基于SPIN算法的异步电机转子断条故障信号分离

Application of in Fault Signal Separation of Broken Rotor Bar in Asynchronous Motor Based on the SPIN Algorithm

目前采用通信技术对异步电动机状态进行监测及故障诊断过程中,常需要加大信息采集量来确保诊断结果的可靠性,造成数据传输占用带宽变大、数据存储占用空间增多的问题.由于考虑到在故障诊断系统中正常信号并不具有故障分析的意义,若将故障信号进行分离提取后作为系统的输入,则数据传输和处理过程可节省出更多时间与空间.因此,提出将SPIN算法应用到异步电机转子断条故障诊断中提取故障信号的过程中,运用MCSA方法将异步电机定子电流信号建模为故障分量与正常分量信号的总和,对原始信号进行梯度投影将分量信号分别恢复出来,进而达到分离故障特征信号的目的.

In the process of using communication technology to monitor and diagnose the state of asynchronous motor,it is often necessary to increase information collection to ensure the reliability of the diagnosis results,which causes the larger bandwidth of the data transmission and the increase of the occupancy space of the data storage.Considering that the normal signal in the fault diagnosis system is not useful for fault analysis,if the fault signal is separated and extracted as the input of the system,the data transmission and processing help to save more time and space.Therefore,it was proposed in this paper to use the SPIN algorithm to extract the fault signal in the fault diagnosis of broken rotor bars in asynchronous motor and use the MCSA methods to model the stator current signal of the asynchronous motor into the sum of the fault component and the normal component signal.The original signal was projected by gradient projection to recover the component signals,so as to achieve the purpose of separating the fault characteristic signals.