Research on Long-Term Load Forecasting of the Power System Based on Multi-Factor Combination Analysis
Many factors influence long-term load forecasting of the power system,but the accuracy of the load forecasting using only a single factor is often low.Therefore,a power system long-term load forecasting method based on multi factor combination analysis is proposed in this paper.After the power load data is collected through the discernibility matrix,the ACO-PAM comprehensive algorithm is used to cluster the power data to obtain valuable load data.After the power load data obtained by clustering is quantified by data factors and non data factors,the correlation between multiple factors and load is analyzed,and the obtained multiple factors are used as the input of genetic algorithm to improve neural network to output the long-term load forecasting results of the power system.The experimental results show that under the influence of many factors,the long-term load forecasting result of the power system based on this method is close to the actual value.Compared with the two comparison methods,the average absolute error of this method is 1 309 800 and 416 500 tons of standard coal respectively,and the average relative error is 3.77%and 1.19%less than the two comparison methods,respectively.
multiple factorscombination analysispower systemlong term load forecastingdata clusteringneural network
NETL
NSTL
万方数据
