Bandpass Effect and Its Compensation Method for Diagnosis in a Closed-Loop Control System

Motor current signal analysis is an effective technique for fault diagnosis in the motor drive system. However, in the closed-loop control (CLC) system, the amplitude of periodic faulty harmonic is suppressed by the impact of CLC characteristics, so the severity assessment of faults may fail. In this article, a model reference method based on the theoretical analysis of CLC characteristics is proposed to deal with this problem. First, the faulty harmonic in the q-axis current is extracted from the field-oriented control process as a fault indicator. Then, the bandpass effect is revealed to describe the variation of amplitudes of harmonics in the current spectrum. The variation is caused by the control parameter and architecture in the CLC system. Finally, to improve diagnostic performance, a model reference method is proposed to compensate for the aforementioned variation. The model is established by experimentally measuring the amplitude-frequency response of the closed-loop speed control system. The bandpass effect of the CLC system is validated by simulations in the MATLAB/Simulink environment. Furthermore, experimental validations are conducted on a ball screw platform. The results show that the proposed method is effective to compensate the bandpass effect and helps to achieve better diagnostic performance.