应用最小二乘支持向量机在线预测绝缘子等值附盐密度

On-line Forecasting Insulator’s ESDD with Least Squares Support Vector Machine

污秽等级评定方法是绝缘子泄漏电流在线监测系统的重要研究内容,等值附盐密度是确定污秽等级的唯一依据,而泄漏电流与绝缘子表面污秽状况密切相关。笔者采用最小二乘支持向量机(LS-SVM)方法,建立以泄漏电流有效值、泄漏电流峰值、脉冲电流次数、环境湿度、温度5个变量作为输入参数,ESDD作为输出参数的智能预测模型。实验结果表明,该方法有效、模型预测精度高,能实现绝缘子表面污秽程度在线评估。

The method of contamination grades assessment is very important content for the on-line leakage current （LC）monitoring system of insulator, however the equivalent salt deposit density（ ESDD ）is the only basis of defining pollution classes, and leakage current is strongly correlated to the contamination condition of insulator surface. Based on experimental data, it is investigated to build a intelligent model to predict ESDD with the least squares support vector machine（LS-SVM）in this paper. The r.m.s, value of the LC, the peak value of the LC, the amplitude and times of the pulses of the LC, environmental humidity and temperature are chosen as five input variables, the value of ESDD is chosen as one output variable. Experimental results show that the method is effective, the model has high forecast accuracy, and on-line evaluation method of insulator surface contamination can be realized.