Aiming at the high dimension and complexity of parameters identification problem, a whale optimization algorithm based on event-triggered and dimension learning scheme (EDWOA) is presented, which is specifically designed for the proton exchange membrane fuel cell (PEMFC) model. Drawing inspiration from group optimization strategies, a novel dimensional learning method is introduced to enhance the dynamic search capabilities of the algorithm. To assess the efficacy of the proposed algorithm, benchmark function testing was conducted, and its fitness surpassed common heuristic algorithms on over 15 objective functions. The results clearly indicate that the EDWOA algorithm outperforms its counterparts in terms of global search performance. Its ability to navigate complex search spaces sets it apart from other algorithms. Finally, the proposed EDWOA algorithm is successfully applied to parameter identification in the PEMFC model. Through a comparative analysis with existing research findings, it was found that the identified PEMFC model exhibited a notable enhancement in fitness, ranging from 0.02 to 0.93. This underscores the effectiveness of the EDWOA algorithm in improving the performance and dynamic output of PEMFC models.