A Novel Ultra Short-Term Load Forecasting Method for Regional Electric Vehicle Charging Load Using Charging Pile Usage Degree

Electric vehicle (EV) charging load is greatly affected by many traffic factors, such as road congestion. Accurate ultra short-term forecasting (STLF) results for regional EV are important to the scheduling plan of load, which can be derived realize optimal grid benefit. In this paper, a regional-level STLF method proposed and discussed. The usage degree all piles firstly defined us based on frequency piles, then constructed our collected transaction data in field. Secondly, these degrees combined with historical values form input matrix deep learning prediction model. Finally, long memory (LSTM) neural network used construct model, trained formed matrix. comparison experiment proves that paper has higher accuracy compared traditional methods. addition, characteristic index fluctuation adjacent day week us, factors assess together mean absolute percentage error (MAPE).