![]() | Only 14 pages are availabe for public view |
Abstract Synchronous generator is one of the most important equipment in power systems, it can be used in different ratings, and connections. Therefore, the continuity of its operation is vital importance in maintaining the reliability of power system. Any unscheduled repair work, especially replacement of a faulty generator, is very expensive and time consuming. The differential protection provides the best overall protection for a synchronous generator. This thesis focuses on the application of artificial neural network (ANN) on protection of 3-phase synchronous generators. In the proposed scheme, different states of internal and external faults are considered as different patterns, which are recognized by an ANN based algorithm. The application of ANN makes the proposed scheme more secure against specific non-fault phenomena related to synchronous generator, and can provide a better detection of internal faults and external faults. Two types of ANN in this study are used, namely Multi layer feed forward neural network and recurrent neural network. The ANN application in synchronous generator protection has verified through using LABVIEWTM software and data acquisition card (DAQ). Experimental results confirm the feasibility of the techniques for detection and classification of faults in synchronous generator. |