In order to improve the classification precision of Support Vector Machines(SVM), to solve the difficult problem of its parameter selection, a parameter optimization method based on improvedartificialbeecolony algo...
详细信息
In order to improve the classification precision of Support Vector Machines(SVM), to solve the difficult problem of its parameter selection, a parameter optimization method based on improved artificial bee colony algorithm is proposed to solve this problem. Then the proposed method is applied to the diagnosis of four types of transformer faults: low-temperature overheating, high-temperature overheating, high-energy discharge and low-energy discharge. Using hydrogen, methane, ethane, ethylene, and acetylene five gases as the feature vectors, and the inverse of classification error rate is used as fitness value, and the improved artificial bee colony algorithm is used to optimize the penalty factor and kernel parameter of SVM. Then the optimized SVM is used to classify the transformer fault type classification, and the recognition accuracy rate is 91.2281%. The experimental results show that this method had high accuracy and fast convergence. It has a more universal value and is suitable for power transformer fault diagnosis.
暂无评论