Developed teamwork optimizer for model parameter estimation of the proton exchange membrane fuel cell

This paper proposes a new optimal methodology for model parameters estimation of the Proton Exchange Membrane Fuel Cell. The main purpose here is to design a newly developed metaheuristic technique to deliver a model with higher accuracy. In this study, we utilized two modifications for the Teamwork Optimizer to get higher accuracy. The two modifiers are opposition-based learning and chaotic mechanism. The results show that using the opposition-based learning, the population diversity has been kept, owing to the greater population size due to the solution space, and using the Chaos theory, the population diversity has been increased. This is proved by applying the Improved Teamwork Optimizer to minimize the Root Mean Square Error and Integral Absolute Error between the suggested model and empirical data. The validation has been done by applying the proposed Improved Teamwork Optimizer to two studied cases, which are Nexa Proton Exchange Membrane Fuel Cell and NedSstack PS6 Proton Exchange Membrane Fuel Cell, and comparing it with other published works. Simulation results showed that the proposed method with 1.14 Integral Absolute Error and 0.21 Root Mean Square Error for NedSstack PS6 Proton Exchange Membrane Fuel Cells and with 12 Integral Absolute Error and 0.17 Root Mean Square Error for Nexa Proton Exchange Membrane Fuel Cells provides the minimum error value among the other optimization techniques. This shows the higher potential of the proposed method for use as the parameter estimator for Proton Exchange Membrane Fuel Cells