基于Bagging改进算法变压器油中气体故障诊断研究

Research on Gas Fault Diagnosis in Transformer Oil Based on Improved Bagging Algorithms

针对变压器故障征兆与故障类型间映射关系的不确定性及模糊性问题,以变压器油中溶解气体数据作为变压器故障类型的判断依据,利用Bagging算法把弱分类器变为强分类器的特点,提出了Bagging的改进算法,并对该算法的性能进行了测试,测试结果表明该方法具有较好的分类精度。将Bagging改进算法应用到变压器油中气体故障诊断中,仿真实验结果表明,基于Bagging的改进算法优于boost集成算法及BP神经网络和支持向量机等最新方法。该方法精度达到90.67%。

In view of the uncertainty and fuzziness of the mapping relationship between the fault symptoms and the fault types of transformers,the dissolved gas data in the transformer oil was took as the judgment basis of the fault types of transformers,and the characteristics of the bagging algorithm were used to change the weak classifier into the strong one,the improved algorithm of bagging was proposed,and the performance of the algorithm was tested.The test results show that the method is effective and has better classification accuracy.The improved algorithm of bagging is applied to gas fault diagnosis in transformer oil.The simulation results show that the improved algorithm based on bagging is better than the latest methods such as boost integration algorithm,BP neural network and support vector machine.The accuracy of this method is 90.67%.