Balanced version of Slime Mold Algorithm: A study on PEM fuel cell system parameters identification

Abstract

A new technique is proposed in this study for optimum selection of the unknown variables in the Proton Exchange Membrane Fuel Cells (PEMFCs). The major purpose is to present an efficient method for minimizing the error between the estimated and the empirical output voltages by the optimal valuation of the model parameters. To do so, a balanced version of the Slime Mold Algorithm (bSMA) is introduced and validated. The method is then performed into three standard benchmarks including NedStack PS6, Horizon H-12, and 5 kW Ballard Mark V and the results are compared with three other state-of-the-art algorithms including Blackhole Algorithm, Locust Swarm Optimization Algorithm, and the original Slime mold algorithm. The results show that the proposed bSMA with 3.01, 1.75, and 0.104 for NedStack PS6, Horizon H-12, and 5 kW Ballard Mark V, respectively has the minimum value of error. Therefore, this proves that the proposed technique gives the best results against the others for model identification. Also, to provide more analysis, the consistency of the suggested method is investigated under different temperature and pressure conditions which shows that the suggested technique has reliable results toward different conditions.

Keywords

• PEMFC
• Optimization
• Balanced Slime Mold Algorithm
• Test cases
• Model identification