Three-phase Asynchronous Motor Fault Diagnosis Based On Neural Fuzzy Petri Nets

As the most widely used power-dragging equipment,three-phase asynchronous motors play an important role in the national production and life under the background of rapid economic development in our country,especially in the fields of industry and agriculture.In addition to jeopardizing the motor itself,the failure of the motor may also lead to the collapse of the entire power system,seriously affecting the life and property of the people.Because of the special status of the three-phase asynchronous motor,how to ensure its stable operation is of great significance.Therefore,it is very important to study an effective fault diagnosis method for three-phase asynchronous motors.For the fault diagnosis of three-phase asynchronous motor,the author does the following research:1.Based on the traditional Petri net theory,fuzzy theory and neural network algorithm,the fault prediction and diagnosis method is analyzed on the basis of adaptive fuzzy fault Petri net.The paper also describes the composition,structure and modeling rules of NFPN in detail,highlighting the advantages of the research methods and ideas.At the same time,the main research directions and contents of this article are elaborated.2.On the basis of previous research results,the paper abandons the more serious empirical transition confidence,using Sigmoid to replace it and better adapt to neural network adaptive adjustment to make the data more clear.Through forward reasoning,the three-phase asynchronous motor fault information propagation status is demonstrated and the fault propagation path is accurately reflected.According to the failures of the three-phase asynchronous motor,the cause of the fault can be quickly found to avoid the blindness and experientiality of the diagnosis through the reverse reasoning matrix.3.Based on the analysis of three-phase asynchronous motor working principle,fault type and structure,the NFPN model of three-phase asynchronous motor is established.The data is statistically processed by using the FMEA method and Bayesian probability method.4.The example of NFPN fault sub-models is used for forward and reverse reasoning demonstrations to realize the fault prediction and diagnosis of the three-phase asynchronous motor.Finally,using Labview programming software,the three-phase asynchronous motor fault prediction and diagnosis system based on NFPN is designed,which can help the staff to investigate and maintain quickly and effectively.