Sensorless Control for IPMSM Based on Adaptive Super-Twisting Sliding-Mode Observer and Improved Phase-Locked Loop

Abstract

In traditional sensorless control of the interior permanent magnet synchronous motors(IPMSMs) for medium and high speed domains, a control strategy based on a sliding-mode observer(SMO) and phase-locked loop (PLL) is widely applied. A new strategy for IPMSM sensorless controlbased on an adaptive super-twisting sliding-mode observer and improved phase-locked loop isproposed in this paper. A super-twisting sliding-mode observer (STO) can eliminate the chatteringproblem without low-pass filters (LPFs), which is an effective method to obtain the estimated backelectromotive forces (EMFs). However, the constant sliding-mode gains in STO may causeinstability in the high speed domain and chattering in the low speed domain. The speed-relatedadaptive gains are proposed to achieve the accurate estimation of the observer in wide speed rangeand the corresponding stability is proved. When the speed of IPMSM is reversed, the traditionalPLL will lose its accuracy, resulting in a position estimation error of 180°. The improved PLL basedon a simple strategy for signal reconstruction of back EMF is proposed to ensure that the motor canrealize the direction switching of speed stably. The proposed strategy is verified by experimentaltesting with a 60-kW IPMSM sensorless drive.

Improved Linear Active Disturbance Rejection Controller Control Considering Bus Voltage Filtering in Permanent Magnet Synchronous Generator

“…Then we have β 1 = 3ω 0 , β 2 = 3ω 2 0 , β 3 = ω 3 0 . It can be known from equation (26) that the time constant of the filtering link in the feedback channel does not affect the anti-interference performance, tracking performance and noise suppression performance.…”

“…From equation (26), the transfer function of the feedback channel to the control output is obtained by equation (45), and the transfer function of the observed noise to the system output is expressed by equation 46:…”

“…The scientific research work of filtering the measured signal includes: in [25], Kalman filter is proposed to filter the measured signal of the controlled system, so that the reduced-order extended state observer has higher observation accuracy, but the computation of this filtering algorithm is too large to affect its convergence speed. In [26], a robust differentiator based on super twisting algorithm is proposed, which can filter the input signal, and its convergence speed is more dependent on the selection of control parameters, and no parameter tuning rules are given. All of the above methods can effectively filter the measured signal, but the difference of amplitude and phase between the filtered signal and the actual system output is not considered, which affects the output of the controller and reduces the system performance.…”

Second‐order lead compensator‐based quadrature PLL for sensorless interior permanent magnet synchronous motor control

“…Model-based methods, which obtain the rotor position information by utilising the back electromotive force (EMF) or flux linkage model, are commonly suitable for medium-and high-speed domains [7][8][9][10][11][12][13][14][15][16][17][18][19]. Model-based methods are primarily divided into 'active flux' observers [7,8], Kalman filtering methods [9,10], model adaptive methods [11,12], and sliding mode observers [13][14][15][16][17][18][19] etc. Among these methods, the sliding mode observer is more spaciously utilised in sensorless control strategy on account of its simple structure and strong robustness.…”

A Super-Twisting Sliding-Mode Stator Flux Observer for Sensorless Direct Torque and Flux Control of IPMSM

“…With the speed increasing to the medium domain, model-based schemes are widely utilized to observe the position [12][13][14][15][16]. Due to its ease of implementation, plenty of research is focused on the designation of the back-emf observer like the extended Kalman filter [12] and the sliding-mode observer (SMO) [13]. However, the capability of the back-emf-based sensorless scheme is poor at low speeds because of the non-negligible noise.…”