Research And Application Of Equipment Health Status Assessment Method For Electric Power Dispatching Automation Equipment

With the rapid development of intelligent automation technology,many industries have benefited.The continuous application of intelligent automation technology in power dispatching systems has also promoted the continuous upgrading of power dispatching systems.The power dispatching system has gradually changed from traditional manual monitoring and manual control to an automated dispatching system that integrates automatic monitoring and automation control.The operation of the power dispatching automation system is inseparable from a large number of information equipment,so it is crucial to ensure the stable and normal operation of these information equipment.In order to reduce the possibility of unexpected equipment failure,the key is to preventive maintenance of the equipment,and the core of preventive maintenance is equipment status assessment.Therefore,the focus of this paper is to evaluate the status of power dispatch automation equipment.Firstly,in order to obtain the high-quality sample data directly available,this paper preprocesses the collected raw data,which reduces the interference of noise data on model training.Then,based on the problem of the proportion of various types of samples and their imbalance,this paper uses the improved SMOTE oversampling technique to oversample the datasets after data preprocessing,and adjust the original unbalanced datasets to equalized datasets.The interference of the unbalanced sample set on model training.Then,in order to solve the nonlinear and fuzzy problems in the state evaluation of power dispatching automation equipment,this paper uses a fuzzy neural network combined with neural network and fuzzy theory to analyze the complex nonlinear relationship between various influencing factors and equipment states.It is also possible to accurately describe the state of the device between the two states.Then,in order to further improve the recognition accuracy of the model and the convergence speed of the model,the particle swarm optimization method is used to optimize the initial values of the weight parameters of the fuzzy neural network and the initialvalues of the membership function parameters,and then based on the searched optimal initial parameters.Training of neural network models.At the same time,the decreasing strategy of the inertia factor of the particle swarm is improved,which further improves the global search ability and the late local search ability of the particle swarm optimization algorithm.The experimental results verify the feasibility and effectiveness of fuzzy neural network in the health assessment of power dispatching automation equipment.At the same time,the experimental results show that the improved particle swarm optimization algorithm further improves the accuracy of the fuzzy neural network and the convergence speed of the model.