Scholarlink

<p>Permanent Magnet Synchronous Generators (PMSGs) combined with Cascaded H-Bridge Inverters (CHBIs) are widely adopted in wind energy systems due to their high efficiency and superior power quality. However, even five-level CHBIs retain noticeable low-order harmonic components and output ripple under nonlinear PMSG wind conditions, indicating that further refinement of switching-angle control is required to maximize performance. This paper introduces a dual optimization–prediction framework to address these challenges. The proposed method integrates the Greater Cane Rat Algorithm (GCRA) for adaptive switching-angle optimization with a Visual Relational Spatio-Temporal Neural Network (VRSTNN) for predictive control under dynamic operating conditions. By jointly minimizing harmonic distortion and forecasting system responses under varying wind and load scenarios, the framework ensures high-quality voltage output and stable operation. Across 10 independent simulation runs, the system achieved a mean THD of 2.10% ± 0.04, voltage ripple of 1.6% ± 0.12, and response time improvement from 0.035&#xa0;s to 0.012&#xa0;s, confirming consistent performance with low variability. MATLAB results further demonstrate reduced power losses, improved efficiency, and faster transient stabilization compared with ANN, RERNN-LSE, RPOA-DTRN, GA–PSO, and CNN-based methods. These findings highlight the potential of the dual optimization–prediction strategy as a robust and scalable solution for next-generation intelligent PMSG–CHBI wind energy conversion systems.</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

Harmonic distortion reduction and dynamic stability in PMSG-CHBI wind energy systems via a dual optimization–prediction approach

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年1月26日

查看全文
  • Lijo Jacob Varghese,
  • G. Venkatesan,
  • Aymen Flah,
  • Monia Hamdi

摘要

Permanent Magnet Synchronous Generators (PMSGs) combined with Cascaded H-Bridge Inverters (CHBIs) are widely adopted in wind energy systems due to their high efficiency and superior power quality. However, even five-level CHBIs retain noticeable low-order harmonic components and output ripple under nonlinear PMSG wind conditions, indicating that further refinement of switching-angle control is required to maximize performance. This paper introduces a dual optimization–prediction framework to address these challenges. The proposed method integrates the Greater Cane Rat Algorithm (GCRA) for adaptive switching-angle optimization with a Visual Relational Spatio-Temporal Neural Network (VRSTNN) for predictive control under dynamic operating conditions. By jointly minimizing harmonic distortion and forecasting system responses under varying wind and load scenarios, the framework ensures high-quality voltage output and stable operation. Across 10 independent simulation runs, the system achieved a mean THD of 2.10% ± 0.04, voltage ripple of 1.6% ± 0.12, and response time improvement from 0.035 s to 0.012 s, confirming consistent performance with low variability. MATLAB results further demonstrate reduced power losses, improved efficiency, and faster transient stabilization compared with ANN, RERNN-LSE, RPOA-DTRN, GA–PSO, and CNN-based methods. These findings highlight the potential of the dual optimization–prediction strategy as a robust and scalable solution for next-generation intelligent PMSG–CHBI wind energy conversion systems.