A novel filtering algorithm for the resolver position signal based on recursive least squares and phase-locked loop

This paper presents a novel filtering algorithm for the resolver position signal based on recursive least squares (RLS) and phase-locked loop (PLL). The proposed RLS-PLL-based algorithm designs a harmonic iterative estimation and elimination mechanism, which can accurately estimate and eliminate the specified harmonic contained in the resolver online and obtain an ideal reconstructed position signal. The convergence analysis, parameter design and pseudo code of the proposed algorithm are all provided. The advantage of the presented method mainly lies in that it has excellent position signal filtering performance in the entire operating condition, including low speed, high speed, steady state, and dynamic. Experimental results demonstrate that compared to the results achieved with the conventional look-up table (LUT) method and the advanced type-III angle tracking observe (ATO), the proposed RLS-PLL-based algorithm has improved filtering performance. Under some typical operating conditions, the harmonic content of the position signal reconstructed by the new method is reduced by an order of magnitude.