Ultra\|Short\|term Forecasting for Photovoltaic Power Output Based on Least Square Support Vector Machine

With the connecting of large-scale photovoltaic(PV)power station to the distribution network, it is necessary to strengthen the study of photovoltaic power output prediction in order to mitigate the impacts of randomness on power system.An ultra-short-term forecasting model based on least square support vector machine(LS-SVM)is proposed.To predict the PV output for every quarter ahead of 1h,the inputs of the model are the latest meteorologic information.To accurately reflect the weather condition of predicted day, a proper weight value is set to each meteorological factor which affects the PV output,then the training samples are determined by calculating the weighted Euclid distance. In the end, the trained model is tested and evaluated by using weather data with sudden changes.The results show that the proposed model has high precision, and can provide reference for dispatching department to formulate reasonable schedule.