Scholarlink

<p>The study presents an intelligent, model-free current control strategy that eliminates the need for explicit plant models while efficiently reducing the effect of plant parameter perturbation. By employing a data-driven approach with fewer input features, the proposed scheme reduces the computational burden during training while maintaining high control performance. Unlike conventional model predictive current control (MPCC), which is computationally expensive because of solving optimization problems at each sample time, and requires precise plant models, the proposed method enhances system performance by addressing plant model discrepancies through data-driven techniques. Additionally, adaptive particle swarm optimization (APSO) is used to optimize the gain parameters of the outer speed control loop for improved dynamic performance. To verify the effectiveness of the data-driven control scheme, a comparative study with a conventional control scheme is presented. We verify that the switching states obtained from the model-based control design are learned with an accuracy of 94.8% using the proposed model-free data-driven approach. Test results show that the proposed approach outperforms traditional methods, offering superior steady-state performance, lower harmonic distortion, and increased robustness.</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

Model-free current control solution employing intelligent control for enhanced motor drive performance

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年1月2日

查看全文
  • Muhammad Usama,
  • Jaehong Kim

摘要

The study presents an intelligent, model-free current control strategy that eliminates the need for explicit plant models while efficiently reducing the effect of plant parameter perturbation. By employing a data-driven approach with fewer input features, the proposed scheme reduces the computational burden during training while maintaining high control performance. Unlike conventional model predictive current control (MPCC), which is computationally expensive because of solving optimization problems at each sample time, and requires precise plant models, the proposed method enhances system performance by addressing plant model discrepancies through data-driven techniques. Additionally, adaptive particle swarm optimization (APSO) is used to optimize the gain parameters of the outer speed control loop for improved dynamic performance. To verify the effectiveness of the data-driven control scheme, a comparative study with a conventional control scheme is presented. We verify that the switching states obtained from the model-based control design are learned with an accuracy of 94.8% using the proposed model-free data-driven approach. Test results show that the proposed approach outperforms traditional methods, offering superior steady-state performance, lower harmonic distortion, and increased robustness.