为探讨污闪电压的有效预测方法,采用支持向量机回归方法,建立盐密、灰密与污闪电压的关系模型,对污闪电压进行预测。实例分析表明,其预测结果与实测结果的误差很小,表明支持向量机是一种非常有前景的预测工具。
To discuss the effective prediction methods of pollution flashover voltage, the model of correlation between salt and non-soluble deposit densities and pollution flashover voltage was established based on support vector machine (SVM) regression to predict the pollution flashover voltage. The practical analysis shows that the errors between predicted results and actual measurement results are small. Thus, SVM appears to be a very promising prediction tool.