Electric Equipment Fault Diagnosis Based On Bayesian Network In Power Grid

With the continuous expansion of the grid scale,how to ensure the safe operation of the power grid has become a very important issue.For this problem,the intelligent method can be applied for the power grid equipment fault diagnosis and forewarning analysis,which ensure safe and effective operation of power systems and attach an important significance to power equipment.So how to identify the electric equipment failures is currently the most major problem for the fault diagnosis system.Based on the characteristics of network equipment failures,by using fuzzy set theory and processing class imbalance,and with Bayesian information criterion scores-climbing search algorithm for Bayesian networks,the essay constructs a network equipment fault diagnosis model.And then do an effective verification by using T City data.The main content can be summarized as follows:(1)For the class imbalance issue of electric equipment failure data,the under-sampling and the synthetic minority oversampling techniques are used to balance the data categories.And with using minority as the standard training data to multiply,compare the accurate rate of each category to find the optimal model.(2)The theory of fuzzy sets are used to solve the power equipment failure data in border to issues and retain data information by constructing appropriate membership functions,and the minimum description length-minimum evaluation standard of information entropy are used for data discretization to enhance the accuracy of the model.(3)After data preprocessing and class balanced,based on the way of Bayesian scores search algorithm and maximum likelihood estimation parameters method,design and imply a state for power equipment fault diagnosis Bayesian network model.Use real data of T city to do an experiment to verify the status fault diagnosis model,make an effective solution to the fault classification problems and improve the power grid fault diagnosis capabilities.(4)By a combination of different equipment failure state,put forward a different maintenance guidance strategy which avoids equipment damage and expansion of failure scope and emphasis due to unplanned downtime.