基于组合决策树的油浸式电力变压器故障诊断

提出了一种利用属于模式识别范畴的决策树C4.5法进行变压器故障诊断的方法。由于C4．5方法可方便地处理连续特征模式且有从样本学习判定规则的功能，因此应用中显示了该方法对于变压器故障诊断的适用性。在讨论变压器故障空间的基础上，针对已积累的故障变压器的大量油中溶解气体等数据，考察了各类故障的特征偏置，并在此基础上构造出组合决策树诊断模型，实现了变压器故障由粗到细的逐级划分，有利于提高诊断的准确性。实例表明该模型的有效性。

FAULT DIAGNOSIS OF OIL-IMMERSED POWER TRANSFORMER USING COMBINATORIAL DECISION TREE

Based on C4.5 method which belongs to pattern recognition, a power transformer fault diagnosis model is proposed. As the C4,5 is capable of handling the continuous numerical value feature patterns and has the function that can learn decision rules from samples, the application of the proposed method to some realistic samples show that the C4.5 method fit for transformer fault diagnosis perfectly. Based on the discussion of fault classification methods and a bias analysis of dissolved gas data of thirteen usual transformer faults, a decision tree is introduced to realize the multi-resolution recognition of the insulation faults, which not only can make the fault diagnosis be more exact, but also is helpful to establish a significant strategy for the maintenance, The recognition results show that this model is effective.