基于GRA-IGA-BP的输电线路负荷预测  

Transmission Line Load Forecasting Based on GRA-IGA-BP

作　　者：陈静[1] 蔡金 Chen Jing;Cai Jin(School of Electrical and Information Engineering,Anhui University of Science and Technology,Huainan,Anhui 232001,China)

机构地区：[1]安徽理工大学电气与信息工程学院,安徽淮南232001

出　　处：《黑龙江工业学院学报（综合版）》2023年第6期103-108,共6页Journal of Heilongjiang University of Technology(Comprehensive Edition)

基　　金：国家自然科学基金项目(项目编号:51874010);安徽省教育厅高校自然科学研究项目(项目编号:KJ2018A0087)。

摘　　要：针对输电线路负荷预测模型中参数选取困难以及在确定影响因素时主观性较强导致预测精度低的问题,提出一种考虑灰色关联权重分析与IGA-BP模型相结合的输电线路负荷预测模型。该预测模型首先采用灰色关联分析对影响输电线路负荷的因素进行相关性分析,提取权重较大的特征,从而降低模型的复杂度;其次,将Tent映射加入到遗传算法的初始种群中,用于生成分布均匀的混沌序列;最后,基于改进的遗传算法寻优BP模型的权值和阈值,得到最优的IGA-BP输电线路负荷预测模型。基于实际输电线路负荷数据,进行IGA-BP预测模型与传统BP、GA-BP预测模型对比实验。实验结果表明,基于IGA-BP的输电线路负荷预测精度高达91.789%,远高于BP和GA-BP算法,对输电线路负荷预测能力明显提升。This paper proposes a transmission line load forecasting algorithm that combines grey correlation weight analysis with IGA-BP model to address the difficulties in parameter selection in transmission line load forecasting models and the low prediction accuracy caused by strong subjectivity in determining influencing factors.The prediction model firstly uses grey correlation analysis to perform correlation analysis on the factors that affect the load of the transmission line,extracting features with larger weights,thereby reducing the complexity of the model;Secondly,to improve the iteration speed and prediction accuracy,an improved genetic algorithm is used to optimize the weights and thresholds in the BP model to obtain the optimal IGA-BP prediction model;Finally,the optimal IGA-BP transmission line load prediction model is obtained based on the weights and thresholds of the improved genetic algorithm search optimization BP model.Based on the actual transmission line load data,the IGA-BP prediction model is compared with the traditional BP and GA-BP prediction models for experiments.The experimental results show that the accuracy of transmission line load forecasting based on IGA-BP is as high as 91.789%,much higher than BP and GA-BP algorithms,and the ability to predict transmission line load is significantly improved.

关 键 词：输电线路 负荷预测 遗传算法 BP神经网络 

分 类 号：TM714[电气工程—电力系统及自动化]