Faults diagnosis for power transformer based on support vector machine

The power transformer is very important equipment in a power system, and it is necessary to carry through faults diagnosis for it. Support vector machine is a machine learning algorithm based on statistical learning theory, which can get good classification effects with a few learning samples. A new power transformer fault diagnosis method based on support vector machine is presented in this paper. The method has many advantages for transformer faults diagnosis, such as simple algorithm, good classification and high efficiency. This faults diagnosis method finally has been proved by many practical faults data of power transformer. Compared experiment results with the traditional three-ratio method, this method has higher diagnosis right ratio. So it shows that such method is very feasible and is very suitable for power transformer faults diagnosis.