Research on power prediction based on significance influencing factor screening and PKO-RBF
ZHOU Ying
BAI Xuefeng
WANG Yongli
QIAN Xiaorui
ZHANG Jiapu
LI Yiming
WEI Yusi
Abstract：The impact of climate change and the irregularity of electricity consumption changes in recent years have put forward higher requirements for electricity consumption forecasting,and traditional forecasting methods cannot achieve ideal forecasting results.In order to effectively capture the key features of time series data and improve the accuracy of electricity consumption prediction,this study proposes an electricity consumption prediction method based on significance influencing factor screening and Pied Kingfisher Optimizer-Radial Basis Function(PKO-RBF).Firstly,the data of influencing factors are corrected based on the Almont polynomial and coupling degree model;then,the set of significant influencing factors is initially determined by integrating the four correlation coefficients;finally,the corrected data of the significant influencing factors are taken as inputs,and the RBF hyper-parameters are optimised by the PKO for RBF neural network prediction,and the analysis of the cases shows that the proposed prediction model has high prediction accuracy,which offers a new approach for the traditional electricity forecasting method.electricity prediction method provides a new idea.
Keywords：time lag effectcoupling effectAlmont polynomial modelcorrelation analysiskey influencing factor screeningRBF neural network
Publication Date：2025-08-28
Online Publishing Date：2025-09-29(First online date of this platform, not the publication date of the document)
Pages：7( 169-175 )