基于最小二乘支持向量机的绝缘子等值附盐密度预测

等值附盐密度是确定污秽等级和绘制电网污区分布图的主要依据，而绝缘子污秽在线监测系统主要监测绝缘子表面泄漏电流和环境参数。研究表明，泄漏电流除了和绝缘子表面的污秽状况有关外，还受温度、湿度等环境因素的影响，并且和各因素之间存在着复杂的非线性关系。文中在实验室模拟试验和现场实测数据基础上，利用最小二乘支持向量机，建立了以泄漏电流有效值、泄漏电流脉冲峰值、泄漏电流脉冲频度、环境湿度、温度等五个变量作为输入参数，等值附盐密度作为输出参数的智能预测模型。并通过部分实验数据验证了该方法的可行性。

Forecasting the ESDD of Insulator Based on Least Squares Support Vector Machine

The equal salt deposit density (ESDD) is the source of defining pollution classes and mapping pollution areas. The surface leakage current (LC) and environment factors are detected in the online monitoring system. Investigation shows that the LC is affected not only by the contamination of insulator surface but also by the environment factors including temperature, humidity and so on. The nonlinear relationship between the LC and the various factors is complicated. Based on laboratory simulation experiments and field data, the LC R.M.S., the pulse peak value of the LC, the amplitude and times of the pulses of the LC, temperature and humidity of environment are chosen as five input variables, the ESDD is chosen as one output variable, the intelligent prediction model using least squares support vector machine (LS-SVM)is built. The feasibility of the method is proved by tests in the laboratory and the field.