Scholarlink

<p>The performance and durability of Proton Exchange Membrane Fuel Cell (PEMFC) systems are strongly affected by the oxygen excess ratio (OER). Optimizing the OER is essential for improving energy conversion efficiency, maximizing net power output, and extending operational lifetime. Accurately measuring and controlling the OER remains challenging due to the complex dynamic interactions within air supply and consumption mechanisms. This study addresses these challenges by proposing a neuro-adaptive model-reference controller featuring an embedded neural-network observer (NNO). The controller incorporates a Maximum Power Point Tracking (MPPT) algorithm for real-time OER optimization and a model-free observer that estimates unmeasured state variables without prior training data. A model-reference adaptive controller based on a radial basis function (RBF) neural network is implemented to enhance robustness and improve OER tracking under uncertainties and disturbances. Lyapunov’s stability theory is used to demonstrate that the tracking error remains uniformly bounded. Simulation results in the MATLAB Simulink environment indicate that the proposed controller outperforms traditional PI and Active Disturbance Rejection Control (ADRC) strategies. The neuro-adaptive controller achieves lower mean absolute error (MAE), root mean square error (RMSE), and standard deviation (SD) in OER tracking, highlighting its effectiveness in maintaining target values despite external 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

Neuro-adaptive control of optimal excess oxygen ratio of proton exchange membrane fuel cell

  • 期刊论文

  • 发表日期：

    2026年4月3日

查看全文
  • Abdelaziz EL Aoumari,
  • Hamid Ouadi,
  • Jamal El-Bakkouri,
  • Fouad Giri

摘要

The performance and durability of Proton Exchange Membrane Fuel Cell (PEMFC) systems are strongly affected by the oxygen excess ratio (OER). Optimizing the OER is essential for improving energy conversion efficiency, maximizing net power output, and extending operational lifetime. Accurately measuring and controlling the OER remains challenging due to the complex dynamic interactions within air supply and consumption mechanisms. This study addresses these challenges by proposing a neuro-adaptive model-reference controller featuring an embedded neural-network observer (NNO). The controller incorporates a Maximum Power Point Tracking (MPPT) algorithm for real-time OER optimization and a model-free observer that estimates unmeasured state variables without prior training data. A model-reference adaptive controller based on a radial basis function (RBF) neural network is implemented to enhance robustness and improve OER tracking under uncertainties and disturbances. Lyapunov’s stability theory is used to demonstrate that the tracking error remains uniformly bounded. Simulation results in the MATLAB Simulink environment indicate that the proposed controller outperforms traditional PI and Active Disturbance Rejection Control (ADRC) strategies. The neuro-adaptive controller achieves lower mean absolute error (MAE), root mean square error (RMSE), and standard deviation (SD) in OER tracking, highlighting its effectiveness in maintaining target values despite external disturbances.