Research On Short-term Power Load Forecasting Method Based On Neural Network

Short-term load forecasting has become an important basis for the safe and stable operation of power systems and the maintenance of supply and demand balance.With the rapid growth of load in recent years and the diversification of power grids,power supply tensions have intensified,causing various power companies to face economic and technological challenges.Therefore,the study of short-term load forecasting has important practical significance in grid operation and scheduling optimization.In order to improve the prediction accuracy and prediction efficiency of short-term power load forecasting,this paper proposes a prediction method based on improved artificial bee colony algorithm to optimize BP neural network.Firstly,the data of electric load and meteorology in Xi'an area are analyzed in detail.The multivariate regression statistical method is used to analyze the correlation between load and meteorological factors,and the key influencing factors of load forecasting are selected to provide theoretical support for the subsequent integration model.Secondly,the data is preprocessed by the data bidirectional comparison method to ensure the rationality and integrity of the data in the modeling and prediction process.Finally,aiming at the initial value sensitivity of BP neural network and the vulnerability to local extremum,an improved artificial bee colony algorithm(IABC)is proposed to optimize BP initial parameter weights and thresholds.The improved artificial bee colony algorithm uses four standard functions.Test and verify the advancement and feasibility of the improved algorithm.The IABC-BP prediction model was constructed by determining the optimal parameters of the network,and the Xi'an ground power load and meteorological data were used for case prediction.The results show that the average absolute percentage error of the IABC-BP model is 1.353%,which is 4.570% and 2.279% lower than the BP model and the ABC-BP model,and the root mean square error is 1.9%.The optimization algorithm is used to get the optimal initial parameters of the BP network,so the convergence speed is further improved.It can be seen that the prediction results under the optimization model have the best accuracy and stability and can be applied to the actual power grid.