Scholarlink

<p>The traditional full-order sliding mode observer (FOSMO) applied to the surface-mounted permanent magnet synchronous motors (SPMSM) is typically based on low-order sliding mode algorithms, which inevitably result in a contradiction between dynamic response and chattering suppression. In addition, achieving speed reversal through the extraction of position information from the estimated back electromotive force presents significant difficulties. To address these issues, this paper proposes a fast super-twisting algorithm-based full-order sliding mode observer (FSTA-FOSMO), in conjunction with a bandwidth adaptive feedforward phase-locked loop (BAFPLL), to enhance the sensorless control performance of SPMSMs. The proposed FSTA-FOSMO integrates additional linear terms into the traditional super-twisting algorithm (STA) and replaces the discontinuous sign function with the continuous saturation function, resulting in enhanced steady-state and dynamic performance compared to the traditional FOSMO. Moreover, the proposed BAFPLL normalizes the input, eliminates the sign term of the electrical angular velocity, and enables adaptive bandwidth adjustment based on the position error. The inclusion of electrical angular velocity feedforward mitigates electrical angle errors and enhances position estimation accuracy. Comprehensive evaluations conducted on an experimental test bench confirm that the proposed FSTA-FOSMO integrated with the BAFPLL significantly enhances transient performance and steady-state precision while exhibiting superior speed reversal performance.</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

Fast super-twisting full-order sliding mode observer for SPMSM with bandwidth adaptive feedforward PLL

  • 期刊论文

  • 发表日期：

    2026年4月30日

查看全文
  • Fujiong Zhao,
  • Gaosheng Li,
  • Jianfeng Mao,
  • Baisong Pan,
  • Yingjie Xu,
  • Rongsheng Jia,
  • Chunhui Ye,
  • Weigang Wang

摘要

The traditional full-order sliding mode observer (FOSMO) applied to the surface-mounted permanent magnet synchronous motors (SPMSM) is typically based on low-order sliding mode algorithms, which inevitably result in a contradiction between dynamic response and chattering suppression. In addition, achieving speed reversal through the extraction of position information from the estimated back electromotive force presents significant difficulties. To address these issues, this paper proposes a fast super-twisting algorithm-based full-order sliding mode observer (FSTA-FOSMO), in conjunction with a bandwidth adaptive feedforward phase-locked loop (BAFPLL), to enhance the sensorless control performance of SPMSMs. The proposed FSTA-FOSMO integrates additional linear terms into the traditional super-twisting algorithm (STA) and replaces the discontinuous sign function with the continuous saturation function, resulting in enhanced steady-state and dynamic performance compared to the traditional FOSMO. Moreover, the proposed BAFPLL normalizes the input, eliminates the sign term of the electrical angular velocity, and enables adaptive bandwidth adjustment based on the position error. The inclusion of electrical angular velocity feedforward mitigates electrical angle errors and enhances position estimation accuracy. Comprehensive evaluations conducted on an experimental test bench confirm that the proposed FSTA-FOSMO integrated with the BAFPLL significantly enhances transient performance and steady-state precision while exhibiting superior speed reversal performance.