Parameters Extraction of PEMFC Model Using Evolutionary Based Optimization Algorithms

In this article, evolutionary based algorithms are used to extract the optimum values of parameters of a proton exchange membrane fuel cell (PEMFC). The mathematical model of fuel cell which is a complex non-linear model, has been developed and optimized using evolutionary based optimization algorithm such as Genetic Algorithm (GA), Mematic Algorithm (MA), Evolutionary Programming (EP), Flower Pollination Algorithm (FPA), and Coral Reefs Optimization (CRO). The optimized models are applied and tested on a NedStack PS6 (6 kW) fuel cell stack. The objective function is considered as minimization of sum of square error between the experimental and estimated stack voltages. The statistical measurements such as mean and standard deviation are determined for the optimized model. The best algorithm has also been reported for the proposed model using box plots for all the algorithms. The polarization curves including I–V curves, power curves and efficiency curves are plotted for both measured and optimized points. Each algorithm has also been compared with other algorithms for evaluating the most effective evolutionary algorithm in solving the fuel cell optimization problem. The results reveal that Memetic Algorithm (MA) algorithm is the best suited evolutionary algorithm among all considered algorithms with least sum of square error, mean and standard deviation values of 2.37, 3.267 and 0.533 respectively. The box plot also shows the lowest median and smallest Interquartile Range (IQR) for Mematic Algorithm (MA). The results of this model can help in determining the optimal parameters of a practical fuel cell stack.