Ultra-short-term forecasting of microgrid surplus loadbased on hybrid intelligence techniques

As the accurate forecasting of distributed generation and short-term load is the basis of operational control and energy management of microgrid,the concept and calculation method of microgrid surplus load are proposed and the features and influencing factors of its ultra-short-term forecasting are discussed.With the consideration of historical power outputs,loads and weather data,an ultra-short-term forecasting model of microgrid surplus load is developed,which integrates k-means clustering analysis,genetic algorithm and artificial neural network.A simulation model of microgrid with wind farms,micro-turbines and fuel cells is established and results show that,the forecasting outputs of distributed generators and the forecasting load of microgrid are very close to the measured data,which verifies the accuracy of the proposed forecasting model.