Monthly Electricity Forecast Based on Electricity Consumption Characteristics Analysis and Multiple Effect Factors

The accurate prediction of users' monthly electricity is the basis for electric power department to allocate power resources and for electric power company to make reasonable sales plans. Based on the in-depth mining of historical electricity data and comprehensively electricity consumption characteristics analysis, a monthly electricity forecast method considering gross domestic product(GDP), temperature and Spring Festival is proposed. Firstly, the prediction model is built based on historical electricity, GDP and temperature data using Elman neural network. Then, the monthly electricity data affected by the Spring Festival is modified to obtain a more accurate monthly electricity prediction result. The monthly electricity data of A city is used to do the verification. By comparing the forecast error of Elman neural network without considering effect factors with the forecast error of proposed algorithm, the accuracy of the proposed algorithm is verified to be improved. It verifies the effectiveness of the prediction algorithm．