A Computationally Efficient Quasi-Centralized DMPC for Back-to-Back Converter PMSG Wind Turbine Systems Without DC-Link Tracking Errors

Quasi-centralized direct model predictive control (QC-DMPC) scheme may serve as an effective alternative for back-to-back power converter in permanent magnet synchronous generator (PMSG) wind turbine systems. However, model errors and imperfect power efficiency lead to evident dc-link voltage tracking offset. This paper proposes a revised quasi-centralized direct model predictive control (RQC-DMPC) scheme for back-to-back converter PMSG wind turbine systems, within which, the dc-link voltage is directly controlled by a grid side predictive controller with a flexibly designed cost function using a revised dynamic reference generation concept. The dc-link voltage steady status tracking errors are eliminated. To reduce the computational efforts of the classical scheme, a computational efficient concept is incorporated into the proposed method. The proposed scheme is implemented on an entirely field programmable gate array-based platform. The effectiveness of the proposed method is verified through experimental data. The dc-link control performance comparison with classical proportional-integration controller-based methods and the QC-DMPC scheme under different scenarios are also experimentally investigated. The results emphasize the improvement of the proposed RQC-DMPC scheme.