Optimal SVM transformer fault diagnosis method based on empire colonial competition algorithm

Abstract

The invention relates to the technical field of the electrical equipment fault diagnosis, and specifically relates to an optimal SVM (support vector machine) transformer fault diagnosis method based on an empire colonial competition algorithm. The method comprises the following steps: analyzing a ratio of dissolved gas (DGA) characteristic quantity in selected transformer oil, performing normalization preprocessing to obtain DGC three-ratio characteristic quantity data, thereby realizing the effectiveness and conciseness of the three-ratio characteristic quantity. The nonlinear and multi-classification changes are performed on the SVM in one to one by adopting OAO, a k-fold average classification accuracy rate target function is constructed, and the fault diagnosis is performed on the transformer by combining a cross validation principle, and parameter optimization of a kernel function is performed by adopting the empire colonial competition algorithm so as to improve the accuracy rateof the fault diagnosis; the binary-classification SVM is expanded as multi-classification SVM, various fault types of the transformer can be diagnosed and identified at one time; the operation is simple and convenient, and a new method for judging a running state of the transformer, assessing a fault condition of the transformer, and guaranteeing the safety and stable running of the transformer is provided.

Description

technical field [0001] The invention relates to the technical field of electrical equipment fault diagnosis, in particular to an optimized SVM transformer fault diagnosis method based on an empire-colonial competition algorithm. Background technique [0002] With the rapid development of computer, communication and sensing technologies, information systems and power systems have gradually realized deep integration and cooperation, and power systems have become one of the important fields for the application of cyber physical fusion system CPS (Cyber ​​Physical System). In the power system, the power transformer is one of the most important equipment. Its operating status is related to the safe and stable operation of the entire power system. Once it fails, it will cause local or even large-scale power outages, which will inevitably cause huge economic losses. . Therefore, it is of great significance to accurately and effectively diagnose the faults of power transformers to ...

AI Technical Summary

Problems solved by technology

For example, the kernel function of the support vector machine diagnosis method has problems such as difficulty in determining the regularization parameters, and the accuracy of fault diagnosis is not high enough.

Images