The Research On Broken Rotor Bar Fault Detection In Induction Motors Based On HFRSET And Optimization Algorithm

Induction motors are widely used in many fields of the national produ ction, such as industrial and agricultural, because of its high performance-cost ratio, good reliability and convenient maintenance. A motor failure can not only result in damage to the motor, but also lead to heavy financial losses and catastrophic failure when it is serious. Statistical studies have shown that the broken rotor bar fault is one of the most familiar faults for induction motors. Consequently, there are great theoretical significance and socioeconomic benefits to do research on broken r otor bar detection for induction motor. Because of the occurrences of broken rotor bar fault mostly need to experience a process of gradual development, obviously, the fast and reliable detection of broken rotor bar fault in induction motors has been of profound importance when the incipient failure occurs. That is the main work of this paper.In order to realize the rotor broken bar quickly and reliably detect incipient faults, the combination of high frequency resolution spectrum estimation technique and optimization algorithms is applied to induction motor broken rotor bar fault detection in this article. Firstly, combining singular value decomposition and signal parameters via rotational invariance technique to extract the broken rotor bar fault feature frequency component in stator current signal, and the simulation result demonstrate that it is effective. But it can't estimate the amplitude and initial phase of each component accurately. However, it is crucial to judge the degree of broken rotor bar that the size of fault feature component's amplitude. Therefore, the paper applied optimization algorithm to complete the relevant work. Then, it introduces the principle of two kinds of improved algorithm of particle swarm algorithm and gravitational search algorithm respectively. It is ideal that the new methods' feasibility verification with the simulation signal of broken rotor bar fault. And the new methods combine the three kinds of optimization algorithm with the method mentioned above. Thus new methods of induction motor broken rotor bar fault detection based on ESPRIT of singular value decomposition filtering and YPSO/SAPSO/GSA are formed. Finally, The experimental result of an induction motor with the broken rotor bar fault demonstrate that new meth ods of induction motor broken rotor bar fault detection based on ESPRIT of singular value decomposition filtering and YPSO/SAPSO/GSA are effective. D ue to the fact that it could quickly work out the accurate results under guaranteeing the premise of high frequency resolution even for short-time sampling signal, all of them are also suitable for online detection for induction motors operating under severe interference, such as fluctuant load and noise. Therefore, the new methods have wide engineering application prospect.