "      4:     4,        3, 1.000,  0.000\n",
      "     10:    10,        2, 1.000,  0.063\n",
      "\n",
      "  default values:\n",
      "    base_kv: 230.000\n",
      "    vmin: 0.900\n",
      "    vmax: 1.100\n",
      "    area: 1\n",
      "    source_id: [(2)]\n",
      "    zone: 1\n",
      "\n",
      "\n",
      "\u001b[1mTable: load\u001b[0m\n",
      "         load_bus,    pd,    qd\n",
      "      1:        2, 3.000, 0.986\n",
      "      2:        3, 3.000, 0.986\n",
      "      3:        4, 4.000, 1.315\n",
      "\n",
      "  default values:\n",
      "    status: 1\n",
      "    source_id: [(2)]\n",
      "\n",
      "\n",
      "\u001b[1mTable: gen\u001b[0m\n",
      "         gen_bus,    pg,     qg,    vg,  pmax,   qmin,  qmax\n",
      "      1:       1, 0.400,  0.300, 1.078, 0.400, -0.300, 0.300\n",
      "      2:       1, 1.700,  1.275, 1.078, 1.700, -1.275, 1.275\n",
      "      3:       3, 3.245,  3.900, 1.100, 5.200, -3.900, 3.900\n",
      "      4:       4, 0.000, -0.108, 1.064, 2.000, -1.500, 1.500\n",
      "      5:      10, 4.707, -1.650, 1.069, 6.000, -4.500, 4.500\n",
      "\n",
      "  default values:\n",
      "    mbase: 100.000\n",
      "    pmin: 0.000\n",
      "    gen_status: 1\n",
      "    model: 2\n",
      "    ncost: 2\n",
      "    cost: [(2)]\n",
      "    startup: 0.000\n",
      "    shutdown: 0.000\n",
      "    source_id: [(2)]\n",
      "\n",
      "\n",
      "\u001b[1mTable: branch\u001b[0m\n",
      "         f_bus, t_bus,  br_r,  br_x,  b_fr,  b_to,   tap, shift, rate_a, rate_b, rate_c, transformer\n",
      "      1:     1,     2, 0.003, 0.028, 0.004, 0.004, 1.000, 0.000,  4.000,  4.000,  4.000,       false\n",
      "      2:     1,     4, 0.003, 0.030, 0.003, 0.003, 1.000, 0.000,  4.260,  4.260,  4.260,       false\n",
      "      3:     1,    10, 0.001, 0.006, 0.016, 0.016, 1.000, 0.000,  4.260,  4.260,  4.260,       false\n",
      "      4:     2,     3, 0.001, 0.011, 0.009, 0.009, 1.000, 0.000,  4.260,  4.260,  4.260,       false\n",
      "      5:     3,     4, 0.003, 0.030, 0.003, 0.003, 1.050, 0.017,  4.260,  4.260,  4.260,        true\n",
      "      6:     3,     4, 0.003, 0.033, 0.003, 0.004, 0.952, 0.017,  4.260,  4.260,  4.260,        true\n",
      "      7:     4,    10, 0.003, 0.030, 0.003, 0.003, 1.000, 0.000,  2.400,  2.400,  2.400,       false\n",
      "\n",
      "  default values:\n",
      "    g_fr: 0.000\n",
      "    g_to: 0.000\n",
      "    br_status: 1\n",
      "    angmax: 0.524\n",
      "    angmin: -0.524\n",
      "    source_id: [(2)]\n"
     ]
    }
   ],
   "source": [
    "PowerModels.print_summary(data)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7a977319-2d0c-4ffa-a1d8-fe3082c23be7",
   "metadata": {},
   "source": [
    "We can modify the `data` dictionary by just modifying the values inside the dictionary, for example:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "1ed1f45e-827a-4f28-8925-7123b65f1e46",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 6 entries:\n",
       "  \"source_id\" => Any[\"bus\", 4]\n",
       "  \"load_bus\"  => 4\n",
       "  \"status\"    => 1\n",
       "  \"qd\"        => 1.3147\n",
       "  \"pd\"        => 4.0\n",
       "  \"index\"     => 3"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Let's see the old value\n",
    "data[\"load\"][\"3\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "70ef9371-0220-4ded-9c2e-5df5581434e6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3.5"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Modify pd (active power)\n",
    "data[\"load\"][\"3\"][\"pd\"] = 3.5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "0619af0d-bc5e-4de4-9ff2-6bb67505fb75",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 6 entries:\n",
       "  \"source_id\" => Any[\"bus\", 4]\n",
       "  \"load_bus\"  => 4\n",
       "  \"status\"    => 1\n",
       "  \"qd\"        => 1.3147\n",
       "  \"pd\"        => 3.5\n",
       "  \"index\"     => 3"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Let's see the new value\n",
    "data[\"load\"][\"3\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5937d1fc-945b-483a-8a8b-73b721c0b7c6",
   "metadata": {},
   "source": [
    "### Now, let's solve the Optimal Power Flow (OPF) for this problem \n",
    "\n",
    "**But**, before solving this optimization problem, we need to pick a `solver`. The solver you choose will depend on the type of problem you want to solve. For example, solving **mixed integer** problems requires a different solver than solving **continuous nonlinear** problems.\n",
    "\n",
    "In this case, we are interested in solving a Continous **Nonlinear Program (NLP)**, so let's use `Ipopt`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "535fafed-2e34-48fe-b960-abd7c7f7b0ae",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "MathOptInterface.OptimizerWithAttributes(Ipopt.Optimizer, Pair{MathOptInterface.AbstractOptimizerAttribute, Any}[MathOptInterface.RawOptimizerAttribute(\"print_level\") => 3])"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# import solver package\n",
    "using Ipopt\n",
    "\n",
    "# define Ipopt solver and options/attributes\n",
    "ipopt_solver = optimizer_with_attributes(Ipopt.Optimizer, \"print_level\" => 3)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "34e8a5c9-ab8c-4226-b513-3e94777f08f8",
   "metadata": {},
   "source": [
    "We can then use this **solver** and the already **parsed file** to solve the **OPF** problem. We will use the `ACP` (AC Polar) formulation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "5b14edcb-3829-4a5e-bb28-f32333f9aa00",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "******************************************************************************\n",
      "This program contains Ipopt, a library for large-scale nonlinear optimization.\n",
      " Ipopt is released as open source code under the Eclipse Public License (EPL).\n",
      "         For more information visit https://github.com/coin-or/Ipopt\n",
      "******************************************************************************\n",
      "\n",
      "Total number of variables............................:       48\n",
      "                     variables with only lower bounds:        0\n",
      "                variables with lower and upper bounds:       43\n",
      "                     variables with only upper bounds:        0\n",
      "Total number of equality constraints.................:       39\n",
      "Total number of inequality constraints...............:       20\n",
      "        inequality constraints with only lower bounds:        0\n",
      "   inequality constraints with lower and upper bounds:        6\n",
      "        inequality constraints with only upper bounds:       14\n",
      "\n",
      "\n",
      "Number of Iterations....: 24\n",
      "\n",
      "                                   (scaled)                 (unscaled)\n",
      "Objective...............:   4.1259617215094033e+02    1.6503846886037612e+04\n",
      "Dual infeasibility......:   1.0604012696469439e-11    4.2416050785877754e-10\n",
      "Constraint violation....:   8.4376949871511897e-15    8.4376949871511897e-15\n",
      "Variable bound violation:   1.9635816972396469e-08    1.9635816972396469e-08\n",
      "Complementarity.........:   2.5059410794473296e-09    1.0023764317789317e-07\n",
      "Overall NLP error.......:   2.5059410794473296e-09    1.0023764317789317e-07\n",
      "\n",
      "\n",
      "Number of objective function evaluations             = 25\n",
      "Number of objective gradient evaluations             = 25\n",
      "Number of equality constraint evaluations            = 25\n",
      "Number of inequality constraint evaluations          = 25\n",
      "Number of equality constraint Jacobian evaluations   = 25\n",
      "Number of inequality constraint Jacobian evaluations = 25\n",
      "Number of Lagrangian Hessian evaluations             = 24\n",
      "Total seconds in IPOPT                               = 1.046\n",
      "\n",
      "EXIT: Optimal Solution Found.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 8 entries:\n",
       "  \"solve_time\"         => 1.13196\n",
       "  \"optimizer\"          => \"Ipopt\"\n",
       "  \"termination_status\" => LOCALLY_SOLVED\n",
       "  \"dual_status\"        => FEASIBLE_POINT\n",
       "  \"primal_status\"      => FEASIBLE_POINT\n",
       "  \"objective\"          => 16503.8\n",
       "  \"solution\"           => Dict{String, Any}(\"baseMVA\"=>100.0, \"branch\"=>Dict{St…\n",
       "  \"objective_lb\"       => -Inf"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result = solve_opf(data, ACPPowerModel, ipopt_solver)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8ec26041-65fa-4007-9dc9-0959faa62902",
   "metadata": {},
   "source": [
    "The result dictionary contains useful information about the optimization problem solved, including the **objective value** and **wall clock solve time**. Detailed documentation of the result dictionary is available [here](https://lanl-ansi.github.io/PowerModels.jl/stable/result-data/).\n",
    "\n",
    "Let's check out the solution data dictionary. This is where the actual result of the OPF is stored."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "1f7a6129-4aaa-46a6-9085-22c600e9444f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 7 entries:\n",
       "  \"baseMVA\"             => 100.0\n",
       "  \"branch\"              => Dict{String, Any}(\"4\"=>Dict{String, Any}(\"qf\"=>-1.22…\n",
       "  \"gen\"                 => Dict{String, Any}(\"4\"=>Dict{String, Any}(\"qg\"=>-0.12…\n",
       "  \"multiinfrastructure\" => false\n",
       "  \"multinetwork\"        => false\n",
       "  \"bus\"                 => Dict{String, Any}(\"4\"=>Dict{String, Any}(\"va\"=>4.279…\n",
       "  \"per_unit\"            => true"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result[\"solution\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a3537b32-d2b1-495b-9338-5387142717ae",
   "metadata": {},
   "source": [
    "Let's use the function `print_summary`,"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "f4854770-a64d-486b-b1d7-d2180990b55f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1mMetadata\u001b[0m\n",
      "  baseMVA: 100.000\n",
      "  multiinfrastructure: false\n",
      "  multinetwork: false\n",
      "  per_unit: true\n",
      "\n",
      "\u001b[1mTable Counts\u001b[0m\n",
      "  bus: 5\n",
      "  gen: 5\n",
      "  branch: 7\n",
      "\n",
      "\n",
      "\u001b[1mTable: bus\u001b[0m\n",
      "            vm,     va\n",
      "      1: 0.913,  0.070\n",
      "      2: 0.911, -0.008\n",
      "      3: 0.926, -0.000\n",
      "      4: 0.900,  0.000\n",
      "     10: 0.905,  0.088\n",
      "\n",
      "\n",
      "\u001b[1mTable: gen\u001b[0m\n",
      "             pg,     qg,  pg_cost\n",
      "      1:  0.400,  0.300,  560.000\n",
      "      2:  1.700,  1.275, 2550.000\n",
      "      3:  2.958,  3.900, 8873.192\n",
      "      4: -0.000, -0.122,   -0.000\n",
      "      5:  4.521, -1.352, 4520.655\n",
      "\n",
      "\n",
      "\u001b[1mTable: branch\u001b[0m\n",
      "             pf,     qf,     pt,     qt\n",
      "      1:  2.295, -0.066, -2.278,  0.238\n",
      "      2:  1.920,  0.259, -1.907, -0.127\n",
      "      3: -2.116,  1.382,  2.121, -1.359\n",
      "      4: -0.722, -1.224,  0.725,  1.234\n",
      "      5: -0.516, -0.489,  0.518,  0.502\n",
      "      6: -0.251,  2.168,  0.267, -2.008\n",
      "      7: -2.379,  0.197,  2.400,  0.007\n"
     ]
    }
   ],
   "source": [
    "PowerModels.print_summary(result[\"solution\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1c8c0dae-d63e-44b5-9047-b39f41bd403f",
   "metadata": {},
   "source": [
    "\n",
    "\n",
    "You can also see the results manually by just accessing the specific keys in the `solution` dictionary.\n",
    "\n",
    "For example, let's access the solution for the `buses` and the `branches`. All results are in **per-unit**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "780f9fe4-e2b7-4d91-952e-96a0bec552a8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 5 entries:\n",
       "  \"4\"  => Dict{String, Any}(\"va\"=>4.27964e-37, \"vm\"=>0.9)\n",
       "  \"1\"  => Dict{String, Any}(\"va\"=>0.0701399, \"vm\"=>0.912897)\n",
       "  \"10\" => Dict{String, Any}(\"va\"=>0.0876162, \"vm\"=>0.904738)\n",
       "  \"2\"  => Dict{String, Any}(\"va\"=>-0.00775197, \"vm\"=>0.910533)\n",
       "  \"3\"  => Dict{String, Any}(\"va\"=>-5.65684e-5, \"vm\"=>0.925845)"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result[\"solution\"][\"bus\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "13150100-9773-49cf-9cc4-f1f20071ef69",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 7 entries:\n",
       "  \"4\" => Dict{String, Any}(\"qf\"=>-1.22401, \"qt\"=>1.23447, \"pt\"=>0.724912, \"pf\"=…\n",
