Power price forecast play a key role in operations of all actors across the power value chain. Utilities set the operation point of their generating fleet according to the expected power price and demand; distribution system operator and provider will base their sourcing strategy and tariffs on forecast of prices for several power products on the EPEX Spot and EEX stock exchange.
Based on scientific literature on power market modeling, this bot offers forecasts for spot and futures prices in the French market, using two years of spot prices and temperature history as well as futures prices over the forecast period. Forecasts are computed in three steps. A first component is computed to capture seasonal and temperature effects and a second component is computed to capture the influence of futures prices over the forecast period: this component result in the initial forward curve. Based on this curve, a 2-factor diffusion model is applied, leading to an envelope of solutions: the median is considered as the forecast, and a quantile envelope can be set to assess the uncertainty of the forecast.
If you want to know more about our AI solutions, check out our Heka website: https://heka.sia-partners.com/