Short-Term Power Load Forecasting Based On Ant Lion Algorithm Optimized Extreme Learning Machine

With the development of the power market,the short-term load forecasting of the power system will directly affect the decision-making of the power market and the dispatching of the power grid.For the power grid,accurate short-term load forecasting can make effective power generation plans and avoid unnecessary waste of power.For the power market,it is very important to improve the accuracy of short-term load forecasting because the power sales plan can be made according to the load forecasting results,thus maximizing the benefits of all power enterprises.Therefore,on the basis of analyzing the influencing factors of short-term load forecasting and comparing with the existing forecasting models,this paper proposes a short-term load forecasting model of extreme learning machine optimized by ant lion algorithm.First of all,this paper introduces the research status,overview and classification of power load forecasting in detail,analyzes the main factors affecting the change of power load,and summarizes the characteristics of daily load and holiday load.Secondly,the basic principles of several classical model algorithms of artificial neural network are introduced.Through the analysis of examples,the single model with the most accurate and stable prediction in artificial neural network model is found,and then the shortcomings of the single model are studied.Therefore,a solution using swarm intelligence optimization algorithm is proposed.Then,the basic principle of ant lion algorithm is introduced,and an ant lion algorithm optimization extreme learning machine model is constructed.Compared with the improved particle swarm optimization extreme learning machine model in swarm intelligence optimization algorithm,the superiority and effectiveness of the method are verified through application examples.Finally,the ant lion algorithm optimization extreme learning machine is applied to short-term power load forecasting,and the case analysis proves that it has higher forecasting accuracy.