       "  \"1\" => Dict{String, Any}(\"qf\"=>-0.06603, \"qt\"=>0.237913, \"pt\"=>-2.27769, \"pf\"…\n",
       "  \"5\" => Dict{String, Any}(\"qf\"=>-0.488503, \"qt\"=>0.502359, \"pt\"=>0.518347, \"pf…\n",
       "  \"2\" => Dict{String, Any}(\"qf\"=>0.258976, \"qt\"=>-0.127376, \"pt\"=>-1.90656, \"pf…\n",
       "  \"6\" => Dict{String, Any}(\"qf\"=>2.16794, \"qt\"=>-2.00838, \"pt\"=>0.267301, \"pf\"=…\n",
       "  \"7\" => Dict{String, Any}(\"qf\"=>0.196768, \"qt\"=>0.00673953, \"pt\"=>2.39999, \"pf…\n",
       "  \"3\" => Dict{String, Any}(\"qf\"=>1.38205, \"qt\"=>-1.35855, \"pt\"=>2.12066, \"pf\"=>…"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result[\"solution\"][\"branch\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "3bc28b69-e193-4809-b8a7-c702a2aba1bd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 4 entries:\n",
       "  \"qf\" => 2.16794\n",
       "  \"qt\" => -2.00838\n",
       "  \"pt\" => 0.267301\n",
       "  \"pf\" => -0.250755"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result[\"solution\"][\"branch\"][\"6\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c4a045f3-8ee5-484d-8b2b-0daa4c1a1d47",
   "metadata": {},
   "source": [
    "In the result `solution` dictionary we can see things like the **active** and **reactive** power flowing through the branches, the **voltage magnitudes and angles** of the buses, and the **active and reactive power dispatch** from each generating unit. \n",
    "\n",
    "Let's us try now to solve the same problem using the **DC polar linear approximation** (`DCPPowerModel`)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "7d195e8f-f1e8-4131-94be-239041855562",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total number of variables............................:       17\n",
      "                     variables with only lower bounds:        0\n",
      "                variables with lower and upper bounds:       12\n",
      "                     variables with only upper bounds:        0\n",
      "Total number of equality constraints.................:       13\n",
      "Total number of inequality constraints...............:        6\n",
      "        inequality constraints with only lower bounds:        0\n",
      "   inequality constraints with lower and upper bounds:        6\n",
      "        inequality constraints with only upper bounds:        0\n",
      "\n",
      "\n",
      "Number of Iterations....: 14\n",
      "\n",
      "                                   (scaled)                 (unscaled)\n",
      "Objective...............:   3.9597181813173097e+02    1.5838872725269239e+04\n",
      "Dual infeasibility......:   1.0540670187408245e-13    4.2162680749632982e-12\n",
      "Constraint violation....:   4.4408920985006262e-16    4.4408920985006262e-16\n",
      "Variable bound violation:   2.3981676200435231e-08    2.3981676200435231e-08\n",
      "Complementarity.........:   2.5076932351238464e-09    1.0030772940495385e-07\n",
      "Overall NLP error.......:   2.5076932351238464e-09    1.0030772940495385e-07\n",
      "\n",
      "\n",
      "Number of objective function evaluations             = 15\n",
      "Number of objective gradient evaluations             = 15\n",
      "Number of equality constraint evaluations            = 15\n",
      "Number of inequality constraint evaluations          = 15\n",
      "Number of equality constraint Jacobian evaluations   = 1\n",
      "Number of inequality constraint Jacobian evaluations = 1\n",
      "Number of Lagrangian Hessian evaluations             = 1\n",
      "Total seconds in IPOPT                               = 0.003\n",
      "\n",
      "EXIT: Optimal Solution Found.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 8 entries:\n",
       "  \"solve_time\"         => 0.0030179\n",
       "  \"optimizer\"          => \"Ipopt\"\n",
       "  \"termination_status\" => LOCALLY_SOLVED\n",
       "  \"dual_status\"        => FEASIBLE_POINT\n",
       "  \"primal_status\"      => FEASIBLE_POINT\n",
       "  \"objective\"          => 15838.9\n",
       "  \"solution\"           => Dict{String, Any}(\"baseMVA\"=>100.0, \"branch\"=>Dict{St…\n",
       "  \"objective_lb\"       => -Inf"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result = solve_opf(data, DCPPowerModel, ipopt_solver)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "435c4605-3df0-40f8-820d-c4be183574ce",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1mMetadata\u001b[0m\n",
      "  baseMVA: 100.000\n",
      "  multiinfrastructure: false\n",
      "  multinetwork: false\n",
      "  per_unit: true\n",
      "\n",
      "\u001b[1mTable Counts\u001b[0m\n",
      "  bus: 5\n",
      "  gen: 5\n",
      "  branch: 7\n",
      "\n",
      "\n",
      "\u001b[1mTable: bus\u001b[0m\n",
      "             va\n",
      "      1:  0.057\n",
      "      2: -0.016\n",
      "      3: -0.012\n",
      "      4:  0.000\n",
      "     10:  0.072\n",
      "\n",
      "  default values:\n",
      "    vm: 1.000\n",
      "\n",
      "\n",
      "\u001b[1mTable: gen\u001b[0m\n",
      "             pg,  pg_cost\n",
      "      1:  0.400,  560.000\n",
      "      2:  1.700, 2550.000\n",
      "      3:  2.664, 7993.309\n",
      "      4: -0.000,   -0.000\n",
      "      5:  4.736, 4735.564\n",
      "\n",
      "  default values:\n",
      "    qg: NaN0000\n",
      "\n",
      "\n",
      "\u001b[1mTable: branch\u001b[0m\n",
      "             pf,     pt\n",
      "      1:  2.583, -2.583\n",
      "      2:  1.853, -1.853\n",
      "      3: -2.336,  2.336\n",
      "      4: -0.417,  0.417\n",
      "      5: -0.395,  0.395\n",
      "      6: -0.358,  0.358\n",
      "      7: -2.400,  2.400\n",
      "\n",
      "  default values:\n",
      "    qf: NaN0000\n",
      "    qt: NaN0000\n"
     ]
    }
   ],
   "source": [
    "PowerModels.print_summary(result[\"solution\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b1041ebc-fbf3-4bdf-873a-ca0de6868298",
   "metadata": {},
   "source": [
    "As seen, in these results there are no `vm`, `qf` or `qt`. This is because, as you may remember from Power Lectures 101, the DC linear approximation formulation assumes (or approximates) all `vm = 1.0` and do not considers reactive power."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4d927fc0-08af-4c1d-95c6-ab3d3074597a",
   "metadata": {},
   "source": [
    "### -----------------------------------------------------------------------------------------------------------\n",
    "### QUIZ #1: let's solve the Optimal Power Flow (OPF)  using different formulations and compute optimality gaps.\n",
    "\n",
    "Similar to what we have done, now let's solve the OPF using other formulations:\n",
    "\n",
    "- `ACRPowerModel` (NLP) -- AC rectangular coordinates\n",
    "- `SOCWRPowerModel` (RELAXATION) -- Second-order cone W rectangular coordinates\n",
    "- `NFAPowerModel` (APPROXIMATION) -- Network flow approximation\n",
    "- `DCPPowerModel` (APPROXIMATION) -- DC polar approximation\n",
    "\n",
    "**Definition of _optimality gap_**: the difference between the objective value of the best feasible solution currently found (local optima for NLPs) and the optimal objective value of the problem you are testing.\n",
    "\n",
    "**Note**: the optimality gap is is normally used to compare NLPs with approximations/relaxations. NLPs vs NLPs should give a value close to 0.0%\n",
    "\n",
    "**Note 2**: Make sure that the solver terminates nominally!, i.e., `result[\"termination_status\"] == OPTIMAL` ||  `result[\"termination_status\"] == LOCALLY_SOLVED`\n",
    "\n",
    "To compare the optimality gap using the `AC polar` as baseline, use the following function:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "578c98aa-09c0-4760-a406-45fcb9033e0a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "optimality_gap (generic function with 1 method)"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "function optimality_gap(nlp_result, relax_result)\n",
    "    # by convention optimality gaps are given as a percentage\n",
    "    gap = 100*(nlp_result[\"objective\"] - relax_result[\"objective\"])/nlp_result[\"objective\"]\n",
    "    return gap\n",
    "end"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c9277201-6662-4dd4-bfb7-aaa9f64d4fee",
   "metadata": {},
   "source": [
    "I will provide an example below between **AC Polar vs. AC Rectangular**\n",
    "\n",
    "**Hint**: To compare against other formulations, change the type of formulation in the line `result_relax` to solve for the formulation you want to compare against the `ACPPowerModel` formulation, and _re-run_ the notebook cells."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "c2ef617b-bc00-4175-8940-57ad62cf6e1b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total number of variables............................:       48\n",
      "                     variables with only lower bounds:        0\n",
      "                variables with lower and upper bounds:       43\n",
      "                     variables with only upper bounds:        0\n",
      "Total number of equality constraints.................:       39\n",
      "Total number of inequality constraints...............:       20\n",
      "        inequality constraints with only lower bounds:        0\n",
      "   inequality constraints with lower and upper bounds:        6\n",
      "        inequality constraints with only upper bounds:       14\n",
      "\n",
      "\n",
      "Number of Iterations....: 24\n",
      "\n",
      "                                   (scaled)                 (unscaled)\n",
      "Objective...............:   4.1259617215094033e+02    1.6503846886037612e+04\n",
      "Dual infeasibility......:   1.0604012696469439e-11    4.2416050785877754e-10\n",
      "Constraint violation....:   8.4376949871511897e-15    8.4376949871511897e-15\n",
      "Variable bound violation:   1.9635816972396469e-08    1.9635816972396469e-08\n",
      "Complementarity.........:   2.5059410794473296e-09    1.0023764317789317e-07\n",
      "Overall NLP error.......:   2.5059410794473296e-09    1.0023764317789317e-07\n",
      "\n",
      "\n",
      "Number of objective function evaluations             = 25\n",
      "Number of objective gradient evaluations             = 25\n",
      "Number of equality constraint evaluations            = 25\n",
      "Number of inequality constraint evaluations          = 25\n",
      "Number of equality constraint Jacobian evaluations   = 25\n",
      "Number of inequality constraint Jacobian evaluations = 25\n",
      "Number of Lagrangian Hessian evaluations             = 24\n",
      "Total seconds in IPOPT                               = 0.005\n",
      "\n",
      "EXIT: Optimal Solution Found.\n",
      "----------------------------------------------------------------------------------------\n",
      "----------------------------------------------------------------------------------------\n",
      "Total number of variables............................:       48\n",
      "                     variables with only lower bounds:        0\n",
      "                variables with lower and upper bounds:       48\n",
      "                     variables with only upper bounds:        0\n",
      "Total number of equality constraints.................:       39\n",
      "Total number of inequality constraints...............:       31\n",
      "        inequality constraints with only lower bounds:        6\n",
      "   inequality constraints with lower and upper bounds:        5\n",
      "        inequality constraints with only upper bounds:       20\n",
      "\n",
      "\n",
      "Number of Iterations....: 21\n",
      "\n",
      "                                   (scaled)                 (unscaled)\n",
      "Objective...............:   4.1259617224913086e+02    1.6503846889965233e+04\n",
      "Dual infeasibility......:   1.0608287236644614e-11    4.2433148946578457e-10\n",
      "Constraint violation....:   1.2301440189066962e-14    2.3647750424515834e-14\n",
      "Variable bound violation:   1.9635816528307259e-08    1.9635816528307259e-08\n",
      "Complementarity.........:   2.5059410879254664e-09    1.0023764351701866e-07\n",
      "Overall NLP error.......:   2.5059410879254664e-09    1.0023764351701866e-07\n",
      "\n",
      "\n",
      "Number of objective function evaluations             = 22\n",
      "Number of objective gradient evaluations             = 22\n",
      "Number of equality constraint evaluations            = 22\n",
      "Number of inequality constraint evaluations          = 22\n",
      "Number of equality constraint Jacobian evaluations   = 22\n",
      "Number of inequality constraint Jacobian evaluations = 22\n",
      "Number of Lagrangian Hessian evaluations             = 21\n",
      "Total seconds in IPOPT                               = 0.005\n",
      "\n",
      "EXIT: Optimal Solution Found.\n",
      "----------------------------------------------------------------------------------------\n",
