Adaptive PLL-Based Sensorless Control for Improved Dynamics of High-Speed PMSM

• Published in: IEEE transactions on power electronics Vol. 37; no. 9; pp. 10154 - 10165
• Main Authors: Novak, Zdenek, Novak, Martin
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.09.2022; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Acceleration; Adaptive control; Adaptive filters; Algorithms; Bandwidth; Bandwidths; Error correction; Field programmable gate arrays; field-programmable gate arrays (FPGAs); Harmonic analysis; High speed; Observers; Permanent magnet motors; permanent magnet synchronous machine (PMSM); Permanent magnets; Phase locked loops; phase-locked loops (PLL); Quadratures; Rotors; sensorless control; Synchronous motors
• ISSN: 0885-8993; 1941-0107
• DOI: 10.1109/TPEL.2022.3169708

This article presents an adaptive quadrature phase-locked loop (AQPLL) for improving the dynamic performance of sensorless high-speed permanent magnet synchronous motors (PMSMs). A conventional quadrature phase-locked loop (QPLL) operating with a fixed bandwidth tends to have a tradeoff between antidisturbance performance and phase delay. The AQPLL can start to operate with a low bandwidth, while the bandwidth is adapted during motor acceleration in real time. Thus, the position estimation error is quickly minimized, leading to correctly decoupled rotor currents and higher available torque. Key parameters of the adaptation mechanism are examined. Sensorless field-oriented control (FOC) with a full-order observer and an AQPLL has been implemented on a field-programmable gate array and verified using a 100 kr/min PMSM. When motor dynamics is unknown or the QPLL cannot be tuned properly, the AQPLL can auto-tune the bandwidth, facilitating up to 25% faster acceleration from 5 to 100 kr/min. The steady-state position estimation error is less than 3<inline-formula><tex-math notation="LaTeX">{}^{\circ }</tex-math></inline-formula> at 100 kr/min. Experimental results confirm the effectiveness of the method for achieving a high-performance sensorless FOC of high-speed PMSMs. Furthermore, the AQPLL algorithm has the potential to replace the ordinary QPLL, helping to avoid time-consuming online phase-locked loop tuning in standard drives. This article is accompanied by a MATLAB script demonstrating proposed AQPLL fundamentals.