控制理论(社会学)
伺服电动机
同步电动机
磁铁
观察员(物理)
伺服机构
永磁同步电动机
计算机科学
伺服
物理
控制工程
工程类
人工智能
电气工程
量子力学
控制(管理)
作者
Shen Wei,Luwei Shao,Dongchen Liu,Junzheng Wang,Chuanyu Ge
标识
DOI:10.1109/tase.2024.3377639
摘要
This work proposes an event-triggered finite control set model predictive control (FCS-MPC) strategy with a sliding mode observer (SMO) for permanent magnet synchronous motor (PMSM) drives to reduce calculation burden and switching losses, which arise from redundant control actions, without sacrificing the servo tracking performances. A sliding mode observer is utilized to identify load torque to enhance the robustness against loading disturbance for the PMSM control system. The estimation outputs are utilized as feed-forward compensation for the references, which are determined or constructed by the outer-loop controller. By introducing an electrical angular velocity region and decoupling the time-varying system function, a developed mathematical description for the PMSM model is reconstructed. An event-triggered FCS-MPC scheme with the triggering condition derived from the proposed PMSM model is carried out to deal with the calculation burden problem, which is the main challenge for applying the FCS-MPC theory to power electronics. Extensive comparative experiments are completed to validate the stability and efficiency of the proposed strategy Note to Practitioners —This study is motivated by the difficulties in practical applications for the FCS-MPC methodology to PMSM drives. In recent years, a large amount of FCS-MPC strategies have been investigated with respect to the servo motion and tracking performance for PMSM. Realized by enumeration-based predictions, the control algorithm brings exponentially increased calculation burden for long prediction horizons, which are grown to benefit the important closed-loop performance. The efficient traditional FCS-MPC algorithm for PMSM drives puts forward considerable demands on hardware computation resources, which leads to high expense and harmful time delay. To our practical knowledge, there exist repetitive optimized control inputs provided by the FCS-MPC controller for PMSM drives, such that the resulting redundant actions in electronic circuits contain challenges in energy consumption and heat dissipation. Furthermore, coupling disturbances are inevitable for industrial applications of PMSM, among which the load torque has a major impact on control performances. To address these problems, an event-triggering mechanism and a slide mode observer (SMO) are introduced into the FCS-MPC scheme for PMSM servo systems. The triggering condition is designed with guaranteed input-to-state stability. The estimation results from the observer are used for the feed-forward channel, and the enumeration-based prediction for FCS-MPC is processed only when the triggering condition is satisfied. The experimental results demonstrate the effectiveness in reducing the computational burden and switching losses, as well as strengthening system robustness against unknown loading disturbances.
科研通智能强力驱动
Strongly Powered by AbleSci AI