Fault Tree And Neural Network-based Power System Fault Diagnosis System Design And Implementation

With the rapid development of power systems, power equipment, the functions become more perfect and the automation of power systems show increasingly high degree of intelligence, the coupling between the various power subsystems growing, but the structure more complex structure, Once the equipment failure during operation, causing power failure and enormous economic with social loss, for the production of life and even catastrophic consequences. The long period of outdoor power equipment running state, operations officer for its supervision is relatively weak, a large number of monitoring data into the dispatch center, to the operating personnel identify the fault in the complex circumstances cause difficulties. It is necessary to strengthen the theory of power system fault diagnosis Research and practical application techniques to achieve automatic rapid diagnosis of network faults.Based on existing research paper, applied fault tree reasoning and neural networks, that is artificial intelligence. Accompany with the information processing techniques to identify the fault spot in power system to provide optimal condition-based maintenance with decision support, which includes the contents of the following:1. Discussion of a common electrical failure and its type. Summarizes the practical common engineering fault identification and fault modes.2. Based on those introduced classical power system fault diagnosis algorithm, including the classic knowledge-based qualitative fault tree analysis algorithms, neural network analysis and PERTI networks, Bayesian networks.3. Introduce the mature SCADA monitoring technology and IEEE COMTRADE standard .And to explore the fault diagnosis system for auxiliary support.4. Using Microsoft.NET to implement the diagnosis system and communicate with the SCADA system interconnection after system failure.