Regular Article

Refining efficiency in standalone proton exchange membrane fuel cell systems through gross hopper optimization-based maximum power point tracking control

¹ School of Electrical and Electronics Engineering, REVA University, Bangalore 560064, India
² School of Rural Management, KIIT Deemed to be University, Bhubaneswar 751024, India
³ School of Electrical Engineering, KIIT Deemed to be University, Bhubaneswar 751024, India
⁴ School of Electrical Engineering, Vellore Institute of Technology, Vellore 632014, India
⁵ Department of Mechanical Engineering, Siksha ‘O’ Anusandhan Deemed to be University, Bhubaneswar 751030, India

^* Corresponding author: abinashmahapatro@gmail.com

Received: 23 October 2024
Accepted: 26 March 2025

Abstract

This study introduces a novel Maximum Power Point Tracking (MPPT) technique for Proton Exchange Membrane Fuel Cell (PEMFC) systems, leveraging the Gross Hopper Optimization (GHO) algorithm to achieve enhanced performance. The proposed method is applied to a stand-alone PEMFC system with a power capacity of 1.2 kW. The primary problem addressed is the challenge of achieving efficient and reliable MPPT in dynamic operating conditions, which is critical for optimizing PEMFC performance and extending its lifespan. Unlike conventional optimization techniques, the GHO algorithm is parameter-independent, making it highly adaptive and suitable for diverse and fluctuating operational scenarios. To further improve prediction accuracy, the GHO algorithm incorporates a natural cubic-spline prediction model within its iterative mechanism, which enhances power generation predictions under dynamic conditions such as abrupt changes in fuel cell temperature and reactant partial pressure. The performance of the system is evaluated through extensive simulations under steady-state and transient conditions. The key findings reveal that the proposed method achieves a tracking efficiency of more than 98.3% under standard operating conditions and maintains an efficiency greater than 96.5% during dynamic changes, outperforming the controllers based on the adaptive Neural Network (NN) and the Adaptive Neuro-Fuzzy Inference System (ANFIS). Furthermore, the GHO-based controller demonstrates faster response times with a 30% improvement in settle time and greater robustness to parameter variations compared to the benchmarks.