论文部分内容阅读
针对变压器油中溶解气体浓度预测中存在的变量取值范围影响预测精度问题,提出了基于核目标度量规则(Kernel Target Alignment,KTA)和支持向量机(Support Vector Machines,SVM)的油中溶解气体浓度预测方法。在分析油中溶解气体产生机理的基础上选取输入变量,采用KTA对输入变量进行尺度缩放来避免变量的取值范围影响SVM泛化性能问题,利用交叉验证法选择SVM的参数,建立油中溶解气体浓度的KTA-SVM预测模型。将所提出的方法与SVM和灰色模型进行比较,均方根误差分别为0.156 8、0.179 1、0.220 5,实验结果表明了所提出的方法具有较优的预测精度和泛化性能。
Aiming at the problem that the range of variables in the prediction of the concentration of dissolved gas in transformer oil affects prediction accuracy, the dissolved gas in oil based on KTA (Target Kernel Target Alignment) and Support Vector Machines (SVM) Concentration prediction method. Based on the analysis of dissolved gas generation mechanism in oil, input variables are selected and KTA is used to scale the input variables to avoid the range of variables from affecting SVM generalization performance. The parameters of SVM are selected by cross-validation to establish the oil dissolution KTA-SVM prediction model of gas concentration. The proposed method is compared with SVM and gray model, the root mean square error is 0.156 8,0.179 1,0.220 5, the experimental results show that the proposed method has better prediction accuracy and generalization performance.