Power System Load Prediction Based On Wavelet And RBF Neural Network

The power industry is the lifeblood of the country,and is closely related to the development of the country’s politics,economy and people’s livelihood.Load forecasting plays an important role in the power industry.Load forecasting is related to the daily operation of the power system and the control of the power system.Accurate load forecasting can not only ensure the power consumption for daily life and production,but also reduce the cost of power generation.It is also an important index to control the power supply plan of each region and maintain the stable and safe operation of the power grid.However,many factors affecting power load,such as season,environment,political and economic policies,will affect the result of load forecasting,thus increasing the difficulty of forecasting.This paper expounds the background and significance of power system load forecasting research,analyzes the research situation and analyzes the advantages and disadvantages of various prediction methods.The load prediction is classified and the load characteristics,influencing factors and prediction error are analyzed.Study the algorithm and characteristics of RBF neural network,and construct a power system load prediction model using RBF neural network.For the low load prediction accuracy of power system,RBF neural network method is used to predict power load,design model structure and parameters,and predict the long-term load and long-term load of some region.Through Matlab simulation,the load prediction accuracy is high.Considering the impact of seasonal,environmental and other factors on the load forecast results,bad data reduces the accuracy of load prediction.The method combining wavelet and RBF neural network,using wavelet to handle bad load data as the input of the load prediction model,through Matlab simulation analysis and compared to the simulation results of RBF method,the two methods can effectively improve the load prediction accuracy.