Project Definition

Project Description:

• Host a web-based coastal forecast machine learning model with a goal of using the website as the user interface. Wind wave data to be pulled from selected NOAA stations to be updated at a predetermined interval. NOAA station 41013, Frying Pan Shoals, will be used as a base case during offline set-up of the script for the machine learning model.

Team Communication:

• Email
• Zoom

Members and Role Assignments:

• Dr. Zhangping Wei - Project Manager
• Andrew Davison - Lead Developer

Feature List:

• The system will scrape NOAA station data on a predetermined interval (every 6 hours).
• The system will clean the raw data and prepare for use in the model.
• The system will plot forecast data and ground truth NOAA measurement for display in the application.
• Prediction plot will update on the backend.
• The user can select from a list of available NOAA stations to see predicted data for that specific station.
• Functionality for users to request a current prediction plot that does not affect the predetermined interval.
• The application will retrieve the current prediction display for the user selected location, location default will be station 41013.

Data Required:

• NOAA station raw data.
• Optimized and functional machine learning model.
• Prediction plot storage.

Platform:

• Web-based platform

Implementation Technologies:

• Google Cloud Platform
• AWS
• Flask

User Roles:

• User - individual who will access application