Short Term Power Load Forecasting Based On Improved Whale Algorithm

The accuracy of load forecasting is related to whether the social production and living order can be carried out normally.For power enterprises,relatively high accuracy of load forecasting can effectively reduce power generation costs and improve the economic benefits of enterprises.In the current situation of social development,in order to improve the level of short-term load forecasting,we should work hard on the factors affecting load forecasting and the selection of forecasting methods that keep pace with the times.With the smart grid and digital grid proposed by the State Grid Corporation of China,all major enterprises in the society are carrying out digital transformation.The relatively easy access of various influencing load forecasting data provides the basis for adopting more advanced and complex multi factor forecasting methods.BP neural network is one of the hot directions at present.Based on the short-term power load forecasting in Dalian,according to the changes of regional power grid,combined with the actual work and the requirements of current regional power grid short-term load forecasting.(1)This paper studies the current research and development status and short-term load characteristics of short-term load forecasting at home and abroad.At the same time,it goes deep into theoretical research and combines with practice to study how different factors such as climate characteristics,temperature and humidity,date type and so on are incorporated into the short-term load forecasting model,and collects and arranges the relevant influence characteristics of observation as the attribute of forecasting historical samples.(2)An adaptive weight adjustment method is proposed to improve the whale optimization algorithm.On the basis of consolidating the inherent efficient and stable convergence of the algorithm,the global convergence is further improved,and the convergence speed and optimization ability are further increased.Then the improved whale optimization algorithm is combined with the original neural network to propose bp-gwoa algorithm.The combination of the two can effectively improve the prediction accuracy and reasonably solve the problems existing in the prediction of traditional neural network.(3)By analyzing the results of power grid short-term load forecasting,the maximum error of the improved algorithm is 1.126%,which is lower than the minimum error of the basic algorithm of 2.31%.It is proved that the improved whale optimization algorithm proposed in this paper has a good guiding role in improving the ability of short-term load forecasting.