基于ACA-SVM融合的变压器故障诊断

Faults diagnosis of transformer based on fusion of ACA-SVM

变压器故障诊断是一项复杂而有难度的技术,受到很多因素的影响.本文提出了基于ACA-SVM融合的变压器故障诊断方法,即将蚁群优化算法(ACA)用于SVM参数优化.它不仅具有很强的全局搜索能力,而且容易实现.经实验结果证明,基于ACA-SVM融合的变压器故障诊断结果与实测数据基本一致,其预测精度高于普通的SVM和IEC三比值法,能有效诊断变压器内部潜伏性故障.

Affected by many factors,fault diagnosis of transformer is a complex and difficult technology.The proposed ACA-SVM fusion method is applied to fault diagnosis of transformer in the paper,in which the ant colony algorithm（ACA） is used to determine free parameters of support vector machine.The method not only has strong global search capability,but also is very easy to implement.The experimental results indicate that the ACA-SVM method can achieve the nearly same result as the measured data and higher diagnostic accuracy than normal SVM classifier and IEC three ratios.Consequently,it can diagnose the internal latent faults in oil-filled transformer effectively.