Accurate forecasting of electricity production by solar power plants enables PVPP owners to minimize costs for imbalances regulation.
To increase the accuracy of PV generation forecasts, we use artificial intelligence within the PV Forecast software module, of our RnD team development. Forecasting system error for a new plant is about 20-25% for PV generation forecasting for the next day. Forecast generation schedule refinement allows to significantly reduce such an error. As the service life of the complex increases, the forecast error decreases due to the continuous training of the neural network.
PV generation forecasting comprehensive solution by KNESS includes:
Forecasting the generation of PVPP from 2 hours to 7 days in advance, with the help of artificial intelligence.
Delivering forecasted data in any format to any platform.
Training of artificial intelligence individually for each station.
Taking into account the operational data of the PVPP service by using own software (PV.SCADA).
Reliability of saving meteorological forecasts` parameters and service data on Cloud Storage with mirrors.