基于改进鲸鱼算法优化的接地网腐蚀速率预测  

Corrosion Rate Prediction of Grounding Grid Based on Improved Whale Algorithm Optimization

作　　者：张海博 潘鹏程 郑峰 ZHANG Haibo;PAN Pengcheng;ZHENG Feng(College of Electrical Engineering and New Energy,China Three Gorges University,Yichang 443002,China;Nanyang Power Supply Company,State Grid Henan Electrical Power Company,Nanyang 473000,China)

机构地区：[1]三峡大学电气与新能源学院,湖北宜昌443002 [2]国网河南省电力公司南阳供电公司,河南南阳473000

出　　处：《电子科技》2025年第5期8-14,30,共8页Electronic Science and Technology

基　　金：国家水运安全工程技术研究中心开放基金(B2022002);宜昌市自然科学基金(A22-3-008)。

摘　　要：为提高变电站接地网腐蚀速率预测的精确度,解决传统BP(Back Propagation)神经网络易陷入局部最优解且传统优化算法随机初始化种群影响预测精度等问题,文中提出了一种改进鲸鱼算法优化BP神经网络方法来预测变电站接地网腐蚀速率。通过混沌映射思想初始化鲸鱼种群、改进非线性因子调整自适应权重、强化搜索的莱维飞行3种策略得到改进后的鲸鱼优化算法,建立基于改进鲸鱼算法优化BP神经网络的接地网腐蚀预测模型,以72座变电站的接地网腐蚀数据进行仿真分析。结果表明,改进后的模型平均相对误差为1.84%,全局最大相对误差为3.86%,均方根误差为0.13902,其误差相比传统模型显著降低,证明了所提预测模型的可行性。In order to improve the accuracy of predicting the corrosion rate of substation grounding grids and solve the problems of traditional BP(Back Propagation)neural networks easily falling into local optima and traditional algorithms randomly initializing populations affecting prediction accuracy,this study proposes an improved whale optimization algorithm to optimize the BP neural network for predicting the corrosion rate of substation grounding grids.An improved whale optimization algorithm is developed using chaotic mapping to initialize whale population,improving nonlinear factor to adjust adaptive weights,and enhancing search by Levay flight.Corrosion prediction model of grounding grid based on the improved whale optimization algorithm to optimize the BP neural network is established.The corrosion data of grounding network of 72 substations are simulated and analyzed.The results show that the average relative error of the improved model is 1.84%,the global maximum relative error is 3.86%,and the root-mean-square error is 0.13902,which is significantly lower than that of the traditional model,proving the feasibility of the proposed model.

关 键 词：变电站 接地网 腐蚀速率 改进鲸鱼算法 BP神经网络 预测模型 混沌映射 非线性因子 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]