Abstract
To tackle audible noise caused by sensorless control operation of the flux leakage controllable permanent magnet (FLC-PM) motor under typical operating conditions in the low-speed range, a multimodal low-noise sensorless control strategy is proposed in this article. Combining the structural characteristics of the FLC-PM and based on the driving conditions of electric vehicles in the low-speed range, three typical operating conditions are determined, and a multimodality design is carried out for these conditions. This strategy effectively solves the problem of deterioration in rotor position estimation accuracy that exists in traditional noise reduction methods. With the injection of the proposed variable-frequency voltage signal, significant noise reduction can be achieved. Correspondingly, a delayed effects-aware demodulated signal is designed to reduce the impact of system delay. Furthermore, an online variable-frequency current signal detection scheme is proposed to improve the signal detection quality, leading to high accuracy in rotor position detection. As a result, the steady-state and dynamic performances of the motor drive system can be enhanced. Finally, the sensorless control strategy is verified by experiments on the FLC-PM motor drive platform.
| Original language | English |
|---|---|
| Journal | IEEE Transactions on Industrial Electronics |
| DOIs | |
| Publication status | Accepted/In press - Oct 2025 |
Keywords
- Audible noise
- flux leakage controllable permanent magnet (FLC-PM) motor
- iterative learning control (ILC)
- online variable-frequency detection filter
- variable-frequency signal injection
ASJC Scopus subject areas
- Control and Systems Engineering
- Electrical and Electronic Engineering