      "----------------------------------------------------------------------------------------\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "-2.3798209925953233e-8"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Baseline nlp to compare\n",
    "result_nlp = solve_opf(data, ACPPowerModel, ipopt_solver)\n",
    "\n",
    "println(\"----------------------------------------------------------------------------------------\")\n",
    "println(\"----------------------------------------------------------------------------------------\")\n",
    "\n",
    "# Approx. or Relax. to compare (in this case is another nlp formulation)\n",
    "result_relax = solve_opf(data, ACRPowerModel, ipopt_solver)\n",
    "\n",
    "println(\"----------------------------------------------------------------------------------------\")\n",
    "println(\"----------------------------------------------------------------------------------------\")\n",
    "\n",
    "# Calculate optimality gap\n",
    "op_gap = optimality_gap(result_nlp, result_relax) # result is in %\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "2e93c9be-1b0b-4357-99cc-d09e203e173c",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mInfo: \u001b[22m\u001b[39mNLP: 16503.846886037612\n",
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mInfo: \u001b[22m\u001b[39mNLP: LOCALLY_SOLVED\n",
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mInfo: \u001b[22m\u001b[39mRELAX: 16503.846889965233\n",
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mInfo: \u001b[22m\u001b[39mRELAX: LOCALLY_SOLVED\n",
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mInfo: \u001b[22m\u001b[39mOptimality Gap = -2.3798209925953233e-8 %\n"
     ]
    }
   ],
   "source": [
    "@info \"NLP: $(result_nlp[\"objective\"])\"\n",
    "@info \"NLP: $(result_nlp[\"termination_status\"])\"\n",
    "\n",
    "@info \"RELAX: $(result_relax[\"objective\"])\"\n",
    "@info \"RELAX: $(result_relax[\"termination_status\"])\"\n",
    "\n",
    "@info \"Optimality Gap = $(op_gap) %\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cdca7e37-1047-4c95-a205-7cd658507f2b",
   "metadata": {},
   "source": [
    "# Tutorial **PowerModelsDistribution**"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e9964fd0-f447-44ac-a52d-b3c82094d941",
   "metadata": {},
   "source": [
    "`PowerModelsDistribution` is designed to do the same 'thing' that `PowerModels` does but for **multiconductor unbalanced distribution systems**.\n",
    "\n",
    "Due to requiring more details for modeling multiconductor unbalanced systems, PMD makes use of an [**ENGINEERING MODEL**](https://lanl-ansi.github.io/PowerModelsDistribution.jl/stable/manual/eng-data-model.html) and a [**MATHEMATICAL MODEL**](https://lanl-ansi.github.io/PowerModelsDistribution.jl/stable/manual/math-model.html).\n",
    "\n",
    "In short, the engineering model is designed to provide a more 'engineering' modeling perspective, so it is easier to read and modify by the user, while the mathematical model is what is used to create the JuMP mathematical optimization model. We encourage readers to read the [PowerModelsDistribution Documentation](https://lanl-ansi.github.io/PowerModelsDistribution.jl/stable/index.html) to get a better perspective on PMD functionality. \n",
    "\n",
    "For this tutorial, we will only focus on the basics, i.e., how to run an OPF, similar to what we did for PowerModels. In this case, let's put everything together, to avoid repetitions.\n",
    "\n",
    "For more detailed tutorials and guides, check out:\n",
    "\n",
    "- [Quick Start Guide](https://lanl-ansi.github.io/PowerModelsDistribution.jl/stable/manual/quickguide.html)\n",
    "- [Beginners Guide](https://lanl-ansi.github.io/PowerModelsDistribution.jl/stable/tutorials/Beginners%20Guide.html)\n",
    "\n",
    "**PowerModelsDistribution** uses the file format used in OpenDSS, i.e., `OpenDSS (\".dss\")`. You can find the `.dss` file we use in this example [here](https://github.com/lanl-ansi/PowerModelsDistribution.jl/blob/main/test/data/opendss/case3_balanced.dss)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "4dbd3d7d-07ae-4e63-8d7b-8c27ed5a3465",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING: using PowerModelsDistribution.parse_file in module Main conflicts with an existing identifier.\n",
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mPowerModelsDistribution | Info ] : \u001b[22m\u001b[39mCircuit has been reset with the 'clear' on line 1 in 'case3_balanced.dss'\n",
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mPowerModelsDistribution | Info ] : \u001b[22m\u001b[39mCommand 'calcvoltagebases' on line 36 in 'case3_balanced.dss' is not supported, skipping.\n",
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mPowerModelsDistribution | Info ] : \u001b[22m\u001b[39mCommand 'solve' on line 38 in 'case3_balanced.dss' is not supported, skipping.\n",
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mPowerModelsDistribution | Info ] : \u001b[22m\u001b[39mbasemva=100 is the default value, you may want to adjust sbase_default for better convergence\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total number of variables............................:       63\n",
      "                     variables with only lower bounds:        9\n",
      "                variables with lower and upper bounds:        0\n",
      "                     variables with only upper bounds:        0\n",
      "Total number of equality constraints.................:       63\n",
      "Total number of inequality constraints...............:       30\n",
      "        inequality constraints with only lower bounds:        9\n",
      "   inequality constraints with lower and upper bounds:        0\n",
      "        inequality constraints with only upper bounds:       21\n",
      "\n",
      "\n",
      "Number of Iterations....: 3\n",
      "\n",
      "                                   (scaled)                 (unscaled)\n",
      "Objective...............:   1.8344980704278438e-02    1.8344980704278438e-02\n",
      "Dual infeasibility......:   1.4224867400962002e-08    1.4224867400962002e-08\n",
      "Constraint violation....:   1.5674554394954330e-14    5.0076175844848692e-13\n",
      "Variable bound violation:   0.0000000000000000e+00    0.0000000000000000e+00\n",
      "Complementarity.........:   0.0000000000000000e+00    0.0000000000000000e+00\n",
      "Overall NLP error.......:   8.6864427797129750e-09    1.4224867400962002e-08\n",
      "\n",
      "\n",
      "Number of objective function evaluations             = 4\n",
      "Number of objective gradient evaluations             = 4\n",
      "Number of equality constraint evaluations            = 4\n",
      "Number of inequality constraint evaluations          = 4\n",
      "Number of equality constraint Jacobian evaluations   = 4\n",
      "Number of inequality constraint Jacobian evaluations = 4\n",
      "Number of Lagrangian Hessian evaluations             = 3\n",
      "Total seconds in IPOPT                               = 0.042\n",
      "\n",
      "EXIT: Optimal Solution Found.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[33m\u001b[1m┌ \u001b[22m\u001b[39m\u001b[33m\u001b[1mPowerModelsDistribution | Warning ] : \u001b[22m\u001b[39mbasefreq=60.0 on line.quad does not match circuit basefreq=50.0\n",
      "\u001b[33m\u001b[1m└ \u001b[22m\u001b[39m\u001b[90m@ PowerModelsDistribution ~/.julia/packages/PowerModelsDistribution/EOTxp/src/data_model/transformations/dss2eng.jl:422\u001b[39m\n",
      "\u001b[33m\u001b[1m┌ \u001b[22m\u001b[39m\u001b[33m\u001b[1mPowerModelsDistribution | Warning ] : \u001b[22m\u001b[39mbasefreq=60.0 on line.ohline does not match circuit basefreq=50.0\n",
      "\u001b[33m\u001b[1m└ \u001b[22m\u001b[39m\u001b[90m@ PowerModelsDistribution ~/.julia/packages/PowerModelsDistribution/EOTxp/src/data_model/transformations/dss2eng.jl:422\u001b[39m\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 8 entries:\n",
       "  \"solve_time\"         => 0.0433922\n",
       "  \"optimizer\"          => \"Ipopt\"\n",
       "  \"termination_status\" => LOCALLY_SOLVED\n",
       "  \"dual_status\"        => FEASIBLE_POINT\n",
       "  \"primal_status\"      => FEASIBLE_POINT\n",
       "  \"objective\"          => 0.018345\n",
       "  \"solution\"           => Dict{String, Any}(\"voltage_source\"=>Dict{String, Any}…\n",
       "  \"objective_lb\"       => -Inf"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# import package\n",
    "using PowerModelsDistribution\n",
    "\n",
    "# import solver package\n",
    "using Ipopt\n",
    "\n",
    "# solver attributes/options\n",
    "ipopt_solver = optimizer_with_attributes(Ipopt.Optimizer, \"print_level\" => 3)\n",
    "\n",
    "# define the path to be the PowerModelsDistribution path (optional)\n",
    "files_path = joinpath(dirname(pathof(PowerModelsDistribution)), \"..\")\n",
    "pmd_file = joinpath(files_path, \"test/data/opendss/case3_balanced.dss\")\n",
    "\n",
    "# parse the file into dictionary\n",
    "data_pmd = PowerModelsDistribution.parse_file(pmd_file)\n",
    "\n",
    "# solve the multi-conductor (mc) OPF\n",
    "result_pmd = PowerModelsDistribution.solve_mc_opf(data_pmd, ACPUPowerModel, ipopt_solver)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e1ac885d-a1c4-432d-aded-bc4628d304f1",
   "metadata": {},
   "source": [
    "The result `solution` dictionary has a very similar structure as the one in `PowerModels`, but for a multiphase network. You can access this information through the dictionary keys as follows:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "93edd612-d10e-45d6-a3ad-552e77947a89",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 6 entries:\n",
       "  \"voltage_source\" => Dict{String, Any}(\"source\"=>Dict{String, Any}(\"qg_bus\"=>[…\n",
       "  \"line\"           => Dict{String, Any}(\"quad\"=>Dict{String, Any}(\"qf\"=>[3.0345…\n",
       "  \"settings\"       => Dict{String, Any}(\"sbase\"=>100000.0)\n",
       "  \"load\"           => Dict{String, Any}(\"l2\"=>Dict{String, Any}(\"qd_bus\"=>[3.0]…\n",
       "  \"bus\"            => Dict{String, Any}(\"primary\"=>Dict{String, Any}(\"va\"=>[-0.…\n",
       "  \"per_unit\"       => false"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result_pmd[\"solution\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "9746057d-4dff-4a0a-abfa-d65dc0ee9ba9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 3 entries:\n",
       "  \"primary\"   => Dict{String, Any}(\"va\"=>[-0.0329191, -120.033, 119.967], \"vm\"=…\n",
       "  \"sourcebus\" => Dict{String, Any}(\"va\"=>[-8.99303e-7, -120.0, 120.0], \"vm\"=>[0…\n",
       "  \"loadbus\"   => Dict{String, Any}(\"va\"=>[-0.0663874, -120.066, 119.934], \"vm\"=…"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result_pmd[\"solution\"][\"bus\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "8f551ce8-b17c-4086-8307-d9bafc6327f2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 3 entries:\n",
       "  \"l2\" => Dict{String, Any}(\"qd_bus\"=>[3.0], \"pd_bus\"=>[6.0], \"qd\"=>[3.0], \"pd\"…\n",
       "  \"l3\" => Dict{String, Any}(\"qd_bus\"=>[3.0], \"pd_bus\"=>[6.0], \"qd\"=>[3.0], \"pd\"…\n",
       "  \"l1\" => Dict{String, Any}(\"qd_bus\"=>[3.0], \"pd_bus\"=>[6.0], \"qd\"=>[3.0], \"pd\"…"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result_pmd[\"solution\"][\"load\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "fcc83ad9-114d-46e3-bec7-228c9a4afaf0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 2 entries:\n",
       "  \"quad\"   => Dict{String, Any}(\"qf\"=>[3.03456, 3.03456, 3.03456], \"qt\"=>[-3.0,…\n",
       "  \"ohline\" => Dict{String, Any}(\"qf\"=>[3.06468, 3.06468, 3.06468], \"qt\"=>[-3.03…"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result_pmd[\"solution\"][\"line\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "964a801e-ad3a-4fbb-a7eb-8b0730e9c1d3",
   "metadata": {},
   "source": [
    "### -----------------------------------------------------------------------------------------------------------\n",
    "### QUIZ #2: let's solve the Optimal Power Flow (OPF)  using different formulations and compute optimality gaps.\n",
    "\n",
    "Let's solve the OPF using other formulations:\n",
    "\n",
    "- `ACRUPowerModel` (NLP) -- AC rectangular coordinates\n",
    "- `LinDist3FlowPowerModel` (APPROXIMATION) -- LindDist3Flow approximation\n",
    "- `NFAUPowerModel` (APPROXIMATION) -- Network flow approximation\n",
    "- `IVRUPowerModel` (NLP) -- Current-Voltage rectangular coordinates\n",
    "\n",
    "**Note**: the optimality gap is normally used to compare NLPs with approximations/relaxations. NLPs vs NLPs should give a value close to 0.0%\n",
    "\n",
    "**Note 2**: Make sure that the solver terminates nominally!, i.e., `result[\"termination_status\"] == OPTIMAL` ||  `result[\"termination_status\"] == LOCALLY_SOLVED`\n",
    "\n",
    "I will provide an example below between **AC Polar Unbalaced vs. AC Rectangular Unbalanced**\n",
