Research Of Short-term Load Forecasting Of Anyang Power Grid

The power load is the key to the normal operation of a region and the lifeblood of local economic development.For the power grid,load forecasting is an important part of the whole power grid system scheduling and planning.Timely and accurate forecasting of short-term load of the power grid can ensure the normal operation of the power grid and the stability of regional production and life.With the rapid development of China's economy,the demand for power in urban areas is increasing,the power grid structure is becoming more complex,and the load nature is becoming more variable.The establishment of a reasonable short-term load forecasting model is conducive to improving the operation efficiency and power supply capacity of the power grid.This paper summarizes the traditional and intelligent power load forecasting methods.In the establishment of the short-term load forecasting model of power grid,it is necessary to correctly and comprehensively understand the various factors affecting the power grid load.Therefore,we collected the economic and social development,historical power grid load data,and basic power grid data of anyang area,and analyzed the power structure and load characteristics of Anyang area.Taking weather factor,population factor,emergency factor and time factor into full consideration,an accurate short-term load forecasting model of Anyang power grid is established through the neural network model.In order to obtain the ideal prediction effect and improve the prediction accuracy and applicability of the model,the following work was done in this paper :(1)abnormal data processing was conducted on the historical power grid load data,and the load data and various influencing factors were normalized.(2)to determine appropriate neural network topology and parameters,the model of through continuous learning,constantly to some data for calibration,get some abnormal data points,so that by comparing the difference between predicted values and actual values,selective model improvement,constantly improve the prediction accuracy of the models.(3)in order to achieve rapid prediction and improve the prediction efficiency as far as possible,the data of each node of the neural network model are parallelized.Through experimental verification,it is found that the accuracy of the model software is in line with our expectations,which can be used in the short-term load forecasting analysis of Anyang power grid.