Optimal parameter estimation for PEMFC using modified monarch butterfly optimization

The present study develops a new optimization method called monarch butterfly optimization algorithm for optimal parameter estimation of the polymer electrolyte membrane fuel cell (PEMFC). After designing the proposed methodology, it is implemented to 250 MW PEMFC stack under different operating conditions to show the system efficiency and the results are compared with some state-of-the-art methods including Grass Fibrous Root Optimization Algorithm, hybrid Teaching Learning Based Optimization-Differential Evolution Algorithm, and the basic MBO algorithm. Two operational conditions in 3/5 bar and 80°C and 2.5/3 bar, 70°C are used for model verification. The main idea is to minimize the sum of square error (SSE) between the estimated and the actual data. Simulation results in the first condition give an SSE of value 7.277667729 with 9.28434e−16 SD value and in the second condition, an SSE of value 4.52810115 with 0.043581628 standard deviations has been reached as the minimum value among the other compared methods that indicate the accuracy and the robustness of the suggested method toward the analyzed methods. The algorithm also gives a convergence speed of 540 iterations and 370 iterations for conditions 1 and 2, respectively that are the fastest in the study.