    "\n",
    "**Hint**: To compare against other formulations, change the type of formulation in the line `result_relax` to solve for the formulation you want to compare against the `ACPUPowerModel` formulation, and _re-run_ the notebook cells."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "1887c93d-e916-4d08-9092-d833d79e7355",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total number of variables............................:       63\n",
      "                     variables with only lower bounds:        9\n",
      "                variables with lower and upper bounds:        0\n",
      "                     variables with only upper bounds:        0\n",
      "Total number of equality constraints.................:       63\n",
      "Total number of inequality constraints...............:       30\n",
      "        inequality constraints with only lower bounds:        9\n",
      "   inequality constraints with lower and upper bounds:        0\n",
      "        inequality constraints with only upper bounds:       21\n",
      "\n",
      "\n",
      "Number of Iterations....: 3\n",
      "\n",
      "                                   (scaled)                 (unscaled)\n",
      "Objective...............:   1.8344980704278438e-02    1.8344980704278438e-02\n",
      "Dual infeasibility......:   1.4224867400962002e-08    1.4224867400962002e-08\n",
      "Constraint violation....:   1.5674554394954330e-14    5.0076175844848692e-13\n",
      "Variable bound violation:   0.0000000000000000e+00    0.0000000000000000e+00\n",
      "Complementarity.........:   0.0000000000000000e+00    0.0000000000000000e+00\n",
      "Overall NLP error.......:   8.6864427797129750e-09    1.4224867400962002e-08\n",
      "\n",
      "\n",
      "Number of objective function evaluations             = 4\n",
      "Number of objective gradient evaluations             = 4\n",
      "Number of equality constraint evaluations            = 4\n",
      "Number of inequality constraint evaluations          = 4\n",
      "Number of equality constraint Jacobian evaluations   = 4\n",
      "Number of inequality constraint Jacobian evaluations = 4\n",
      "Number of Lagrangian Hessian evaluations             = 3\n",
      "Total seconds in IPOPT                               = 0.002\n",
      "\n",
      "EXIT: Optimal Solution Found.\n",
      "----------------------------------------------------------------------------------------\n",
      "----------------------------------------------------------------------------------------\n",
      "Total number of variables............................:       66\n",
      "                     variables with only lower bounds:        0\n",
      "                variables with lower and upper bounds:        6\n",
      "                     variables with only upper bounds:        0\n",
      "Total number of equality constraints.................:       63\n",
      "Total number of inequality constraints...............:       48\n",
      "        inequality constraints with only lower bounds:       14\n",
      "   inequality constraints with lower and upper bounds:        0\n",
      "        inequality constraints with only upper bounds:       34\n",
      "\n",
      "\n",
      "Number of Iterations....: 21\n",
      "\n",
      "                                   (scaled)                 (unscaled)\n",
      "Objective...............:   1.8344980762983396e-02    1.8344980762983396e-02\n",
      "Dual infeasibility......:   8.9164267302673548e-11    8.9164267302673548e-11\n",
      "Constraint violation....:   1.6113129645571360e-14    4.5474735088646412e-13\n",
      "Variable bound violation:   0.0000000000000000e+00    0.0000000000000000e+00\n",
      "Complementarity.........:   3.9772815329178049e-09    3.9772815329178049e-09\n",
      "Overall NLP error.......:   3.9772815329178049e-09    3.9772815329178049e-09\n",
      "\n",
      "\n",
      "Number of objective function evaluations             = 25\n",
      "Number of objective gradient evaluations             = 22\n",
      "Number of equality constraint evaluations            = 25\n",
      "Number of inequality constraint evaluations          = 25\n",
      "Number of equality constraint Jacobian evaluations   = 22\n",
      "Number of inequality constraint Jacobian evaluations = 22\n",
      "Number of Lagrangian Hessian evaluations             = 21\n",
      "Total seconds in IPOPT                               = 0.006\n",
      "\n",
      "EXIT: Optimal Solution Found.\n",
      "----------------------------------------------------------------------------------------\n",
      "----------------------------------------------------------------------------------------\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "-3.2000555633226365e-7"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Base nlp to compare\n",
    "result_nlp = solve_mc_opf(data_pmd, ACPUPowerModel, ipopt_solver)\n",
    "\n",
    "println(\"----------------------------------------------------------------------------------------\")\n",
    "println(\"----------------------------------------------------------------------------------------\")\n",
    "\n",
    "# Approx. or Relax. to compare (in this case is another nlp formulation)\n",
    "result_relax = solve_mc_opf(data_pmd, ACRUPowerModel, ipopt_solver)\n",
    "\n",
    "println(\"----------------------------------------------------------------------------------------\")\n",
    "println(\"----------------------------------------------------------------------------------------\")\n",
    "\n",
    "# Optimality Gap\n",
    "op_gap = optimality_gap(result_nlp, result_relax) # result is in %\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "7146b032-b6c3-4abb-a2e8-f577c967ccaf",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mMain | Info ] : \u001b[22m\u001b[39mNLP: 0.018344980704278438\n",
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mMain | Info ] : \u001b[22m\u001b[39mNLP: LOCALLY_SOLVED\n",
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mMain | Info ] : \u001b[22m\u001b[39mRELAX: 0.018344980762983396\n",
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mMain | Info ] : \u001b[22m\u001b[39mRELAX: LOCALLY_SOLVED\n",
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mMain | Info ] : \u001b[22m\u001b[39mOptimality Gap: -3.2000555633226365e-7 %\n"
     ]
    }
   ],
   "source": [
    "@info \"NLP: $(result_nlp[\"objective\"])\"\n",
    "@info \"NLP: $(result_nlp[\"termination_status\"])\"\n",
    "\n",
    "@info \"RELAX: $(result_relax[\"objective\"])\"\n",
    "@info \"RELAX: $(result_relax[\"termination_status\"])\"\n",
    "\n",
    "@info \"Optimality Gap: $(op_gap) %\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "49730098-55cb-4479-8053-283488a55441",
   "metadata": {},
   "source": [
    "# Tutorial **PowerModelsITD** (PowerModels Integrated Transmission-Distribution)\n",
    "\n",
    "Now, let's take a look at [`PowerModelsITD`](https://github.com/lanl-ansi/PowerModelsITD.jl). [PMITD](https://github.com/lanl-ansi/PowerModelsITD.jl) is a Julia package that was designed to combine the capabilities of `PowerModels` and `PowerModelsDistribution` to enable modeling and solving large-scale Transmission and Distribution PF and OPF problems.\n",
    "\n",
    "Similar to `PM` and `PMD`, `PMITD` works under the same principles and data formats.\n",
    "\n",
    "- **Transmission**: Matpower \".m\" and PTI \".raw\" files (PSS(R)E v33 specification)\n",
    "- **Distribution**: OpenDSS \".dss\" files\n",
    "- **Boundary**: JSON \".json\" files\n",
    "\n",
    "The transmission system information is passed in using the same data formats used by PowerModels while the distribution system(s) are passed using the same data format PowerModelsDistribution supports, i.e., OpenDSS.\n",
    "\n",
    "The only additional information that is required is the **boundary** information, which is given using a `.json` file, that indicates which transmission bus connects to which distribution system (bus). \n",
    "\n",
    "For example, in the file shown below, the `voltage_source.source` bus of the `3bus_unbal` distribution system is connected to `bus #5` of the transmission system.\n",
    "\n",
    "```\n",
    "[\n",
    "\t{\n",
    "        \"transmission_boundary\": \"5\",\n",
    "        \"distribution_boundary\": \"3bus_unbal.voltage_source.source\"\n",
    "\t},\n",
    "\n",
    "\t{\n",
    "        \"transmission_boundary\": \"6\",\n",
    "        \"distribution_boundary\": \"3bus_bal.voltage_source.source\"\n",
    "        }\n",
    "]\n",
    "\n",
    "```\n",
    "\n",
    "Similarly, the `voltage_source.source` bus of the `3bus_bal` distribution system is connected to the `bus #6` of the transmission system.\n",
    "\n",
    "**Note**: A detailed description of the File Formats supported in `PowerModelsITD` can be found [here](https://lanl-ansi.github.io/PowerModelsITD.jl/stable/manual/fileformat.html).\n",
    "\n"
   ]
  },
  {
   "attachments": {
    "d0777324-6ef4-46cc-a383-1d6d46b05dff.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "id": "94d106c7-6059-48f9-bf5c-53edd3345de2",
   "metadata": {},
   "source": [
    "### Mathematics at the T&D Boundaries\n",
    "\n",
    "**PowerModelsITD** uses **PowerModels** as the _backend_ that generates the mathematical problem for the transmission system, and it uses **PowerModelsDistribution** as the _backend_ that generates the mathematical problems for the distribution systems. \n",
    "\n",
    "#### But, How **PowerModelsITD** `\"connects\"` or `\"models\"` the T&D boundary mathematically?\n",
    "\n",
    "Each formulation, e.g., ACP-ACPU, ACR-ACRU, has its own set of mathematical constraints defined at the T&D boundary(ies) that allow the mathematical interconnection between the transmission and distribution systems. You can find the mathematical constraints for a list of ITD supported formulations [here](https://lanl-ansi.github.io/PowerModelsITD.jl/stable/manual/formulations.html). \n",
    "\n",
    "Let's us check out the mathematical constraints for the `ACP-ACPU` (AC polar) formulation.\n",
    "\n",
    "##### Sets, Parameters, and (General) Variables\n",
    "\n",
    "\\begin{align}\n",
    "%\n",
    "\\mbox{sets:} & \\nonumber \\\\\n",
    "& N \\mbox{ - Set of buses}\\nonumber \\\\\n",
    "& \\mathcal{T} \\mbox{ - Belongs to transmission network}\\nonumber \\\\\n",
    "& \\mathcal{D} \\mbox{ - Belongs to distribution network}\\nonumber \\\\\n",
    "& \\mathcal{B} \\mbox{ - Set of boundary links}\\nonumber \\\\\n",
    "%\n",
    "\\mbox{parameters:} & \\nonumber \\\\\n",
    "& \\Re \\mbox{ - Real part}\\nonumber \\\\\n",
    "& \\Im \\mbox{ - Imaginary part}\\nonumber \\\\\n",
    "& \\Phi = a, b, c \\mbox{ - Multi-conductor phases}\\nonumber \\\\\n",
    "& \\chi \\rightarrow{\\mathcal{T}},{\\mathcal{D}} \\mbox{ - Belongs to Transmission or Distribution}\\nonumber \\\\\n",
    "& \\beta^{^{\\chi}} \\mbox{ - Boundary bus}\\nonumber \\\\\n",
    "%\n",
    "\\mbox{variables:} & \\nonumber \\\\\n",
    "& P_{\\beta^{^\\mathcal{T}}\\beta^{^\\mathcal{D}}}^{^\\mathcal{T}} \\mbox{ - Active power flow from Transmisison boundary bus to Distribution boundary bus}\\nonumber \\\\\n",
    "& Q_{\\beta^{^\\mathcal{T}}\\beta^{^\\mathcal{D}}}^{^\\mathcal{T}} \\mbox{ - Reactive power flow from Transmisison boundary bus to Distribution boundary bus}\\nonumber \\\\\n",
    "& P_{\\beta^{^\\mathcal{D}}\\beta^{^\\mathcal{T}}}^{{\\mathcal{D},\\varphi}} \\mbox{ - Active power flow from Distribution boundary bus phase $\\varphi$ to Transmission boundary bus}\\nonumber \\\\\n",
    "& Q_{\\beta^{^\\mathcal{D}}\\beta^{^\\mathcal{T}}}^{{\\mathcal{D},\\varphi}} \\mbox{ - Reactive power flow from Distribution boundary bus phase $\\varphi$ to Transmission boundary bus}\\nonumber \\\\\n",
    "& V_i^{^\\mathcal{T}} \\mbox{ - Voltage magnitude at bus $i$}\\nonumber \\\\\n",
    "& \\theta_i^{^\\mathcal{T}} \\mbox{ - Voltage angle at bus $i$}\\nonumber \\\\\n",
    "& v_i^{\\mathcal{D}, \\varphi} \\mbox{ - Voltage magnitude at bus $i$ phase $\\varphi$}\\nonumber \\\\\n",
    "& \\theta_i^{\\mathcal{D}, \\varphi} \\mbox{ - Voltage angle at bus $i$ phase $\\varphi$}\\nonumber \\\\\n",
    "%\n",
    "\\end{align}\n",
    "\n",
    "\n",
    "##### ACP-ACPU Formulation - Mathematical Constraints at the T&D Boundary\n",
    "\n",
    "ACP to ACPU (AC polar to AC polar unbalanced)\n",
    "\n",
    "- **Coordinates**: Polar\n",
    "- **Variables**: Power-Voltage\n",
    "- **Model(s)**: NLP-NLP\n",
    "- **ITD Boundary Math. Formulation**:\n",
    "\n",
    "\\begin{align}\n",
    "%\n",
    "\\mbox{ITD boundaries: } & \\nonumber \\\\\n",
    "& \\sum_{\\varphi \\in \\Phi} P_{\\beta^{^\\mathcal{D}}\\beta^{^\\mathcal{T}}}^{\\mathcal{D},\\varphi} +  P_{\\beta^{^\\mathcal{T}}\\beta^{^\\mathcal{D}}}^{^\\mathcal{T}} = 0, \\ \\forall (\\beta^{^\\mathcal{T}},\\beta^{^\\mathcal{D}}) \\in {\\Lambda} \\mbox{ - Active power flow at boundary} \\\\\n",
