GridSight

The energy transition is a multi-decade megatrend that guarantees one thing: increasing market complexity and volatility. The value in this new market will accrue to those who can navigate this complexity most effectively.

GridSight is a web-based intelligence tool for this new era, providing the speed of insight necessary to manage risk, identify opportunities, and trade with confidence.

Project description

GridSight visualises public data and helps analyse historic and current scenarios in a way that can lead to forming better hypotheses and making more informed trading decisions.

It consists of two core, complementary views: a Live Plant Status Board that answers the tactical question, "What is happening to every major asset on the grid right now?" and a Historical Volatility Analyser which answers the strategic question, "What happened to prices and generation the last time conditions were exactly like this?"

Core features

1. Live Plant Status Board (the "Bingo Board"): A real-time, tile-based dashboard displaying the operational status of every major UK power generation unit. Each tile is colour-coded to show its status (e.g., Generating, Standby, Balancing, Outage) and provides key data at a glance: real-time output vs. planned (FPN) and maximum availability (MEL).

2. Historical Volatility Analyser: An interactive dashboard which allows users to query years of historical grid data. Its primary feature is a Correlation Filter that isolates specific market conditions (e.g., low wind, high demand) and instantly displays the resulting average generation mix, imbalance prices, and thermal profitability.

3. Indicative Spark Spread Calculator: An integrated feature that calculates and visualises the indicative profitability of gas-fired power plants, incorporating user-defined assumptions for thermal efficiency.

4. System Headroom View: A macro-level chart that aggregates MEL and FPN data for the entire CCGT fleet, providing a clear, time-series view of the total spare capacity (headroom) available on the grid.

Getting Started

This guide will walk you through setting up the GridSight project for local development.

Prerequisites

• Python 3.12
• PostgreSQL 16 (or later).
• The TimescaleDB extension must be installed and enabled on your PostgreSQL instance.

Setup Instructions

1. Clone the repository:

   git clone [email protected]:a115/gridsight.git
   cd gridsight

2. Create a virtual environment, activate it and install dependencies:

   python -m venv .venv
   source .venv/bin/activate
   pip install -r requirements.txt
   pip install -r requirements-dev.txt

3. Configure your environment: Create a .env file in the project root by copying the example file. This will store your database credentials and other secrets / env vars.

   APP_NAME=gridsight
   
   DJANGO_SECRET_KEY=your-secret-key-here
   DJANGO_DEBUG=True
   
   # Make sure this is aligned with the docker compose file
   POSTGRES_USER=postgres
   POSTGRES_PASSWORD=postgres
   POSTGRES_DB=postgres
   POSTGRES_HOST=localhost
   POSTGRES_PORT=5433
   
   # Once you have a BMRS API key
   BMRS_API_KEY=your-cool-and-secret-bmrs-key
   
   # Once you have a Logfire token
   LOGFIRE_TOKEN=your-cool-and-secret-logfire-token
   

4. Run database migrations and collect static files: This will create all the necessary tables in your database, including the TimescaleDB hypertables.

   python manage.py migrate
   python manage.py collectstatic

5. Run the development server:

   python manage.py runserver

   The application should now be running at http://127.0.0.1:8000/.

6. Fetch initial data (Optional but Recommended): To populate your local database with some sample data, you can use the custom management command.

   # TODO