Fault Diagnosis Implementation of Induction Machines based on Advanced Digital Signal Processing Techniques

Abstract

In this paper, a comprehensive cross correlation-based fault diagnostic method is proposed for real time DSP implementation. It covers both fault monitoring and decision making stages. In practice, a motor driven by an adjustable speed drive is run at various operating points where the frequency, amplitude and phase of the fault signatures varies with time. These dynamic changes are considered as one of the common factor that yields erroneous fault tracking and unstable fault detection. In this paper, the proposed algorithms deals with the operating point dependent ambiguities and threshold issues. It is theoretically and experimentally verified that the motor fault can continuously be tracked when the operating point changes within a limited range.

A Compact Error Management Algorithm to Minimize False-Alarm Rate of Motor/Generator Faults in (Hybrid) Electric Vehicles

“…, N 2 . The signal averaging technique in [10] and [15] is modified and elaborately demonstrated for this purpose. The frequency analysis of the averaged signal in (8) can be limited from the Nyquist sampling theorem as follows:…”

“…However, in a typical traction system, perfect speed feedback may not be possible, and inaccurate speed feedback might potentially result in diagnosis failure. In [10], the MF-based detection is analytically derived and mathematically proven to have an optimal performance in fault signal detection under extremely noise condition. However, the MF-based method cannot maintain its optimal effectiveness and its performance reduce inevitably by speed feedback error.…”

“…As a commonly employed method, line current analysis has been the most popular technique in providing practical solutions for incipient fault detection [6]- [10]. Though the vibration and thermal analyses provide valid results [11]- [13], current-based monitoring provides low-cost protection in real time without using additional components.…”

“…As a result, by monitoring and analyzing the irregularities of fault harmonic modulations, the diagnostics have been efficiently performed to determine specific fault conditions. Among the various fault diagnostic methods proposed so far, some techniques keeping the track of the characteristic fault signatures in the spectrum have been suggested with reduced computational complexity and low memory occupancy on a microcontroller such as Park transform-based analysis (PTA) [14], [15], the phase sensitive detection (PSD) [9], the correlation-based detections (CD) [9], [15], and the matched filter (MF)-based techniques [10]. The MF-based theory has provided the fundamental background of those signature-based diagnosis techniques [9], [10], [14], [15] in terms of noise immunization and complexity minimization.…”

“…Various faults in a motor have been known to create specific signatures in the motor current spectrum owing to inherent electromechanical coupling. The stator current spectrum exhibits abnormal harmonic modulation under fault condition, where the observed frequencies are generalized as so-called fault characteristic frequencies in [5]- [10]. As a result, by monitoring and analyzing the irregularities of fault harmonic modulations, the diagnostics have been efficiently performed to determine specific fault conditions.…”