Scholarlink

<p>As a critical component in wind energy systems, the doubly fed induction generator’s control performance affects overall power conversion efficiency considerably. To address the limitations in conventional model predictive current control (MPCC) approaches—including substantial current/torque pulsations, limited control precision, and parameter sensitivity—this study proposes an enhanced fast three-vector MPCC (FTV-MPCC) strategy that incorporates online parameter identification through an extended Kalman filter (EKF). The methodology comprises three principal innovations. First, through real-time computation of reference rotor voltage values, the algorithm identifies the primary optimal voltage vector rapidly with minimal error, reducing selection time considerably. Subsequent selection of secondary optimal and zero vectors from remaining options, coupled with precise duration calculation, enables effective voltage vector synthesis. Second, integration with motor mathematical models facilitates construction of EKF iterative equations, enabling real-time identification of stator resistance and inductance parameters. These updated parameters are dynamically fed back to the FTV-MPCC controller, enhancing system robustness against parameter variations. Finally, the comprehensive simulation and experimental validation through comparative analysis of steady-state performance, dynamic response characteristics, and parameter robustness conclusively demonstrate the proposed algorithm’s superior performance over conventional MPCC approaches.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Predictive current control of DFIG model based on EKF parameter identification

  • 期刊论文

  • 发表日期：

    2025年6月27日

查看全文
  • Shenyuan Yu,
  • Shuxi Liu,
  • Ke Liu,
  • Bo Tang,
  • Zhen Wang

摘要

As a critical component in wind energy systems, the doubly fed induction generator’s control performance affects overall power conversion efficiency considerably. To address the limitations in conventional model predictive current control (MPCC) approaches—including substantial current/torque pulsations, limited control precision, and parameter sensitivity—this study proposes an enhanced fast three-vector MPCC (FTV-MPCC) strategy that incorporates online parameter identification through an extended Kalman filter (EKF). The methodology comprises three principal innovations. First, through real-time computation of reference rotor voltage values, the algorithm identifies the primary optimal voltage vector rapidly with minimal error, reducing selection time considerably. Subsequent selection of secondary optimal and zero vectors from remaining options, coupled with precise duration calculation, enables effective voltage vector synthesis. Second, integration with motor mathematical models facilitates construction of EKF iterative equations, enabling real-time identification of stator resistance and inductance parameters. These updated parameters are dynamically fed back to the FTV-MPCC controller, enhancing system robustness against parameter variations. Finally, the comprehensive simulation and experimental validation through comparative analysis of steady-state performance, dynamic response characteristics, and parameter robustness conclusively demonstrate the proposed algorithm’s superior performance over conventional MPCC approaches.