The project team
The intraday power market is very dynamic due to the large number of products, short maturities, and very volatile liquidity. To understand market prices, it is necessary to consider large amounts of data from various sources (markets, weather, production park, fuel prices, regulations, congestions, …)
Market participants are dreaming of high-quality automated price prediction models to increase their competitiveness, and current approaches such as fundamental or technical analysis often struggle to capture real price drivers.
Our technical innovation lies in the use of machine learning, as a way to empower the trader, by giving him a third-eye view on what is currently happening in the market.
Instead of building a single model to analyze the market, we build thousands of such models simultaneously and dynamically assign to each of them a specific influence in the prediction, depending on its ability to extract knowledge from current observable market conditions.
This advanced machine-learning model has been built cooperatively by traders, analysts and data scientists. It is running in production and has already been generating significant positive revenue.
With increased digitalization and available data, the interest for data-driven processes that extract value from the complexity of our data is rising.
Since results of this project have been published, various business units within the group are contacting us. This project really opened up new horizons in the group: machine learning is not only something people talk about at ENGIE, we are using it to generate new revenue streams!
Sign up for the ENGIE Innovation Newsletter