    "& \\sum_{\\varphi \\in \\Phi} Q_{\\beta^{^\\mathcal{D}}\\beta^{^\\mathcal{T}}}^{\\mathcal{D},\\varphi} +  Q_{\\beta^{^\\mathcal{T}}\\beta^{^\\mathcal{D}}}^{^\\mathcal{T}} = 0, \\ \\forall (\\beta^{^\\mathcal{T}},\\beta^{^\\mathcal{D}}) \\in {\\Lambda} \\mbox{ - Reactive power flow at boundary} \\\\\n",
    "& V_{\\beta^{^\\mathcal{T}}} = v_{\\beta^{^\\mathcal{D}}}^{^{a}}, \\ \\forall (\\beta^{^\\mathcal{T}},\\beta^{^\\mathcal{D}}) \\in {\\Lambda} \\mbox{ - Voltage mag. equality - phase a} \\\\\n",
    "& V_{\\beta^{^\\mathcal{T}}} = v_{\\beta^{^\\mathcal{D}}}^{^{b}}, \\ \\forall (\\beta^{^\\mathcal{T}},\\beta^{^\\mathcal{D}}) \\in {\\Lambda} \\mbox{ - Voltage mag. equality - phase b} \\\\\n",
    "& V_{\\beta^{^\\mathcal{T}}} = v_{\\beta^{^\\mathcal{D}}}^{^{c}}, \\ \\forall (\\beta^{^\\mathcal{T}},\\beta^{^\\mathcal{D}}) \\in {\\Lambda} \\mbox{ - Voltage mag. equality - phase c} \\\\\n",
    "& \\theta_{\\beta^{^\\mathcal{T}}} = \\theta_{\\beta^{^\\mathcal{D}}}^{^{a}}, \\ \\forall (\\beta^{^\\mathcal{T}},\\beta^{^\\mathcal{D}}) \\in {\\Lambda} \\mbox{ - Voltage ang. equality - phase a} \\\\\n",
    "& \\theta_{\\beta^{^\\mathcal{D}}}^{^{b}} = (\\theta_{\\beta^{^\\mathcal{D}}}^{^{a}} -120^{\\circ}),  \\ \\forall \\beta^{^\\mathcal{D}} \\in N^{^\\mathcal{B}} \\cap  N^{^\\mathcal{D}} \\mbox{ - Voltage ang. equality - phase b} \\\\\n",
    "& \\theta_{\\beta^{^\\mathcal{D}}}^{^{c}} = (\\theta_{\\beta^{^\\mathcal{D}}}^{^{a}} +120^{\\circ}), \\ \\forall \\beta^{^\\mathcal{D}} \\in N^{^\\mathcal{B}} \\cap  N^{^\\mathcal{D}} \\mbox{ - Voltage ang. equality - phase c} \\\\\n",
    "%\n",
    "\\end{align}\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "![image.png](attachment:d0777324-6ef4-46cc-a383-1d6d46b05dff.png)\n",
    "\n",
    "The figure above depicts the ACP-ACPU formulation mapping between the single-phase transmission system buses and the three-phase distribution system nodes at the T&D boundaries. As seen in the figure, the voltage constraints enforce the mathematical mapping of the boundary buses, and the power flow\n",
    "constraints ensure that the power coming from the transmission system boundary bus is equal to the power flowing towards the rest of the distribution system.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b8f3440a-da17-4e7f-a876-8fbe3c73fad1",
   "metadata": {},
   "source": [
    "### Let's import PowerModelsITD and parse the files we want to solve\n",
    "\n",
    "Here, you can see that it will be very similar to how `PowerModels` and `PowerModelsDistribution` work, with the addition of having to parse the **boundary** information file.\n",
    "\n",
    "In this example, we want to solve the integrated T&D optimal power flow (opfitd) problem for a case where we have the PJM 5-bus system (`case5_withload`) as the transmission system and the IEEE 4 Node Test Feeder (`case3_unbalanced.dss`) as the distribution system. The boundary information that specifies the respective boundary buses for both the transmission and distribution systems can be found in `case5_case3_unbal.json`.\n",
    "\n",
    "The formulation used to solve this optimization problem is the **ACP-ACPU** (i.e., AC polar for transmission and AC polar unbalanced for distribution systems) formulation.\n",
    "\n",
    "**Note**: All files used here can be found in the PowerModelsITD [test/data](https://github.com/lanl-ansi/PowerModelsITD.jl/tree/main/test/data) folder."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "2aa6b4b4-3ed2-46fe-ac5f-6e007a028850",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mPowerModelsDistribution | Info ] : \u001b[22m\u001b[39mbasemva=100 is the default value, you may want to adjust sbase_default for better convergence\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 3 entries:\n",
       "  \"multiinfrastructure\" => true\n",
       "  \"it\"                  => Dict{String, Any}(\"pmd\"=>Dict{String, Any}(\"conducto…\n",
       "  \"per_unit\"            => false"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "using PowerModelsITD\n",
    "\n",
    "# Define the path to be the PowerModelsITD path (optional)\n",
    "files_path = joinpath(dirname(pathof(PowerModelsITD)), \"..\")\n",
    "\n",
    "# Transmission system file\n",
    "pm_file = joinpath(files_path, \"test/data/transmission/case5_withload.m\")\n",
    "\n",
    "# Distribution system file\n",
    "pmd_file = joinpath(files_path, \"test/data/distribution/case3_unbalanced.dss\")\n",
    "\n",
    "# Boundary information file\n",
    "boundary_file = joinpath(files_path, \"test/data/json/case5_case3_unbal.json\")\n",
    "\n",
    "# parse the files into dictionary\n",
    "data = PowerModelsITD.parse_files(pm_file, pmd_file, boundary_file)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "60d60779-a58b-46af-87a5-cb7bfd8751ce",
   "metadata": {},
   "source": [
    "In this case, we need to access the information of the system(s) through the `it` dictionary key, for example:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "8996d248-114e-453f-8f4f-bd987b0ad1e6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 3 entries:\n",
       "  \"pmd\"   => Dict{String, Any}(\"conductor_ids\"=>[1, 2, 3, 4], \"bus\"=>Dict{Any, …\n",
       "  \"pm\"    => Dict{String, Any}(\"bus\"=>Dict{String, Any}(\"4\"=>Dict{String, Any}(…\n",
       "  \"pmitd\" => Dict{String, Any}(\"100001\"=>Dict{String, Any}(\"distribution_bounda…"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[\"it\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "09ef3254-4601-4baf-a91d-8e2460e063f8",
   "metadata": {},
   "source": [
    "Inside the `\"pm\"` key, we find the information related to the **transmission system**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "495dfa11-76fd-4e68-80e9-253f85f842f1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 13 entries:\n",
       "  \"bus\"            => Dict{String, Any}(\"4\"=>Dict{String, Any}(\"zone\"=>1, \"bus_…\n",
       "  \"source_type\"    => \"matpower\"\n",
       "  \"name\"           => \"case5\"\n",
       "  \"dcline\"         => Dict{String, Any}()\n",
       "  \"source_version\" => \"2\"\n",
       "  \"gen\"            => Dict{String, Any}(\"4\"=>Dict{String, Any}(\"pg\"=>0.0, \"mode…\n",
       "  \"branch\"         => Dict{String, Any}(\"4\"=>Dict{String, Any}(\"br_r\"=>0.00108,…\n",
       "  \"storage\"        => Dict{String, Any}()\n",
       "  \"switch\"         => Dict{String, Any}()\n",
       "  \"baseMVA\"        => 100.0\n",
       "  \"per_unit\"       => true\n",
       "  \"shunt\"          => Dict{String, Any}()\n",
       "  \"load\"           => Dict{String, Any}(\"4\"=>Dict{String, Any}(\"source_id\"=>Any…"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[\"it\"][\"pm\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "d19e38b2-ee98-4ba1-b078-1c0c1fb368a4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 6 entries:\n",
       "  \"4\"  => Dict{String, Any}(\"zone\"=>1, \"bus_i\"=>4, \"bus_type\"=>3, \"vmax\"=>1.1, …\n",
       "  \"1\"  => Dict{String, Any}(\"zone\"=>1, \"bus_i\"=>1, \"bus_type\"=>2, \"vmax\"=>1.1, …\n",
       "  \"5\"  => Dict{String, Any}(\"zone\"=>1, \"bus_i\"=>5, \"bus_type\"=>1, \"vmax\"=>1.1, …\n",
       "  \"2\"  => Dict{String, Any}(\"zone\"=>1, \"bus_i\"=>2, \"bus_type\"=>1, \"vmax\"=>1.1, …\n",
       "  \"10\" => Dict{String, Any}(\"zone\"=>1, \"bus_i\"=>10, \"bus_type\"=>2, \"vmax\"=>1.1,…\n",
       "  \"3\"  => Dict{String, Any}(\"zone\"=>1, \"bus_i\"=>3, \"bus_type\"=>2, \"vmax\"=>1.1, …"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[\"it\"][\"pm\"][\"bus\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "31ad3655-d7d1-43d0-9fba-40c32c9dc165",
   "metadata": {},
   "source": [
    "Inside the `\"pmd\"` key, we find the  information related to the **distribution systems**. All of the distribution systems. \n",
    "- _Note_: Each element name has the format `name_of_distro_system.element_name`.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "0698da64-3fee-42b7-b949-2fe2bd92ea35",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 13 entries:\n",
       "  \"conductor_ids\"  => [1, 2, 3, 4]\n",
       "  \"bus\"            => Dict{Any, Any}(\"3bus_unbal.loadbus\"=>Dict{String, Any}(\"r…\n",
       "  \"name\"           => \"3bus_unbal\"\n",
       "  \"ckt_names\"      => [\"3bus_unbal\"]\n",
       "  \"settings\"       => Dict{String, Any}(\"sbase_default\"=>100000.0, \"vbases_defa…\n",
       "  \"files\"          => [\"/Users/dfobes/.julia/packages/PowerModelsITD/09MN1/src/…\n",
       "  \"generator\"      => Dict{Any, Any}(\"3bus_unbal.gen1\"=>Dict{String, Any}(\"vg\"=…\n",
       "  \"voltage_source\" => Dict{Any, Any}(\"3bus_unbal.source\"=>Dict{String, Any}(\"so…\n",
       "  \"line\"           => Dict{Any, Any}(\"3bus_unbal.quad\"=>Dict{String, Any}(\"leng…\n",
       "  \"data_model\"     => ENGINEERING\n",
       "  \"transformer\"    => Dict{Any, Any}(\"3bus_unbal.subxf\"=>Dict{String, Any}(\"pol…\n",
       "  \"load\"           => Dict{Any, Any}(\"3bus_unbal.l2\"=>Dict{String, Any}(\"model\"…\n",
       "  \"linecode\"       => Dict{Any, Any}(\"3bus_unbal.4/0quad\"=>Dict{String, Any}(\"b…"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[\"it\"][\"pmd\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "5c914d3f-a6a4-40a5-a441-aca40df0dc47",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{Any, Any} with 4 entries:\n",
       "  \"3bus_unbal.loadbus\"    => Dict{String, Any}(\"rg\"=>[0.0], \"grounded\"=>[4], \"s…\n",
       "  \"3bus_unbal.substation\" => Dict{String, Any}(\"rg\"=>[0.0], \"grounded\"=>[4], \"s…\n",
       "  \"3bus_unbal.primary\"    => Dict{String, Any}(\"rg\"=>Float64[], \"grounded\"=>Int…\n",
       "  \"3bus_unbal.sourcebus\"  => Dict{String, Any}(\"rg\"=>[0.0], \"grounded\"=>[4], \"s…"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[\"it\"][\"pmd\"][\"bus\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7085c597-bcbc-4351-94c6-454b499c6eb7",
   "metadata": {},
   "source": [
    "Inside the `\"pmitd\"` key, we find the information related to the **boundary(ies)**\n",
    "- _Note_: Each **boundary** is numbered starting from \"100001\" to ....."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "65b5316e-533b-4cfb-9768-96185a55c1de",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 1 entry:\n",
       "  \"100001\" => Dict{String, Any}(\"distribution_boundary\"=>\"3bus_unbal.voltage_so…"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[\"it\"][\"pmitd\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "dda761fb-ee02-419c-8ef3-704a581c3669",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 2 entries:\n",
       "  \"distribution_boundary\" => \"3bus_unbal.voltage_source.source\"\n",
       "  \"transmission_boundary\" => \"5\""
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[\"it\"][\"pmitd\"][\"100001\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3a542ce2-87da-464c-a27c-c978418781cf",
   "metadata": {},
   "source": [
    "### Now, let's solve the T&D Optimal Power Flow (OPFITD) for this problem \n",
    "\n",
    "This will solve the OPFITD, optimal power flow for integrated transmission-distribution system\n",
    "\n",
    "<span style=\"color:red\">**NOTE**: _If you get an error, try reparsing the data. This issue happens because after the problem is solved, the `data` structure is modified and affects re-running the problem again. That is why you need to re-parse the file_.</span>."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "b305e979-377c-4602-9304-7b2c25f4488b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total number of variables............................:      211\n",
      "                     variables with only lower bounds:       27\n",
      "                variables with lower and upper bounds:       71\n",
      "                     variables with only upper bounds:        0\n",
      "Total number of equality constraints.................:      199\n",
      "Total number of inequality constraints...............:       69\n",
      "        inequality constraints with only lower bounds:       18\n",
      "   inequality constraints with lower and upper bounds:        7\n",
      "        inequality constraints with only upper bounds:       44\n",
      "\n",
      "\n",
      "Number of Iterations....: 24\n",
      "\n",
      "                                   (scaled)                 (unscaled)\n",
      "Objective...............:   4.4884366280124215e+02    1.7953746512049685e+04\n",
      "Dual infeasibility......:   7.9856407667097663e-10    3.1942563066839065e-08\n",
      "Constraint violation....:   3.9849375691553443e-14    5.8670117375481112e-12\n",
      "Variable bound violation:   3.0132552453210337e-08    3.0132552453210337e-08\n",
      "Complementarity.........:   2.5059052362800323e-09    1.0023620945120128e-07\n",
      "Overall NLP error.......:   2.5059052362800323e-09    1.0023620945120128e-07\n",
      "\n",
      "\n",
