The Cost of Photovoltaic Forecasting Errors in Microgrid Control with Peak Pricing

Model predictive control (MPC) is widely used for microgrids or unit commitment due to its ability respect the forecasts of loads and generation renewable energies. However, while there are lots approaches accounting uncertainties in these forecasts, their impact rarely analyzed systematically. Here, we use a simplified linear state space model commercial building including photovoltaic (PV) plant real-world data from 30 day period 2020. PV predictions derived weather industry peak pricing assumed. The effect prediction accuracy on resulting cost evaluated by multiple simulations with different errors initial conditions. Analysis shows mainly correlation, exact shape depends treatment at current time step. Furthermore, despite horizon 24h, only first 75min was relevant presented setting.