In the speed control system of an Interior Permanent Magnet Synchronous Motor (IPMSM) without a speed sensor, PI controllers using only a fixed set of parameters cannot achieve accurate tracking of the estimated speed in a wide speed domain and also suffer from step response overshoot. This paper proposes a Compound Variable Structure PI (CVSPI) controller to improve the system control performance. It can choose whether to include an integral term according to the size of the system deviation to speed up the response. It also introduces a Model Reference Adaptive System (MRAS) speed observer in the controller to estimate the speed and adaptively adjust the size of the anti-integration saturation gain to improve the dynamic response following performance and immunity of the system. A feed-forward link is added for a given input differential to achieve an accurate answer to time-varying inputs. As the linear compensation matrix of the conventional MRAS is a unit matrix, the speed can only be accurately observed in a specific speed range. In this paper, a new linear compensation matrix is designed, and a new speed adaptive law is derived, allowing the improved MRAS to measure speed over a wide range accurately. Simulation results validate the excellent control performance of the CVSPI and the accuracy of the enhanced MRAS over a wide speed range.
To improve the accuracy of speed estimation strategy over a wide speed range in the sensorless speed control system of an interior permanent – magnet synchronous motor (IPMSM), a full-order adaptive observer based on a novel fast super-twisting algorithm (NFSTA-AO) is proposed. The conventional model reference adaptive system (MRAS) takes the linear compensation matrix of the unit matrix, which can only achieve speed discrimination within a certain speed range. Therefore, in this paper, a new linear compensation matrix is first derived using Popov’s super stability theory, and then, a new tachograph – adaptive law is obtained. A feedback correction link is also added to the adjustable model to improve the convergence speed of the error between the reference and adjustable model outputs. To further improve the accuracy of the tachograph estimation strategy, a novel fast super-twisting algorithm (NFSTA) convergence law is introduced in place of the adaptive law in the full-order adaptive observer, combining the advantages of both algorithms. The NFSTA added the inverse hyperbolic sine function based on system state variables to the integral term of the fast super-twisting algorithm (FSTA) to effectively suppress the torque pulsation of the system. A soft switching function is also designed to replace the symbolic switching function in order to reduce the system jitter caused by the sliding mode variable structure control. The simulation experiments show that the system using the NFSTA-AO estimation strategy is more resistant to disturbances and robust; additionally, it has better dynamic following than the conventional MRAS in the presence of added load disturbances and sudden speed changes.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.