      "Number of objective function evaluations             = 34\n",
      "Number of objective gradient evaluations             = 25\n",
      "Number of equality constraint evaluations            = 34\n",
      "Number of inequality constraint evaluations          = 34\n",
      "Number of equality constraint Jacobian evaluations   = 25\n",
      "Number of inequality constraint Jacobian evaluations = 25\n",
      "Number of Lagrangian Hessian evaluations             = 24\n",
      "Total seconds in IPOPT                               = 0.022\n",
      "\n",
      "EXIT: Optimal Solution Found.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mPowerModelsDistribution | Info ] : \u001b[22m\u001b[39mbasemva=100 is the default value, you may want to adjust sbase_default for better convergence\n",
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mPowerModelsITD | Info ] : \u001b[22m\u001b[39mpmitd model solution time (instantiate + optimization): 5.668787002563477\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 8 entries:\n",
       "  \"solve_time\"         => 0.0237119\n",
       "  \"optimizer\"          => \"Ipopt\"\n",
       "  \"termination_status\" => LOCALLY_SOLVED\n",
       "  \"dual_status\"        => FEASIBLE_POINT\n",
       "  \"primal_status\"      => FEASIBLE_POINT\n",
       "  \"objective\"          => 17953.7\n",
       "  \"solution\"           => Dict{String, Any}(\"multiinfrastructure\"=>true, \"it\"=>…\n",
       "  \"objective_lb\"       => -Inf"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "using Ipopt\n",
    "\n",
    "# Re-parse data to avoid errors\n",
    "data = PowerModelsITD.parse_files(pm_file, pmd_file, boundary_file)\n",
    "\n",
    "# Define T&D formulation to use\n",
    "formulation = PowerModelsITD.NLPowerModelITD{ACPPowerModel, ACPUPowerModel}\n",
    "\n",
    "# solver\n",
    "ipopt_solver = optimizer_with_attributes(Ipopt.Optimizer, \"print_level\" => 3)\n",
    "\n",
    "# solve the opf itd optimization problem\n",
    "result = solve_opfitd(data, formulation, ipopt_solver)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b8faae6f-9dac-4d95-8257-155dc8ae79d4",
   "metadata": {},
   "source": [
    "#### Let's check the results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "b38e3a62-fd48-4b31-9bca-2e2316d284ae",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mMain | Info ] : \u001b[22m\u001b[39mObjective: 17953.746512049685 [$/hr]\n"
     ]
    }
   ],
   "source": [
    "@info \"Objective: $(result[\"objective\"]) [\\$/hr]\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "5946ef90-28db-4c6d-a422-ecff78761162",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mMain | Info ] : \u001b[22m\u001b[39mTermination Status of Solver: LOCALLY_SOLVED\n"
     ]
    }
   ],
   "source": [
    "@info \"Termination Status of Solver: $(result[\"termination_status\"])\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "ababeed2-9ab7-4849-b9f6-1910ea0a8220",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mMain | Info ] : \u001b[22m\u001b[39mSolve Time: 0.0237119197845459 [seconds]\n"
     ]
    }
   ],
   "source": [
    "@info \"Solve Time: $(result[\"solve_time\"]) [seconds]\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b9c07ce4-7929-4476-8139-33e2676bcff9",
   "metadata": {},
   "source": [
    "##### Solution Values for Transmission system"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "92ff01a0-345e-4b3c-8bde-a2a35831989d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 6 entries:\n",
       "  \"4\"  => Dict{String, Any}(\"va\"=>-9.66631e-28, \"vm\"=>0.9)\n",
       "  \"1\"  => Dict{String, Any}(\"va\"=>3.95571, \"vm\"=>0.917815)\n",
       "  \"5\"  => Dict{String, Any}(\"va\"=>-0.803612, \"vm\"=>0.940088)\n",
       "  \"2\"  => Dict{String, Any}(\"va\"=>-0.656332, \"vm\"=>0.941632)\n",
       "  \"10\" => Dict{String, Any}(\"va\"=>5.01851, \"vm\"=>0.904864)\n",
       "  \"3\"  => Dict{String, Any}(\"va\"=>-0.312228, \"vm\"=>0.967094)"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Buses\n",
    "result[\"solution\"][\"it\"][\"pm\"][\"bus\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "28e9d289-430f-47ba-96f2-04b4c4bacd3a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 7 entries:\n",
       "  \"4\" => Dict{String, Any}(\"qf\"=>-215.457, \"qt\"=>220.015, \"pt\"=>72.7277, \"pf\"=>…\n",
       "  \"1\" => Dict{String, Any}(\"qf\"=>-91.9458, \"qt\"=>113.049, \"pt\"=>-235.968, \"pf\"=…\n",
       "  \"5\" => Dict{String, Any}(\"qf\"=>71.3753, \"qt\"=>-69.0087, \"pt\"=>57.0431, \"pf\"=>…\n",
       "  \"2\" => Dict{String, Any}(\"qf\"=>40.8275, \"qt\"=>-27.519, \"pt\"=>-190.173, \"pf\"=>…\n",
       "  \"6\" => Dict{String, Any}(\"qf\"=>19.2895, \"qt\"=>1.0554, \"pt\"=>239.998, \"pf\"=>-2…\n",
       "  \"7\" => Dict{String, Any}(\"qf\"=>3.79799, \"qt\"=>-4.36718, \"pt\"=>-8.06845, \"pf\"=…\n",
       "  \"3\" => Dict{String, Any}(\"qf\"=>208.618, \"qt\"=>-204.199, \"pt\"=>220.4, \"pf\"=>-2…"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Branches\n",
    "result[\"solution\"][\"it\"][\"pm\"][\"branch\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "2396f427-a23f-4d94-9609-e44c1cf81a72",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 5 entries:\n",
       "  \"4\" => Dict{String, Any}(\"qg\"=>54.2318, \"pg\"=>18.9614)\n",
       "  \"1\" => Dict{String, Any}(\"qg\"=>30.0, \"pg\"=>40.0)\n",
       "  \"5\" => Dict{String, Any}(\"qg\"=>-203.144, \"pg\"=>460.398)\n",
       "  \"2\" => Dict{String, Any}(\"qg\"=>127.5, \"pg\"=>170.0)\n",
       "  \"3\" => Dict{String, Any}(\"qg\"=>390.0, \"pg\"=>315.977)"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Generators\n",
    "result[\"solution\"][\"it\"][\"pm\"][\"gen\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4c227a6b-c8b5-4719-959c-743c755327e6",
   "metadata": {},
   "source": [
    "##### Solution Values for Distribution system(s)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "de04ffac-4e67-4a54-b6ae-6ac5c2078b0d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 4 entries:\n",
       "  \"3bus_unbal.loadbus\"    => Dict{String, Any}(\"va\"=>[-0.954285, -120.825, 119.…\n",
       "  \"3bus_unbal.substation\" => Dict{String, Any}(\"va\"=>[-0.808934, -120.807, 119.…\n",
       "  \"3bus_unbal.primary\"    => Dict{String, Any}(\"va\"=>[-0.877214, -120.816, 119.…\n",
       "  \"3bus_unbal.sourcebus\"  => Dict{String, Any}(\"va\"=>[-0.80642, -120.806, 119.1…"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Buses/Nodes\n",
    "result[\"solution\"][\"it\"][\"pmd\"][\"bus\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "1eb452a6-e368-4077-b057-823603d930cc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 2 entries:\n",
       "  \"3bus_unbal.quad\"   => Dict{String, Any}(\"qf\"=>[1344.78, 1503.94, 1502.44], \"…\n",
       "  \"3bus_unbal.ohline\" => Dict{String, Any}(\"qf\"=>[1354.62, 1507.23, 1504.47], \"…"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Lines (i.e., branches)\n",
    "result[\"solution\"][\"it\"][\"pmd\"][\"line\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "14b77a34-2aa0-4c76-8cee-4bf3828dbc7d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 4 entries:\n",
       "  \"qg_bus\" => [-0.0, -0.0, -0.0]\n",
       "  \"qg\"     => [-0.0, -0.0, -0.0]\n",
       "  \"pg\"     => [666.668, 666.668, 666.668]\n",
       "  \"pg_bus\" => [666.668, 666.668, 666.668]"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Generators (DG1)\n",
    "result[\"solution\"][\"it\"][\"pmd\"][\"generator\"][\"3bus_unbal.gen1\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "06eb18ff-fef9-40cc-9a51-bc37d4e4acec",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 3 entries:\n",
       "  \"3bus_unbal.l2\" => Dict{String, Any}(\"qd_bus\"=>[1500.0], \"pd_bus\"=>[3000.0], …\n",
       "  \"3bus_unbal.l3\" => Dict{String, Any}(\"qd_bus\"=>[1500.0], \"pd_bus\"=>[3000.0], …\n",
       "  \"3bus_unbal.l1\" => Dict{String, Any}(\"qd_bus\"=>[1333.33], \"pd_bus\"=>[4000.0],…"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Loads\n",
    "result[\"solution\"][\"it\"][\"pmd\"][\"load\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "09547e6c-3842-4233-8880-7e9e0b3448cc",
   "metadata": {},
   "source": [
    "##### Solution Values for Boundary(ies) (T&D interconnection)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "c9f27e9f-a18c-408d-b788-9c54db812bfb",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 2 entries:\n",
       "  \"pbound_fr\" => [8068.45]\n",
       "  \"qbound_fr\" => [4367.18]"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Boundary - Transmission (Bus #5) to Distribution (3bus_unbal)\n",
    "result[\"solution\"][\"it\"][\"pmitd\"][\"boundary\"][\"(100001, 5, voltage_source.3bus_unbal.source)\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "beb979ca-fe53-4f3d-a5c7-ed723142d535",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 2 entries:\n",
       "  \"pbound_to\" => [-3367.19, -2346.4, -2354.86]\n",
       "  \"qbound_to\" => [-1355.01, -1507.47, -1504.7]"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Boundary - Distribution (3bus_unbal) to Transmission (Bus #5)\n",
    "result[\"solution\"][\"it\"][\"pmitd\"][\"boundary\"][\"(100001, voltage_source.3bus_unbal.source, 5)\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6e932d9d-7147-4687-8f4f-2210dd87cbb1",
   "metadata": {},
   "source": [
    "# --------------------------------\n",
    "<span style=\"color:red\">**IMPORTANT!**</span>\n",
    "\n",
    "The values with a key name ending in `..._fr` represent the powers flowing from the Transmission to the Distribution. The values ending in `...to` represent the powers flowing from the Distribution to the Transmission (that's why they are negative). For the Distribution values, they represent `[phase A, phase B, phase C]`. The summation of all elements in the `to` vectors must be equal to the value in `fr`. The summation must be multiplied by $\\times$ -1 to make it positive.\n",
    "\n",
    "# --------------------------------"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4ce8b917-2179-4bb8-bcd4-7c481732b7bd",
   "metadata": {},
   "source": [
    "### Let's solve a Multinetwork (Time-series) Multisystem (Multiple Distribution Systems) OPFITD\n",
    "\n",
    "In this example case, we want to solve the integrated T&D optimal power flow (opfitd) problem for a Multinetwork case where we have the PJM 5-bus system (`case5_with2loads.m`) as the transmission system and two copies of the IEEE 4 Node Test Feeder (`case3_unbalanced_withoutgen_mn.dss`) as the distribution systems, where each distribution system has load profiles for their respective loads for **four timesteps**. The boundary information that specifies the respective boundary buses for both the transmission and distribution systems can be found in `case5_case3x2.json`.\n",
    "\n",
    "**Note**: The key `nw` means `network`. The term `multinetwork` is a generalization of `timeseries`, since each timestep in the timeseries can be considered a _different_ network.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "4544aea4-368c-46ef-8337-f24b384f9fba",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[33m\u001b[1m┌ \u001b[22m\u001b[39m\u001b[33m\u001b[1mPowerModelsITD | Warning ] : \u001b[22m\u001b[39mSuppressing information and warning messages output by PowerModels and PowerModelsDistribution for the rest of this session.\n",
      "\u001b[33m\u001b[1m└ \u001b[22m\u001b[39m\u001b[90m@ PowerModelsITD ~/.julia/packages/PowerModelsITD/09MN1/src/PowerModelsITD.jl:18\u001b[39m\n",
      "\u001b[33m\u001b[1m┌ \u001b[22m\u001b[39m\u001b[33m\u001b[1mPowerModelsITD | Warning ] : \u001b[22m\u001b[39mauto_rename option is true, so boundary names in 'pmitd=>' will be overwritten sequentially and may not represent the actual wanted boundary connections.\n",
      "\u001b[33m\u001b[1m└ \u001b[22m\u001b[39m\u001b[90m@ PowerModelsITD ~/.julia/packages/PowerModelsITD/09MN1/src/core/helpers.jl:107\u001b[39m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total number of variables............................:     1488\n",
      "                     variables with only lower bounds:      216\n",
      "                variables with lower and upper bounds:      388\n",
      "                     variables with only upper bounds:        0\n",
      "Total number of equality constraints.................:     1452\n",
      "Total number of inequality constraints...............:      480\n",
      "        inequality constraints with only lower bounds:      144\n",
      "   inequality constraints with lower and upper bounds:       32\n",
      "        inequality constraints with only upper bounds:      304\n",
      "\n",
      "\n",
      "Number of Iterations....: 24\n",
      "\n",
      "                                   (scaled)                 (unscaled)\n",
      "Objective...............:   1.7971067692666493e+03    7.1884270770665971e+04\n",
      "Dual infeasibility......:   8.1648649495225114e-10    3.2659459798090046e-08\n",
