Study On Security Risk Analysis And Assessment Of Transformer Based On Neural Network

In many factors of affecting operation security in electricity power system, the working safely of the equipments plays an important role. In order to strengthen the safety and reliability of power plant, potential risk factors and failure mechanism on thermal power plant tansformer equipment are investigated in this thesis.Firstly, this thesis summarizes and evaluates the existed security riske assessment methods. Furthermore, fault events are modeled and made into analysis based on the risk analysis theory of fault tree. After that equipment failure characters are confirmed and key factors are exacted.Then intelligent knowledge representation and risk evaluation methods based on neural networks are proposed, which is according to the complexity of risk analysis of the characteristics of nonlinear in the evaluation system. The testing results of evaluation and analysis, validate that the proposed model is reasonable and effective. At last through the secureity risk assessment of transformer by neural networks propose some preventive measures.