Research On Fault Detection Method Of Rotor Bar Breaking Of Asynchronous Motor Based On Duffing System

Asynchronous motor is widely used in the production of national industry.Motor failure is not only a loss of a motor,but also a serious breakdown of the production line due to the damage of motor,resulting in huge economic losses and catastrophic consequences.Therefore,it is of great significance to realize the fault detection of the rotor broken bar of the asynchronous motor.A detection method combining Duffing system with extended Prony algorithm is proposed in view of the fault of rotor bar that often occurs in asynchronous motor.First of all,we studied the sensitivity of Duffing system to initial parameters,and compiled the corresponding threshold comparison table according to the principle of detecting fault feature signals in Duffing system.Then,aiming at the intermittent chaos of the Duffing system,the corresponding improvement measures for the sensitivity of the Duffing system and the initial phase angle are proposed.Finally,because the Duffing system at the same time to ensure the frequency resolution of the loss of the ability to effectively estimate the fault feature component amplitude,and amplitude of fault feature component is important for judging the extent of rotor fault,this paper based on the extraction of fault feature components by using extended Prony algorithm to determine the amplitude of the fault component and initial phase angle at Duffing system.Aiming at the feasibility of Duffing-Prony algorithm,Matlab simulation and experimental verification were carried out respectively.Rotor fault detection experiment showed that the rotor bar broken fault detection method based on Duffing and Prony has good detection effect.Unlike those traditional detection methods abiding in frequency domain,a time-domain method to detect broken rotor bar fault in asynchronous motor is presented in this paper,aiming to extend the thinking dimension of this kind of detection.