Wavelet-adaptive ANN forecaster for renewable energy sources for continuous supply in microgrid applications

In this paper, the performance of hybrid power system (HPS) with high penetration of renewable energy sources (RES) was investigated under dominant weather conditions. Hourly solar radiation and wind speed were forecasted for one week ahead (168h) using wavelet - adaptive feed forward artificial neural network. The load was forecasted for the same time horizon. Based on these forecasts, the supervisory control calculates available power from the installed PV modules and wind turbines then send the required reference signal to the voltage source inverter (VSI). The VSI will control the power flow at the point of coupling to guarantee continuous power supply to the loads. For better understanding of the interactions of the microgrid with the main AC grid under weather conditions and to validate the effectiveness of the system, an experiment was carried out in a laboratory based smart power system. The controller response and consequently, power flow were monitored, controlled and discussed.