Scholarlink

<p>Sodium nitroprusside (SNP) is a potent vasodilator used for rapid blood pressure regulation in critical care settings. The goal of SNP-based control is to maintain mean arterial blood pressure (MABP) within a desired physiological range. However, patient-specific variability and drug sensitivity introduce significant control challenges. To address this, the proposed work integrates neural networks and Class Topper Optimization for the optimal tuning of a fractional-order proportional–integral–derivative controller. The cost function is designed to reduce control errors while imposing constraints on integral absolute error, integral square error, and integral time absolute error. A neural network predicts safe operating ranges for these indices, while the optimization fine-tunes controller gains. Robustness is evaluated with Monte Carlo simulations across parametric uncertainties. Fault tolerance is ensured using a parallel fault detection and isolation (FDI) scheme with a fallback controller to maintain stability under extreme sensor and actuator faults. The proposed control strategy maintains stable and reliable performance under sensor and actuator faults through the integration of a parallel FDI-based fallback mechanism. Robustness under parameter variations and uncertainties is confirmed using both <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(\mu \)</EquationSource> <EquationSource Format="MATHML"><math> <mi>μ</mi> </math></EquationSource> </InlineEquation>-analysis and Monte Carlo simulations, which show consistent and bounded system responses across different patient models. Overall, the framework significantly enhances precision, robustness, and safety in automated blood pressure regulation. A graphical representation of the work is presented in Fig. <InternalRef RefID="Fig1">1</InternalRef>. </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

A Fault-Tolerant Adaptive FOPID Controller for Blood Pressure Regulation Using Hybrid Neural-Class Topper Optimization

  • 期刊论文

  • 发表日期：

    2026年4月18日

查看全文
  • K. R. Achu Govind,
  • G. L. Karthik,
  • D. Sakthi Kumar

摘要

Sodium nitroprusside (SNP) is a potent vasodilator used for rapid blood pressure regulation in critical care settings. The goal of SNP-based control is to maintain mean arterial blood pressure (MABP) within a desired physiological range. However, patient-specific variability and drug sensitivity introduce significant control challenges. To address this, the proposed work integrates neural networks and Class Topper Optimization for the optimal tuning of a fractional-order proportional–integral–derivative controller. The cost function is designed to reduce control errors while imposing constraints on integral absolute error, integral square error, and integral time absolute error. A neural network predicts safe operating ranges for these indices, while the optimization fine-tunes controller gains. Robustness is evaluated with Monte Carlo simulations across parametric uncertainties. Fault tolerance is ensured using a parallel fault detection and isolation (FDI) scheme with a fallback controller to maintain stability under extreme sensor and actuator faults. The proposed control strategy maintains stable and reliable performance under sensor and actuator faults through the integration of a parallel FDI-based fallback mechanism. Robustness under parameter variations and uncertainties is confirmed using both \(\mu \) μ -analysis and Monte Carlo simulations, which show consistent and bounded system responses across different patient models. Overall, the framework significantly enhances precision, robustness, and safety in automated blood pressure regulation. A graphical representation of the work is presented in Fig. 1.