Neural network and URED observer based fast terminal integral sliding mode control for energy efficient polymer electrolyte membrane fuel cell used in vehicular technologies. (15th April 2023)

• Record Type: Journal Article
• Authors: Javaid, Usman; Mehmood, Adeel; Iqbal, Jamshed; Uppal, Ali Arshad
• Is Part Of: Energy. Volume 269(2023)
• Publication Date: 2023-04-15
• Subjects: Fuel cell -- Neural network -- Oxygen excess ratio -- Uniform robust exact differentiator -- Fast terminal integral sliding mode control -- Vehicular technology; Power resources -- Periodicals; Power (Mechanics) -- Periodicals; Energy consumption -- Periodicals; 333.7905
• DOI: 10.1016/j.energy.2023.126717
• Languages: English
• ISSNs: 0360-5442

Abstract

Abstract: In this research work, a Neural Network (NN) and Uniform Robust Exact Differentiator (URED) observer-based Fast Terminal Integral Sliding Mode Control (FTISMC) has been proposed for Oxygen Excess Ratio (OER) regulation of a Polymer Electrolyte Membrane Fuel Cell (PEMFC) power systems for vehicular applications. The controller uses URED as an observer for supply manifold pressure estimation. NN is used to estimate the stack temperature which is unavailable. The suggested control method increased the PEMFC's effectiveness and durability while demonstrating the finite-time convergence of system trajectories. By controlling the air-delivery system in the presence of uncertain current requirements and measurement noise, the approach ensures maximum power efficiency. The Lyapunov stability theorem has been used to confirm the stability of the presented algorithm. In addition, the suggested method eliminated the chattering phenomenon and improved power efficiency. Given these noteworthy characteristics, the research has the potential to decrease sensor dependence and production costs while also improving the transient and steady-state response in vehicular applications. Highlights: FTISMC is proposed for PEMFC OER regulation. Lyapunov theorem is used to analyze the proposed controller's stability. URED-based observer is used for supply manifold pressure estimation. Stack temperature is estimated based on a non-auto regressive neural network. Robustness of the proposed technique is evaluated for various uncertain conditions.