End User
EMOT
Description
The Localised PV Generation Forecasting service provides high-resolution photovoltaic generation forecasting for energy communities, combining data-driven machine learning with physics-informed feature engineering. The service produces deterministic and probabilistic forecasts at 15-minute resolution for short-term and day-ahead horizons, enabling community operators to anticipate renewable availability and improve scheduling of flexible loads.
Core Capabilities
Predictive & Prescriptive Analytics
Business Need
PV output exhibits strong intra-day variability driven by weather conditions. Without reliable generation forecasts, energy communities experience unexpected exports, inefficient charging schedules, and increased reliance on external electricity suppliers. Accurate forecasts allow operators and downstream optimisation services to align consumption with renewable availability, improving system efficiency and supporting collective self-consumption objectives.
Key Performance Indicators
Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and R²
Forecast bias and ramp prediction accuracy
Comparison against persistence baselines
Forecast generation latency and endpoint availability
Data Provided
15-minute time-indexed PV power production forecasts per energy community
Probabilistic uncertainty intervals (where enabled)
Derived indicators: expected peak generation timing, ramp-rate estimates, clear-sky reference profiles
Inputs: historical PV measurements, meteorological forecasts (irradiance, temperature, cloud cover), PV system metadata
TEF
TEF EV