%0 Journal Article %T Implementation of Adaptive Neuro-Fuzzy Inference System (Anfis) for Performance Prediction of Fuel Cell Parameters %J Journal of Renewable Energy and Environment %I Materials and Energy Research Center (MERC) Iranian Association of Chemical Engineers (IAChE) %Z 2423-5547 %A Mostafaeipour, Ali %A Qolipour, Mojtaba %A Goudarzi, Hossein %A Jahangiri, Mehdi %A Golmohammadi, Amir-Mohammad %A Rezaei, Mostafa %A Goli, Alireza %A Sadeghikhorami, Ladan %A Sadeghi Sedeh, Ali %A Khalifeh Soltani, Seyad Rashid %D 2019 %\ 07/01/2019 %V 6 %N 3 %P 7-15 %! Implementation of Adaptive Neuro-Fuzzy Inference System (Anfis) for Performance Prediction of Fuel Cell Parameters %K Adaptive Neuro-Fuzzy Inference System (ANFIS) %K Fuel Cell %K optimization %K Proton Exchange Membrane %R 10.30501/jree.2019.98989 %X Fuel cells are potential candidates for storing energy in many applications; however, their implementation is limited due to poor efficiency and high initial and operating costs. The purpose of this research is to find the most influential fuel cell parameters by applying the adaptive neuro-fuzzy inference system (ANFIS). The ANFIS method is implemented to select highly influential parameters for proton exchange membrane (PEM) element of fuel cells. Seven effective input parameters are considered including four parameters of semi-empirical coefficients, parametric coefficient, equivalent contact resistance, and adjustable parameter. Parameters with higher influence are then identified. An optimal combination of the influential parameters is presented and discussed. The ANFIS models used for predicting the most influential parameters in the performance of fuel cells were performed by the well-known statistical indicators of the root-mean-squared error (RMSE) and coefficient of determination (R2). Conventional error statistical indicators, RMSE, r, and R2, were calculated. Values of R2 were calculated as of 1.000, 0.9769, and 0.9652 for three different scenarios, respectively. R2 values showed that the ANFIS could be properly used for yield prediction in this study %U https://www.jree.ir/article_98989_1a0cdf02cb7dc1c7526c2105cb5b1ba7.pdf