基于支持向量机的雾中棒–板间隙击穿电压预测

Breakdown Voltage Prediction of Rod-plane Gap in Fog Based on Support Vector Machine

空气间隙的击穿电压是电场分布和大气环境综合作用的结果。采用支持向量机（support vector machine,SVM）建立了空气间隙在不同大气环境中的击穿电压预测模型,以电场特征量和大气参数作为预测模型的输入,基于二分类的思想,将空气间隙在加载电压耐受或击穿分别表示为-1和1,并作为预测模型的输出。基于正交试验设计选取训练样本集,采用SVM模型对棒–板间隙在雾环境中的工频击穿电压进行了预测,预测结果与试验值符合良好,51个测试样本的平均绝对百分比误差仅为1.04%。根据预测结果,采用多元非线性回归方法拟合得到了击穿电压与雾环境中大气参数的关系式,分析所得影响规律与根据试验数据所得结论基本一致,验证了SVM模型预测结果的合理性。研究结果可为复杂大气环境下的空气间隙击穿电压预测及放电特性研究提供参考。

The breakdown voltage of an air gap is the result of the synthetic effect of electric field distribution and atmospheric environment. We established a model for air gap breakdown voltage prediction under different atmospheric environments by support vector machine（SVM）. The electric field features and atmospheric parameters were taken as the inputs of the prediction model. Based on the idea of binary classification, the withstand or breakdown of an air gap under the applied voltage was characterized by-1 and 1, which was taken as the output of the prediction model. The training sample set was selected based on orthogonal experimental design, and the SVM model was applied to predict the power frequency breakdown voltages of rod-plane gaps in fog environment. The predicted results are in good agreement with the experimental values, and the mean absolute percentage error of 51 test samples is only 1.04%. The relational expression between the breakdown voltage and the atmospheric parameters in fog environment were fitted by a multivariate nonlinear regression method according to the predicted results. The influence laws basically coincide with those obtained by experimental data, which validates the rationality of the predicted results of SVM model. The research results can provide references for breakdown voltage prediction and discharge characteristic study of air gaps in complex atmospheric environment.