INTELLIGENT DISTURBANCE DIAGNOSIS IN POWER TRANSFORMERS BASED ON THE WAVELET TRANSFORM AND SUPPORT VECTOR MACHINE

Abstract

This paper proposes a method based support vector machine and wavelet transform in order to discriminate different disturbances in power transformers appropriately, such as internal and external faults, and transformer energizations. The proposed method recreates the conventional differential function using a disturbance detector, by means of the energies of the wavelet coefficients, which enables the support vector machine (SVM)-based differential phase functions. Furthermore, the proposed method can work in conjunction with other system protections by sending warning signals, for example in external fault conditions of the transformer, so it makes the protection system more reliable and intelligent. Several events were simulated in the alternative transient program, such as external and internal faults, turn-to-turn and turn-to-ground fault with variations of fault resistance, fault inception angle, and fault type parameters, as well as transformer energizations. The method presented better success rate and a faster trip in the internal fault detection then the conventional different protection.