Research On Short-term Power Load Forecasting Based On Improved Least Square Support Vector Machine

Accurate and efficient short-term power load forecasting is an important prerequisite and powerful guarantee for realizing the economics,flexibility and normal production and life of power companies.In recent years,with the rapid development of the smart grid and the large number of distributed power sources in the power supply network,the power load has increased rapidly and the power consumption environment has become more complex.Traditional prediction methods have been unable to meet the actual needs of the power system at this stage.Therefore,it is of great significance to study short-term power load forecasting methods that are more suitable for smart grids.This paper first analyzes the historical load data based on the characteristics of the power supply network in Xi’an,and focuses on the main factors that affect the load characteristics and their influencing laws to provide theoretical support for subsequent predictions.Secondly,methods for processing and improving abnormal load data are proposed,and neighborhood rough sets are used to perform attribute reduction on the processed load data and various influencing factors to reduce the impact of redundant factors on prediction accuracy and speed.In order to solve the problems of low prediction accuracy and slow convergence speed due to the selection of kernel parameters σ and disciplinary parameters η in the least squares support vector machine prediction model,this paper proposes an improved whale algorithm to optimize the prediction of the least square support vector machine method.The introduction of nonlinear convergence factors and adaptive weights to improve the traditional whale optimization algorithm to update the position of the population formula,thereby enhancing the ability of the whale algorithm to coordinate global search and local mining,and then use the standard test function to optimize the performance of the improved method Simulation.Finally,the IWOA-LSSVM load forecasting model based on the above theory is built,and the reduced simulation data is used to carry out the forecasting simulation test.The test results show that the IWOA-LSSVM load forecasting model improves the forecasting accuracy while enhancing the forecasting convergence and stability,and provides a reliable basis for the rational dispatching and stable operation of the power system.