Power System Short-term Load Forecast Research Based On Neural Network

Power system short-term load forecast is the basis of power system optimization running.It can affect safety property,reliability and economy of power system operation.Thus,to find effective method has important meaning for enhancing forecast precision.So far,researchers have come up with many effective methods.This paper used a popular method---neural network to forecast short-term load of power system.The main work included:1.Looking into the present condition of power system short-term load forecast and summarizing research method in the world.2.Study about the knowledge of neural network,designed BP network based on conjugate gradient algorithm used in load forecast and results showed the approach was feasible.Compared with common BP network.It shortened training time and enhanced forecasting precision.3.Studying wavelet knowledge,attempting to combine wavelet and neural network to design WNN(Wavelet Neural Network).The result showed this approach also feasible.4.Creatively bring forward a new type neural network:RAN(Resource Allocating Network).It can add or delete it's hidden layer neural unites based on the input data's complexity.Attempt use it in load forecast,also have better effect,and compared above three approaches.5.At last,summarizing the work had done,coming up with some improved proposals and introducing development possibility of short-term load forecast of power system.