      "Constraint violation....:   5.0333644883944543e-14    6.0395854983852360e-12\n",
      "Variable bound violation:   3.0073922019369093e-08    3.0073922019369093e-08\n",
      "Complementarity.........:   2.5059168630961834e-09    1.0023667452384734e-07\n",
      "Overall NLP error.......:   2.5059168630961834e-09    1.0023667452384734e-07\n",
      "\n",
      "\n",
      "Number of objective function evaluations             = 25\n",
      "Number of objective gradient evaluations             = 25\n",
      "Number of equality constraint evaluations            = 25\n",
      "Number of inequality constraint evaluations          = 25\n",
      "Number of equality constraint Jacobian evaluations   = 25\n",
      "Number of inequality constraint Jacobian evaluations = 25\n",
      "Number of Lagrangian Hessian evaluations             = 24\n",
      "Total seconds in IPOPT                               = 0.138\n",
      "\n",
      "EXIT: Optimal Solution Found.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mPowerModelsITD | Info ] : \u001b[22m\u001b[39mpmitd model solution time (instantiate + optimization): 0.6272449493408203\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 8 entries:\n",
       "  \"solve_time\"         => 0.148122\n",
       "  \"optimizer\"          => \"Ipopt\"\n",
       "  \"termination_status\" => LOCALLY_SOLVED\n",
       "  \"dual_status\"        => FEASIBLE_POINT\n",
       "  \"primal_status\"      => FEASIBLE_POINT\n",
       "  \"objective\"          => 71884.3\n",
       "  \"solution\"           => Dict{String, Any}(\"multiinfrastructure\"=>true, \"it\"=>…\n",
       "  \"objective_lb\"       => -Inf"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "using PowerModelsITD\n",
    "using Ipopt\n",
    "\n",
    "# solver\n",
    "ipopt_solver = optimizer_with_attributes(Ipopt.Optimizer, \"print_level\" => 3)\n",
    "\n",
    "PowerModelsITD.silence!()  # Silences the console, so warning messages or information are not displayed\n",
    "\n",
    "files_path = joinpath(dirname(pathof(PowerModelsITD)), \"..\")                                    # file path for test cases files\n",
    "pm_file = joinpath(files_path, \"test/data/transmission/case5_with2loads.m\")                     # transmission system file\n",
    "pmd_file1 = joinpath(files_path, \"test/data/distribution/case3_unbalanced_withoutgen_mn.dss\")    # distribution system file 1\n",
    "pmd_file2 = joinpath(files_path, \"test/data/distribution/case3_unbalanced_withoutgen_mn_diff.dss\")    # distribution system file 2\n",
    "boundary_file = joinpath(files_path, \"test/data/json/case5_case3x2.json\")                       # boundary file with info. about boundary buses\n",
    "pmd_files = [pmd_file1, pmd_file2]                                                                # vector of distrib. systems (in this case a copy of the same)\n",
    "\n",
    "# type of formulation to be used\n",
    "formulation = PowerModelsITD.NLPowerModelITD{ACPPowerModel, ACPUPowerModel}\n",
    "\n",
    "# solve the multinetwork opfitd\n",
    "result = PowerModelsITD.solve_mn_opfitd(pm_file, pmd_files, boundary_file, formulation, ipopt_solver; auto_rename=true)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "582cba5d-b420-4e65-8950-c0d5b4f3271a",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mMain | Info ] : \u001b[22m\u001b[39mObjective: 71884.27077066597 [$/hr]\n"
     ]
    }
   ],
   "source": [
    "@info \"Objective: $(result[\"objective\"]) [\\$/hr]\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "df496be5-c9c3-44b8-8fb8-ead4a0fe1524",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mMain | Info ] : \u001b[22m\u001b[39mTermination Status of Solver: LOCALLY_SOLVED\n"
     ]
    }
   ],
   "source": [
    "@info \"Termination Status of Solver: $(result[\"termination_status\"])\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "b96efabf-33a6-42f3-adc0-498538d02166",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mMain | Info ] : \u001b[22m\u001b[39mSolve Time: 0.14812207221984863 [seconds]\n"
     ]
    }
   ],
   "source": [
    "@info \"Solve Time: $(result[\"solve_time\"]) [seconds]\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6aebfe82-3844-4a4f-9ddb-fc7e922f5663",
   "metadata": {},
   "source": [
    "##### Solution Values for Transmission system"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "ad05748e-4891-4dff-a917-d84966807bba",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 7 entries:\n",
       "  \"4\"  => Dict{String, Any}(\"va\"=>1.77189e-30, \"vm\"=>0.9)\n",
       "  \"1\"  => Dict{String, Any}(\"va\"=>3.95511, \"vm\"=>0.91789)\n",
       "  \"5\"  => Dict{String, Any}(\"va\"=>-0.769599, \"vm\"=>0.94138)\n",
       "  \"2\"  => Dict{String, Any}(\"va\"=>-0.65741, \"vm\"=>0.94211)\n",
       "  \"6\"  => Dict{String, Any}(\"va\"=>-0.825474, \"vm\"=>0.940969)\n",
       "  \"10\" => Dict{String, Any}(\"va\"=>5.01852, \"vm\"=>0.904864)\n",
       "  \"3\"  => Dict{String, Any}(\"va\"=>-0.327076, \"vm\"=>0.967421)"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Buses of timestep 1 in the timeseries\n",
    "result[\"solution\"][\"it\"][\"pm\"][\"nw\"][\"1\"][\"bus\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "199c080f-8d93-4423-8c28-d0d1f1238dac",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 7 entries:\n",
       "  \"4\"  => Dict{String, Any}(\"va\"=>-2.34906e-30, \"vm\"=>0.9)\n",
       "  \"1\"  => Dict{String, Any}(\"va\"=>3.95596, \"vm\"=>0.91792)\n",
       "  \"5\"  => Dict{String, Any}(\"va\"=>-0.72648, \"vm\"=>0.941916)\n",
       "  \"2\"  => Dict{String, Any}(\"va\"=>-0.651197, \"vm\"=>0.942305)\n",
       "  \"6\"  => Dict{String, Any}(\"va\"=>-0.745434, \"vm\"=>0.941844)\n",
       "  \"10\" => Dict{String, Any}(\"va\"=>5.01854, \"vm\"=>0.904862)\n",
       "  \"3\"  => Dict{String, Any}(\"va\"=>-0.332464, \"vm\"=>0.967549)"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Buses of timestep 2 in the timeseries\n",
    "result[\"solution\"][\"it\"][\"pm\"][\"nw\"][\"2\"][\"bus\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "e1c1080b-21d9-4064-bd16-1a19432c3d97",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 7 entries:\n",
       "  \"4\"  => Dict{String, Any}(\"va\"=>2.90557e-31, \"vm\"=>0.9)\n",
       "  \"1\"  => Dict{String, Any}(\"va\"=>3.95345, \"vm\"=>0.917828)\n",
       "  \"5\"  => Dict{String, Any}(\"va\"=>-0.856195, \"vm\"=>0.940291)\n",
       "  \"2\"  => Dict{String, Any}(\"va\"=>-0.669593, \"vm\"=>0.941712)\n",
       "  \"6\"  => Dict{String, Any}(\"va\"=>-0.986678, \"vm\"=>0.939193)\n",
       "  \"10\" => Dict{String, Any}(\"va\"=>5.01847, \"vm\"=>0.904867)\n",
       "  \"3\"  => Dict{String, Any}(\"va\"=>-0.315789, \"vm\"=>0.967162)"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Buses of timestep 3 in the timeseries\n",
    "result[\"solution\"][\"it\"][\"pm\"][\"nw\"][\"3\"][\"bus\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "1cba98e3-75a1-4160-b22b-ee43911d6b4c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 7 entries:\n",
       "  \"4\"  => Dict{String, Any}(\"va\"=>1.0386e-28, \"vm\"=>0.9)\n",
       "  \"1\"  => Dict{String, Any}(\"va\"=>3.95146, \"vm\"=>0.917749)\n",
       "  \"5\"  => Dict{String, Any}(\"va\"=>-0.965128, \"vm\"=>0.938896)\n",
       "  \"2\"  => Dict{String, Any}(\"va\"=>-0.684337, \"vm\"=>0.941203)\n",
       "  \"6\"  => Dict{String, Any}(\"va\"=>-1.19036, \"vm\"=>0.936924)\n",
       "  \"10\" => Dict{String, Any}(\"va\"=>5.01842, \"vm\"=>0.904872)\n",
       "  \"3\"  => Dict{String, Any}(\"va\"=>-0.300677, \"vm\"=>0.96683)"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Buses of timestep 4 in the timeseries\n",
    "result[\"solution\"][\"it\"][\"pm\"][\"nw\"][\"4\"][\"bus\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "259e8562-026d-4871-9060-43b72ce8984f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 5 entries:\n",
       "  \"4\" => Dict{String, Any}(\"qg\"=>53.0206, \"pg\"=>19.5984)\n",
       "  \"1\" => Dict{String, Any}(\"qg\"=>30.0, \"pg\"=>40.0)\n",
       "  \"5\" => Dict{String, Any}(\"qg\"=>-204.204, \"pg\"=>460.449)\n",
       "  \"2\" => Dict{String, Any}(\"qg\"=>127.5, \"pg\"=>170.0)\n",
       "  \"3\" => Dict{String, Any}(\"qg\"=>390.0, \"pg\"=>313.247)"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Generators of timestep 1 in the timeseries\n",
    "result[\"solution\"][\"it\"][\"pm\"][\"nw\"][\"1\"][\"gen\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "6375fc55-da1d-49d1-b36c-ad4ac206c6c9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 5 entries:\n",
       "  \"4\" => Dict{String, Any}(\"qg\"=>52.5589, \"pg\"=>19.7762)\n",
       "  \"1\" => Dict{String, Any}(\"qg\"=>30.0, \"pg\"=>40.0)\n",
       "  \"5\" => Dict{String, Any}(\"qg\"=>-204.645, \"pg\"=>460.225)\n",
       "  \"2\" => Dict{String, Any}(\"qg\"=>127.5, \"pg\"=>170.0)\n",
       "  \"3\" => Dict{String, Any}(\"qg\"=>390.0, \"pg\"=>311.283)"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Generators of timestep 2 in the timeseries\n",
    "result[\"solution\"][\"it\"][\"pm\"][\"nw\"][\"2\"][\"gen\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d1bf98b5-b2ec-46e7-8d31-90a6e61a9eac",
   "metadata": {},
   "source": [
    "##### Solution Values for Distribution system(s)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "f7b784e3-7ee6-4acb-ba03-d924557fb948",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 8 entries:\n",
       "  \"3bus_unbal_nogen_mn_diff.substation\" => Dict{String, Any}(\"va\"=>[-0.827403, …\n",
       "  \"3bus_unbal_nogen_mn_diff.sourcebus\"  => Dict{String, Any}(\"va\"=>[-0.826499, …\n",
       "  \"3bus_unbal_nogen_mn.primary\"         => Dict{String, Any}(\"va\"=>[-0.799141, …\n",
       "  \"3bus_unbal_nogen_mn.sourcebus\"       => Dict{String, Any}(\"va\"=>[-0.770623, …\n",
       "  \"3bus_unbal_nogen_mn.loadbus\"         => Dict{String, Any}(\"va\"=>[-0.83023, -…\n",
       "  \"3bus_unbal_nogen_mn_diff.primary\"    => Dict{String, Any}(\"va\"=>[-0.855042, …\n",
       "  \"3bus_unbal_nogen_mn.substation\"      => Dict{String, Any}(\"va\"=>[-0.771526, …\n",
       "  \"3bus_unbal_nogen_mn_diff.loadbus\"    => Dict{String, Any}(\"va\"=>[-0.886158, …"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Buses of timestep 1 in the timeseries\n",
    "result[\"solution\"][\"it\"][\"pmd\"][\"nw\"][\"1\"][\"bus\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "fa90f494-401b-49ba-9bf1-d32700d59a04",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 8 entries:\n",
       "  \"3bus_unbal_nogen_mn_diff.substation\" => Dict{String, Any}(\"va\"=>[-0.746074, …\n",
       "  \"3bus_unbal_nogen_mn_diff.sourcebus\"  => Dict{String, Any}(\"va\"=>[-0.745774, …\n",
       "  \"3bus_unbal_nogen_mn.primary\"         => Dict{String, Any}(\"va\"=>[-0.755989, …\n",
       "  \"3bus_unbal_nogen_mn.sourcebus\"       => Dict{String, Any}(\"va\"=>[-0.727503, …\n",
       "  \"3bus_unbal_nogen_mn.loadbus\"         => Dict{String, Any}(\"va\"=>[-0.787042, …\n",
       "  \"3bus_unbal_nogen_mn_diff.primary\"    => Dict{String, Any}(\"va\"=>[-0.75533, -…\n",
       "  \"3bus_unbal_nogen_mn.substation\"      => Dict{String, Any}(\"va\"=>[-0.728405, …\n",
       "  \"3bus_unbal_nogen_mn_diff.loadbus\"    => Dict{String, Any}(\"va\"=>[-0.765676, …"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Buses of timestep 2 in the timeseries\n",
    "result[\"solution\"][\"it\"][\"pmd\"][\"nw\"][\"2\"][\"bus\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "b9a95137-d1b6-4116-9337-73bcdfc0a252",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 8 entries:\n",
       "  \"3bus_unbal_nogen_mn_diff.substation\" => Dict{String, Any}(\"va\"=>[-0.991215, …\n",
       "  \"3bus_unbal_nogen_mn_diff.sourcebus\"  => Dict{String, Any}(\"va\"=>[-0.989088, …\n",
       "  \"3bus_unbal_nogen_mn.primary\"         => Dict{String, Any}(\"va\"=>[-0.885806, …\n",
       "  \"3bus_unbal_nogen_mn.sourcebus\"       => Dict{String, Any}(\"va\"=>[-0.857221, …\n",
       "  \"3bus_unbal_nogen_mn.loadbus\"         => Dict{String, Any}(\"va\"=>[-0.916966, …\n",
       "  \"3bus_unbal_nogen_mn_diff.primary\"    => Dict{String, Any}(\"va\"=>[-1.05591, -…\n",
       "  \"3bus_unbal_nogen_mn.substation\"      => Dict{String, Any}(\"va\"=>[-0.858127, …\n",
       "  \"3bus_unbal_nogen_mn_diff.loadbus\"    => Dict{String, Any}(\"va\"=>[-1.1292, -1…"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Buses of timestep 3 in the timeseries\n",
    "result[\"solution\"][\"it\"][\"pmd\"][\"nw\"][\"3\"][\"bus\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "f8fc02c0-3466-44f9-9cba-5fabce3fa67c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 8 entries:\n",
       "  \"3bus_unbal_nogen_mn_diff.substation\" => Dict{String, Any}(\"va\"=>[-1.19822, -…\n",
       "  \"3bus_unbal_nogen_mn_diff.sourcebus\"  => Dict{String, Any}(\"va\"=>[-1.19454, -…\n",
       "  \"3bus_unbal_nogen_mn.primary\"         => Dict{String, Any}(\"va\"=>[-0.994826, …\n",
