Dynamic Adaptive Correction Strategy for Position Sensorless BLDCM Commutation Error Based on Sign Detection

Correcting commutation error plays a critical role in enhancing the position sensorless brushless DC motor (BLDCM) efficiency. The existing commutation error sign detection (CESD)-based correction method gets rid of the reliance on analog-to-digital converter, reducing the demand on BLDCM system configuration. However, due to the complex structure and excessive power consumption of the existing CESD circuit, it is not conducive to long-term stable system operation. Besides, conventional algorithm with decreasing iterative increment exhibits poor anti-interference. Therefore, it is necessary to further improve the conventional correction method. First, a simplified CESD circuit is proposed by equivalently replacing components in the conventional design, increasing detection reliability and decreasing power consumption. Moreover, the compensation angle is calculated using a dynamically varying correction algorithm based on the commutation error sign. By dynamically adjusting the gain factor according to the correction condition, the proposed algorithm switches between searching and locating modes to improve adaptability to disturbances. Under this correction strategy, the compensation angle inevitably converges to the required value within a tiny accuracy range, which is theoretically proven as well. Finally, experiments on a magnetically suspended control moment gyro prototype validate the effectiveness of the proposed method.