Solar Data Retrieval and Analysis

This project aims to retrieve solar data from the National Solar Radiation Database (NSRDB) API for a specific location on Earth and perform machine learning predictions using the obtained data. The project consists of a Flask backend that interacts with the NSRDB API and processes the retrieved solar data, and a frontend built with Next.js and React that provides a user interface for inputting location coordinates and selecting desired datasets.

Features

• Retrieve solar data from the NSRDB API based on user-provided latitude and longitude coordinates.
• Allow users to select specific datasets, years, and intervals for data retrieval.
• Process the retrieved solar data using pandas and calculate solar generation potential using the SolarTKMaxPowerCalculator class.
• Perform machine learning predictions based on the processed solar data (to be implemented).
• Provide a user-friendly frontend interface for inputting location coordinates and selecting datasets.

Prerequisites

To run this project locally, you need to have the following installed:

• Python 3.x
• Node.js
• bun

Getting Started

Clone the repository:

git clone https://github.com/your-username/fall23-solar-tk-next.git

Install the required Python dependencies:

cd server
conda install -r requirements.txt

Install the required frontend dependencies:

cd client
bun install

Obtain an API key from the NSRDB API here

Start the Flask backend server:

cd ..
python proto.py

In a separate terminal, start the Next.js frontend development server:

cd client
bun dev

Open your web browser and navigate to http://localhost:3000 to access the frontend interface.

Usage

• On the frontend interface, enter email, NSRDB api key, latitude, and longitude coordinates of the desired location.
• Select the desired dataset, years, and intervals for data retrieval.
• Click the "Submit" button to initiate the data retrieval process.
• The backend server will retrieve the solar data from the NSRDB API, process it, and perform machine learning predictions (to be implemented).
• The results will be displayed on the frontend interface (to be implemented).