Robust Energy Management Strategy for Fuel Cell Hybrid Electric Vehicles Based on the RPO-DTRN Framework

Fuel Cell Hybrid Electric Vehicles (FCHEVs) present a promising solution for sustainable transportation, but their widespread usage is limited by challenges like fuel cell stack degradation and inefficient power distribution. Conventional energy management strategies have difficulty in balancing power allocation efficiently while maximizing the fuel cell lifespan. To overcome these constraints, this research introduces a novel hybrid methodology-based energy management strategy for FCHEVs to enhance the fuel cell stack's lifetime. The proposed framework integrates the Red Panda Optimization (RPO) algorithm with a Dual Transformer Residual Network (DTRN), referred to as the RPO-DTRN approach. The RPO algorithm optimizes power distribution in hybrid electric vehicles, ensuring efficient energy utilization, while the DTRN model predicts and improves the remaining useful life of the fuel cell stack. The proposed strategy is evaluated using MATLAB environment and compared against existing techniques, including the Soft Actor-Critic algorithm, Twin Delayed Deep Deterministic Policy Gradient algorithm and Harris Hawks Optimization-Sine Cosine algorithm. Results demonstrate that the proposed RPO-DTRN method outperforms the existing methods, attaining a high efficiency of 97% and the lowest error value of 0.148. Additionally, the proposed strategy minimizes hydrogen consumption to 87.28 g, making it a more fuel-efficient approach. Furthermore, it achieves the highest final state of charge at 79.04%, indicating superior battery energy retention and overall system sustainability. These findings suggest the efficiency of the proposed energy management method in improving the reliability, efficiency, and longevity of FCHEVs.