基于灰色最小二乘支持向量机的变压器油溶解气体预测

变压器油中溶解气体的体积分数是进行变压器绝缘故障诊断的重要依据,对变压器油中溶解气体进行预测有助于及时预测变压器的故障。将灰色预测方法与支持向量机相结合,通过使用对原始数列进行一次累加生成的处理方法,以提取数列所具有的深层规律特征,建立了基于灰色最小二乘支持向量机的变压器油中溶解气体预测模型,并对最小二乘支持向量机参数的选取进行了优化,最终通过实例与BPNN、灰色模型预测结果相比较,验证了该模型的准确性和有效性。

Study on Transformer Oil Dissolved Gas Prediction Based on Gray Least Square Support Vector Machine

Volume fraction of oil dissolved gas in the transformer is an important basis for transformer insulation fault diagnosis.The prediction of the oil dissolved gas is helpful for timely prediction of failure.With advantages of both gray forecasting method and least square support vector machine,two methods are combined to build the model of gray least square support vector machine for transformer dissolved gas prediction and regular features of the DGA data are extracted by a treatment of an AGO to original number of columns.The parameters’ selection of support vector machine is optimized and compared with BPNN and the gray prediction method.The accuracy and validity of the proposed method are verified through examples.