Research On Fault Diagnosis Of Power Grid Based On Firefly Algorithm And Neural Network Method

With the rapid development of the power industry,the power grid is developing in the direction of greater capacity and higher intelligence.At present,the objective scale of the power grid has the risk of large-scale power outages due to failures.Due to the close correlation between electricity and social development and people's lives,once a power outage occurs due to a failure,it will bring great inconvenience to people's daily life and huge losses to social economy.Therefore,the stability of the power grid operation and the diagnosis work before the fault occurs are particularly important.When the power grid fails,uncertain factors such as mis-switching,refusal to operate,and loss of information distortion often make it difficult to obtain satisfactory diagnostic results for grid fault diagnosis based on switch information.Aiming at the problems of incomplete and inaccurate information about power grid protection and circuit breakers,this paper studies the problem of fault diagnosis of transmission lines based on switching values,and uses the advantages of self-learning and nonlinearity of neural networks to realize the function of power grid line detection.At the same time,in order to improve the efficiency and speed of fault diagnosis,a hybrid algorithm using firefly algorithm to optimize BP neural network is proposed,which is applied to power grid fault diagnosis.Combining the global optimization computing power of the firefly algorithm with the BP neural network,the convergence threshold of the BP neural network is optimized to make up for the shortcomings of the BP neural network.It is applied to the diagnosis of areas with frequent grid faults to verify thesuperiority of the combination of the two.In this paper,we work on the above two aspects,using MATLAB software to simulate the BP neural network,firefly and BP neural network hybrid methods,and compare the simulation results to analyze the effectiveness of the proposed method in power grid faults.