Efficient fault diagnosis in smart grid using non conventional mother wavelet function

In today's smart grid environment existing fault diagnosis methodology emphasizes the need of more advanced, accurate and sturdy strategies. New modified methodology using latest sensing, communication and controlling devices is essential in order to make system self healing and resilient. The paper presents use of Symlet mother wavelet function for fault diagnosis method using multi resolution analysis and feature extraction technique of wavelet transform. Fault current signals can be sensed using advanced smart sensors which will be made available in smart grid system. These are used to extract energy feature of fault transients to locate and classify faults. The method is implemented on IEEE standard 15 bus system with distributed generator. MATLAB 2013a is used to simulate the test system and coding is also developed to extract features for fault detection, classification and location. The method is found accurate, reliable and sturdy as compared to other conventional methods and even methods of feature extraction using conventional mother wavelet functions. Technique can be further developed to be used for real time fault diagnosis which can help to improve smart grid's distribution systems reliability, security and quality.