Research On Short-term Power Load Forecasting Based On BFA-Elman Neural Network

As an important lifeline of the national economy,electricity plays an irreplaceable role in maintaining national survival and maintaining the national economy and people's livelihood.Especially with the advancement of modernization,the dependence of national production and life on electric energy is getting stronger and stronger,and the importance of the power system is becoming more and more prominent in the present.Therefore,the accurate prediction of the load not only ensures the safe and stable operation of the power system,but also helps to optimize the overall resource allocation and alleviate the energy pressure,which is of great significance.In order to improve the short-term electric load forecasting accuracy,this paper aims at the existing problems,based on the theoretical basis and research results at home and abroad,combined with the existing data for the short-term load forecasting experimental simulation.The paper mainly studies the following contents.First,this paper introduces the domestic and international research status of power market,clustering method and power load forecasting method,and finds the main problems of short-term power load forecasting.Secondly,according to the characteristics of the electric load,the traditional electric load classification,and the influence of temperature and seasonal holidays on the electric load,the raw data is subjected to horizontal and vertical processing of missing data and abnormal data.Then,the fuzzy C-means clustering is combined with the specific data for case simulation analysis,and the results are evaluated based on the cluster validity index,and the clustering results are obtained.Finally,the combination of bacterial foraging algorithm and Elman neural network is applied to short-term load forecasting,and compared with wavelet neural network and RBF neural network.It has a good predictive effect.Through the study of the above content,this is the first time that the BFA-Elman neural network has been applied to power load forecasting.The simulation results show that the BFA-Elman algorithm improves the accuracy of short-term load forecasting and provides a new idea for future short-term load forecasting.The research results have certain reference significance,which provides a reference for the next step to establish the relationship between load forecasting and electricity price forecasting.