Scholarlink

<p>This paper presents a <i>H</i><sub><i>∞</i></sub> loop-shaping (HLS) controller for regulating the depth of anesthesia (DoA), a critical aspect of automated anesthesia delivery. Given the high inter-patient variability in response to anesthetic drugs, the DoA system is inherently subject to significant uncertainty. The Bispectral Index (BIS), a quantitative measure of hypnotic depth, is used as the controlled variable. To ensure both robust stability and performance while meeting stringent patient safety requirements, the proposed HLS framework is augmented with Particle Swarm Optimization (PSO) to optimally tune the weighting functions. Numerical simulations demonstrate that the proposed optimized HLS architecture delivers superior trajectory tracking relative to standard HLS and PID controllers. The design achieves a 60.25% and 87.0% reduction in settling time, and a 93.0% and 85.0% reduction in rise time, respectively, compared to the baseline methods. Notably, these performance gains are realized while preserving a low-complexity structure and guaranteeing robust closed-loop behavior under model mismatches and load disturbances.</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

Particle swarm optimization based H∞ loop shaping robust control for anesthesia systems

  • 期刊论文

  • 发表日期：

    2026年7月23日

查看全文
  • Rawnaq A. Mahmod,
  • Safanah M. Raafat,
  • Huthaifa Al-Khazraji,
  • Amjad J Humaidi

摘要

This paper presents a H∞ loop-shaping (HLS) controller for regulating the depth of anesthesia (DoA), a critical aspect of automated anesthesia delivery. Given the high inter-patient variability in response to anesthetic drugs, the DoA system is inherently subject to significant uncertainty. The Bispectral Index (BIS), a quantitative measure of hypnotic depth, is used as the controlled variable. To ensure both robust stability and performance while meeting stringent patient safety requirements, the proposed HLS framework is augmented with Particle Swarm Optimization (PSO) to optimally tune the weighting functions. Numerical simulations demonstrate that the proposed optimized HLS architecture delivers superior trajectory tracking relative to standard HLS and PID controllers. The design achieves a 60.25% and 87.0% reduction in settling time, and a 93.0% and 85.0% reduction in rise time, respectively, compared to the baseline methods. Notably, these performance gains are realized while preserving a low-complexity structure and guaranteeing robust closed-loop behavior under model mismatches and load disturbances.