       "  \"3bus_unbal_nogen_mn.sourcebus\"       => Dict{String, Any}(\"va\"=>[-0.966157, …\n",
       "  \"3bus_unbal_nogen_mn.loadbus\"         => Dict{String, Any}(\"va\"=>[-1.02608, -…\n",
       "  \"3bus_unbal_nogen_mn_diff.primary\"    => Dict{String, Any}(\"va\"=>[-1.30979, -…\n",
       "  \"3bus_unbal_nogen_mn.substation\"      => Dict{String, Any}(\"va\"=>[-0.967065, …\n",
       "  \"3bus_unbal_nogen_mn_diff.loadbus\"    => Dict{String, Any}(\"va\"=>[-1.43704, -…"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Buses of timestep 4 in the timeseries\n",
    "result[\"solution\"][\"it\"][\"pmd\"][\"nw\"][\"4\"][\"bus\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ea0a56f6-e399-48fa-89bf-587548ab32da",
   "metadata": {},
   "source": [
    "##### Solution Values for Boundary(ies) (T&D interconnection)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cb211325-9518-48ff-9435-ade6aba7a4e3",
   "metadata": {},
   "source": [
    "###### ------ Boundary #1 ------"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "763cfc0e-d8cc-46de-9fbb-55d6a40d166f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 2 entries:\n",
       "  \"pbound_fr\" => [3008.59]\n",
       "  \"qbound_fr\" => [1298.59]"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Boundary #1 flows from Transmission (Bus #5) to Distribution (3bus_unbal_nogen_mn) for timestep 1 of the timeseries\n",
    "result[\"solution\"][\"it\"][\"pmitd\"][\"nw\"][\"1\"][\"boundary\"][\"(100001, 5, voltage_source.3bus_unbal_nogen_mn.source)\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "f3a87094-d485-4019-ab97-4cd283fac950",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 2 entries:\n",
       "  \"pbound_to\" => [-1204.1, -901.808, -902.676]\n",
       "  \"qbound_to\" => [-400.626, -449.175, -448.787]"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Boundary #1 flows from Distribution (3bus_unbal_nogen_mn) to Transmission (Bus #5) for timestep 1 of the timeseries\n",
    "result[\"solution\"][\"it\"][\"pmitd\"][\"nw\"][\"1\"][\"boundary\"][\"(100001, voltage_source.3bus_unbal_nogen_mn.source, 5)\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "80e4d3d2-a216-4499-bd79-ae95285b5fa0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 2 entries:\n",
       "  \"pbound_fr\" => [3008.58]\n",
       "  \"qbound_fr\" => [1298.57]"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Boundary #1 flows from Transmission (Bus #5) to Distribution (3bus_unbal_nogen_mn) for timestep 2 of the timeseries\n",
    "result[\"solution\"][\"it\"][\"pmitd\"][\"nw\"][\"2\"][\"boundary\"][\"(100001, 5, voltage_source.3bus_unbal_nogen_mn.source)\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "7e4e6a1e-7788-4e1e-b122-bc33a6a5139f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 2 entries:\n",
       "  \"pbound_to\" => [-1204.1, -901.806, -902.673]\n",
       "  \"qbound_to\" => [-400.62, -449.171, -448.784]"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Boundary #1 flows from Distribution (3bus_unbal_nogen_mn) to Transmission (Bus #5) for timestep 2 of the timeseries\n",
    "result[\"solution\"][\"it\"][\"pmitd\"][\"nw\"][\"2\"][\"boundary\"][\"(100001, voltage_source.3bus_unbal_nogen_mn.source, 5)\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "44431acb-fe4b-475f-9bbf-cca8c8b31452",
   "metadata": {},
   "source": [
    "###### ------ Boundary #2 ------"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "6bcb8959-17ce-4a9b-b270-a9cac4e5dfde",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 2 entries:\n",
       "  \"pbound_fr\" => [3008.59]\n",
       "  \"qbound_fr\" => [1298.6]"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Boundary #2 flows from Transmission (Bus #6) to Distribution (3bus_unbal_nogen_mn_diff) for timestep 1 of the timeseries\n",
    "result[\"solution\"][\"it\"][\"pmitd\"][\"nw\"][\"1\"][\"boundary\"][\"(100002, 6, voltage_source.3bus_unbal_nogen_mn_diff.source)\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "e21e6be6-24ef-4a92-a2ca-ea51bc0b60ec",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 2 entries:\n",
       "  \"pbound_to\" => [-1204.1, -901.81, -902.678]\n",
       "  \"qbound_to\" => [-400.63, -449.178, -448.79]"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Boundary #2 flows from Distribution (3bus_unbal_nogen_mn_diff) to Transmission (Bus #6) for timestep 1 of the timeseries\n",
    "result[\"solution\"][\"it\"][\"pmitd\"][\"nw\"][\"1\"][\"boundary\"][\"(100002, voltage_source.3bus_unbal_nogen_mn_diff.source, 6)\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "33b27f7f-cd35-41ce-8c95-822e28335cc2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 2 entries:\n",
       "  \"pbound_fr\" => [1000.95]\n",
       "  \"qbound_fr\" => [427.41]"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Boundary #2 flows from Transmission (Bus #6) to Distribution (3bus_unbal_nogen_mn_diff) for timestep 2 of the timeseries\n",
    "result[\"solution\"][\"it\"][\"pmitd\"][\"nw\"][\"2\"][\"boundary\"][\"(100002, 6, voltage_source.3bus_unbal_nogen_mn_diff.source)\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "a21245b7-fb19-45d7-94f3-d3a6a25aa5ee",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 2 entries:\n",
       "  \"pbound_to\" => [-400.452, -300.2, -300.295]\n",
       "  \"qbound_to\" => [-131.48, -147.986, -147.944]"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Boundary #2 flows from Distribution (3bus_unbal_nogen_mn_diff) to Transmission (Bus #6) for timestep 2 of the timeseries\n",
    "result[\"solution\"][\"it\"][\"pmitd\"][\"nw\"][\"2\"][\"boundary\"][\"(100002, voltage_source.3bus_unbal_nogen_mn_diff.source, 6)\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "ea8a1872-5842-4d4c-b86e-a8538cbdda0d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 2 entries:\n",
       "  \"pbound_fr\" => [7047.38]\n",
       "  \"qbound_fr\" => [3054.79]"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Boundary #2 flows from Transmission (Bus #6) to Distribution (3bus_unbal_nogen_mn_diff) for timestep 3 of the timeseries\n",
    "result[\"solution\"][\"it\"][\"pmitd\"][\"nw\"][\"3\"][\"boundary\"][\"(100002, 6, voltage_source.3bus_unbal_nogen_mn_diff.source)\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "473f2a8d-f412-4fe0-8174-6203abe307fe",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 2 entries:\n",
       "  \"pbound_to\" => [-2822.68, -2109.95, -2114.75]\n",
       "  \"qbound_to\" => [-946.558, -1055.2, -1053.04]"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Boundary #2 flows from Distribution (3bus_unbal_nogen_mn_diff) to Transmission (Bus #6) for timestep 3 of the timeseries\n",
    "result[\"solution\"][\"it\"][\"pmitd\"][\"nw\"][\"3\"][\"boundary\"][\"(100002, voltage_source.3bus_unbal_nogen_mn_diff.source, 6)\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "5dd9631a-1ac1-4385-b88a-5e0c0c40dccd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 2 entries:\n",
       "  \"pbound_fr\" => [12141.4]\n",
       "  \"qbound_fr\" => [5276.86]"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Boundary #2 flows from Transmission (Bus #6) to Distribution (3bus_unbal_nogen_mn_diff) for timestep 4 of the timeseries\n",
    "result[\"solution\"][\"it\"][\"pmitd\"][\"nw\"][\"4\"][\"boundary\"][\"(100002, 6, voltage_source.3bus_unbal_nogen_mn_diff.source)\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "bb210edb-1125-4d2a-8fd0-2654de73aa4a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dict{String, Any} with 2 entries:\n",
       "  \"pbound_to\" => [-4867.93, -3629.56, -3643.95]\n",
       "  \"qbound_to\" => [-1643.84, -1819.75, -1813.27]"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Boundary #2 flows from Distribution (3bus_unbal_nogen_mn_diff) to Transmission (Bus #6) for timestep 4 of the timeseries\n",
    "result[\"solution\"][\"it\"][\"pmitd\"][\"nw\"][\"4\"][\"boundary\"][\"(100002, voltage_source.3bus_unbal_nogen_mn_diff.source, 6)\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "148985b2-f2fb-4de0-8713-41e415da04d2",
   "metadata": {},
   "source": [
    "### -----------------------------------------------------------------------------------------------------------\n",
    "### QUIZ #3: let's solve the Optimal Power Flow ITD (OPFITD) using different formulations and compute optimality gaps.\n",
    "\n",
    "Let's solve the OPF using other formulations:\n",
    "\n",
    "- `ACRPowerModel-ACRUPowerModel` (NLP) -- AC rectangular coordinates --> `NLPowerModelITD{ACRPowerModel, ACRUPowerModel}`\n",
    "- `BFAPowerModel-LinDist3FlowPowerModel` (APPROXIMATION) -- Branch flow - LindDist3Flow approximation --> `BFPowerModelITD{BFAPowerModel, LinDist3FlowPowerModel}`\n",
    "- `NFAPowerModel-NFAUPowerModel` (APPROXIMATION) -- Network flow approximation -->  `LPowerModelITD{NFAPowerModel, NFAUPowerModel}`\n",
    "- `IVRPowerModel-IVRUPowerModel` (NLP) -- Current-Voltage rectangular coordinates --> `IVRPowerModelITD{IVRPowerModel, IVRUPowerModel}`\n",
    "\n",
    "**Note**: the optimality gap is normally used to compare NLPs with approximations/relaxations. NLPs vs NLPs should give a value close to 0.0%\n",
    "\n",
    "**Note 2**: Make sure that the solver terminates nominally!, i.e., `result[\"termination_status\"] == OPTIMAL` ||  `result[\"termination_status\"] == LOCALLY_SOLVED`)\n",
    "\n",
    "I will provide an example below between **ACP-ACPU vs. ACR-ACRU**\n",
    "\n",
    "See all currently supported formulations [here](https://lanl-ansi.github.io/PowerModelsITD.jl/stable/manual/quickguide.html#Accessing-Different-Formulations) or [here](https://lanl-ansi.github.io/PowerModelsITD.jl/stable/manual/formulations.html)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "id": "b24f6d0f-354b-47a9-9d4e-cd649cc9fb35",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mPowerModelsITD | Info ] : \u001b[22m\u001b[39mpmitd model solution time (instantiate + optimization): 0.03621315956115723\n",
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mMain | Info ] : \u001b[22m\u001b[39mNLP: 17953.746512049685\n",
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mMain | Info ] : \u001b[22m\u001b[39mNLP: LOCALLY_SOLVED\n",
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mPowerModelsITD | Info ] : \u001b[22m\u001b[39mpmitd model solution time (instantiate + optimization): 1.876512050628662\n",
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mMain | Info ] : \u001b[22m\u001b[39mRELAX: 17953.746516911073\n",
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mMain | Info ] : \u001b[22m\u001b[39mRELAX: LOCALLY_SOLVED\n",
      "\u001b[36m\u001b[1m[ \u001b[22m\u001b[39m\u001b[36m\u001b[1mMain | Info ] : \u001b[22m\u001b[39mOptimality Gap: -2.7077286742954975e-8 %\n"
     ]
    }
   ],
   "source": [
    "# solver\n",
    "ipopt_solver = optimizer_with_attributes(Ipopt.Optimizer, \"print_level\" => 0)\n",
    "\n",
    "# Define the path to be the PowerModels path (optional)\n",
    "files_path = joinpath(dirname(pathof(PowerModelsITD)), \"..\")\n",
    "pm_file = joinpath(files_path, \"test/data/transmission/case5_withload.m\")\n",
    "pmd_file = joinpath(files_path, \"test/data/distribution/case3_unbalanced.dss\")\n",
    "boundary_file = joinpath(files_path, \"test/data/json/case5_case3_unbal.json\")\n",
    "\n",
    "# Base nlp to compare\n",
    "formulation = PowerModelsITD.NLPowerModelITD{ACPPowerModel, ACPUPowerModel} ## Formulation!!!\n",
    "data_nlp = PowerModelsITD.parse_files(pm_file, pmd_file, boundary_file)\n",
    "result_nlp = solve_opfitd(data_nlp, formulation, ipopt_solver)\n",
    "@info \"NLP: $(result_nlp[\"objective\"])\"\n",
    "@info \"NLP: $(result_nlp[\"termination_status\"])\"\n",
    "\n",
    "# Approx. or Relax. to compare (in this case is another nlp formulation)\n",
    "formulation = PowerModelsITD.NLPowerModelITD{ACRPowerModel, ACRUPowerModel} ## Formulation!!!\n",
    "data_relax = PowerModelsITD.parse_files(pm_file, pmd_file, boundary_file)\n",
    "result_relax = solve_opfitd(data_relax, formulation, ipopt_solver)\n",
    "@info \"RELAX: $(result_relax[\"objective\"])\"\n",
    "@info \"RELAX: $(result_relax[\"termination_status\"])\"\n",
    "\n",
    "op_gap = optimality_gap(result_nlp, result_relax) # result is in %\n",
    "\n",
    "@info \"Optimality Gap: $(op_gap) %\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "986304ad-ea42-44f0-9b28-1864fceac26d",
   "metadata": {},
   "source": [
    "## Other Test Cases\n",
    "\n",
    "You can find other test cases to run and experiment in the [Beginners Guide of PowerModelsITD](https://lanl-ansi.github.io/PowerModelsITD.jl/stable/tutorials/BeginnersGuide.html)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "64e60ab3-1144-4efe-8270-da5151e5b11e",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Julia 1.11.2",
   "language": "julia",
   "name": "julia-1.11"
  },
  "language_info": {
   "file_extension": ".jl",
   "mimetype": "application/julia",
   "name": "julia",
   "version": "1.11.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}