An Improved Adaptive Selected Harmonic Elimination Algorithm for Current Measurement Error Correction of PMSMs

Measurement errors are inevitable in the current sensors, which cause speed ripple including one and twice the stator electrical frequency. For the undesired harmonics, the adaptive selected harmonic elimination (ASHE) algorithm outputs the sum of the sinusoidal signals multiplied by the adaptive weights. In order to solve the current measurement error (CME) issue, this article adopts the ASHE algorithm. According to the deterministic functional relationship of the surface-mounted permanent magnet synchronous motor (SPMSM), the adopted algorithm extracts the harmonics of the steady state speed error, and outputs the q-axis current compensation. However, there is no explicit connection between speed and d-axis current, so their relationship is uncertain. The remaining d-axis CMEs result in poor current performance. Therefore, this article proposes an improved ASHE algorithm, which compensates the dq-axis CMEs simultaneously depending on their mutual deterministic connection. The proposed method does not need motor parameters, additional sensors, or complex calculation process. Finally, the effectiveness of the method is verified by the SPMSM platform. Experimental results also show that the proposed method reduces speed ripple and suppresses the negative effects of CMEs.