Sensorless interior permanent magnet synchronous motor control with rotational inertia adjustment

Yongle Mao,Dejun Yin,Yangsheng Chen,Jiaqiang Yang

doi:10.1177/1687814016684745

Abstract

Mechanical model is generally required in high dynamic sensorless motor control schemes for zero phase lag estimation of rotor position and speed. However, the rotational inertia uncertainty will cause dynamic estimation errors, eventually resulting in performance deterioration of the sensorless control system. Therefore, this article proposes a high dynamic performance sensorless control strategy with online adjustment of the rotational inertia. Based on a synthetic back electromotive force model, the voltage equation of interior permanent magnet synchronous motor is transformed to that of an equivalent non-salient permanent magnet synchronous motor. Then, an extended nonlinear observer is designed for interior permanent magnet synchronous motor in the stator-fixed coordinate frame, with rotor position, speed and load torque simultaneously estimated. The effect of inaccurate rotational inertia on the estimation of rotor position and speed is investigated, and a novel rotational inertia adjustment approach that employs the gradient descent algorithm is proposed to suppress the dynamic estimation errors. The effectiveness of the proposed control strategy is demonstrated by experimental tests.

Highlights

Due to its high torque density, high efficiency and wide constant-power operating range, interior permanent magnet synchronous motor (IPMSM) has been widely used for industrial applications and vehicle propulsions.[1,2,3,4] In high-performance field-oriented control systems of IPMSM, rotor position information is generally required for the coordinate transformation and speed feed-back control
It can be observed from equation (1) that there is a 2ue-dependent inductance matrix contained in the voltage equation of IPMSM due to the rotor saliency, which is inconvenient in applying the nonlinear observer to IPMSM for rotor position and speed estimation
This article proposed a high dynamic performance sensorless control scheme for IPMSM based on an extended nonlinear observer with compensation for rotational inertia uncertainty

Summary

Introduction

Due to its high torque density, high efficiency and wide constant-power operating range, interior permanent magnet synchronous motor (IPMSM) has been widely used for industrial applications and vehicle propulsions.[1,2,3,4] In high-performance field-oriented control systems of IPMSM, rotor position information is generally required for the coordinate transformation and speed feed-back control. The voltage equation of IPMSM in the aÀb frame is directly given as d uab = Rsiab + L0 dt iab + Eab + hab ð1Þ where uab = (uaub)T is the stator voltage vector in the a2b frame, and ua and ub are the stator voltage components of a- and b-axis, respectively; iab = (iaib)T is the stator current vector in the a2b frame, and ia and ib are the stator current components of a- and b-axis, respectively; Eab = Pnvmcf( À sin ue cos ue)T is the back-EMF vector in the a2b frame; Rs is the stator resistance and cf is the rotor flux-linkage; vm and ue are mechanical speed and electrical position, respectively; Pn is the motor pole pairs; and It can be observed from equation (1) that there is a 2ue-dependent inductance matrix contained in the voltage equation of IPMSM due to the rotor saliency, which is inconvenient in applying the nonlinear observer to IPMSM for rotor position and speed estimation. To extract the rotor position and speed of IPMSM from the stator voltages and currents instead of installing a mechanical position sensor, an extended nonlinear observer is built in the aÀb frame employing the estimated state variables of equation (9), as well as the estimated load torque. Estimation errors of the stator currents are used to regulate the estimated state variables

CCCCCCCCCA

CCCCA l1 l2 0

Experiments and analysis

Findings

Conclusion