摘要
为了提高变电站接地网腐蚀速率预测结果的准确性,提出一种基于黑猩猩算法优化支持向量机的变电站接地网腐蚀速率预测方法。首先对变电站接地网腐蚀速率的特征量进行核主成分分析,确定土壤电阻率、Cl^(-)质量分数、含水量、氧化还原电位与腐蚀速率的关联性较大,选择上述四个特征量作为接地网腐蚀速率预测模型的输入量。然后采用黑猩猩算法对支持向量机进行参数寻优,建立变电站接地网腐蚀速率预测模型。最后采用腐蚀试验数据进行算例分析,并与其他方法的预测效果对比。结果表明,所提模型预测结果的平均相对误差为2.984%,均方根误差为0.00889 mm/a,比其他方法误差波动更小,预测精度更高,验证了所提变电站接地网腐蚀速率预测方法的实用性和优越性。
In order to improve the accuracy of predicting the corrosion rate of substation grounding grids,a method based on chimpanzee algorithm optimized support vector machine for predicting the corrosion rate of substation grounding grids is proposed.Firstly,a kernel principal component analysis is conducted on the characteristic variables of the corrosion rate of the substation grounding grid to determine the significant correlation between soil resistivity,mass fraction,moisture content,and redox potential with corrosion rate.The above four characteristic variables are selected as input for the grounding grid corrosion rate prediction model.Then,the chimpanzee algorithm is used to optimize the parameters of the chimpanzee algorithm,and a corrosion rate prediction model for the substation grounding grid is established.Finally,corrosion test data is used for numerical analysis,and the prediction results are compared with those of other methods.The results show that the average relative error of the proposed model′s prediction results is 2.984%,and the root mean square error is 0.00889 mm/a.Compared with other methods,the error fluctuation is smaller and the prediction accuracy is higher,which verifies the practicality and superiority of the proposed method for predicting the corrosion rate of substation grounding grids.
作者
李雨涵
刘燕燕
刘闯
刘海
徐达
LI Yuhan;LIU Yanyan;LIU Chuang;LIU Hai;XU Da(State Grid Hubei Electric Power Direct Current Company,Yichang 443002,China;State Grid Jingmen Power Supply Company,Jingmen 448000,China;China University of Geosciences,Wuhan 430074,China)
出处
《湖南电力》
2024年第2期77-83,共7页
Hunan Electric Power
基金
中国博士后基金面上项目(2021M692992)。
关键词
变电站接地网
腐蚀速率预测
核主成分分析
黑猩猩算法
支持向量机
substation grounding network
corrosion rate prediction
kernel principal component analysis
chimpanzee optimization algorithm
support vector machine
