Wind Power Forecasting Based on BP Neural Network

With the increasing social demand of electric power, electric power companies need to make accurate and reasonable dispatching plans for electric power resources in order to meet the needs of various users. Considering only historical load data and five meteorological factors and date types, the BP neural network model is constructed to simulate and forecast the future seven-day power load. And compare it to the real thing. By comparison and analysis, it is concluded that the forecast accuracy can be improved by considering meteorological factors and date types, which provides an effective reference for decision-makers of electric power companies to make short-term electric power production plan and electric power dispatching arrangement.