An Improved PLL-Based Speed Estimation Method for Induction Motors through Harmonic Separation

Abstract

The real-time speed estimation of induction motors (IMs) is important for the motors’ state monitoring and control. The utilization of rotor slot harmonics (RSHs) due to the inherent cogging effect is regarded as a promising way to realize the speed estimation of IMs. The key to the RSH-based speed estimation method is how to accurately and quickly identify the frequency of an RSH signal. However, as the RSH signal always consists of two side-frequency components that are adjacent to each other, it is actually improper to directly use the conventional phase-locked loop (PLL) method designed for single-frequency tracking. Furthermore, the form of the two side components in the frequency domain also leads to a significant amplitude fluctuation in the time-domain waveform of RSHs, thus resulting in the obvious frequency tracking errors of the conventional PLL method. In this paper, we proposed an improved PLL through harmonic separation to further improve the performance of the RSH-based speed estimation method of multiphase IMs, so that the dynamic tracking errors of PLL due to the reasons mentioned above can be significantly reduced. Simulations and experiments in a wide speed range were also performed, with their results presented as verifications of the proposed method.