Performance analysis of artificial intelligent controllers in PEM fuel cell voltage tracking R. Vinu 1 • Varghese Paul 2 Received: 29 November 2017 / Revised: 30 January 2018 / Accepted: 2 February 2018 Ó Springer Science+Business Media, LLC, part of Springer Nature 2018 Abstract The main criteria to be kept in mind while designing any application using fuel cell is the Voltage Control under sudden load variations. As a standard practice the output voltage of a fuel cell is controlled and maintained to the reference by introducing Intelligent Controllers. This paper shows the performance analysis of various intelligent controllers that can track the output voltage of fuel cell. In this paper, the state space model of Proton Exchange Membrane Fuel cell is considered for analyzing various controllers. Additionally the transient response of the fuel cell is analyzed and compared for the different controllers. The performance of the controllers is evaluated by estimating the time response characteristics of the system and also by calculating the system errors. Keywords Feedback  Intelligent systems  Neural networks  Optimal control  Renewable energy source 1 Introduction With the increase in green house gas emissions due to exorbitant usage of fossil fuels, many research works are carried out with an aim to find an alternative solution for energy generation with minimal green house gas emissions features. Due to the inherent advantages of Proton Exchange Membrane Fuel cell (PEMFC) [1] like minimal pollutant discharge, high efficiency, increased lifetime and high power density, it is considered as a significant renewable energy source and alternate to fossil fuel as compared with fossil fuels and other types of fuel cells [2]. PEM Fuel Cell consists of electrodes anode and cathode separated by an electrolyte called proton exchange mem- brane. The hydrogen pumped in the anode splits into pro- tons and electrons, of which the protons are allowed to flow through the thin membrane and the negative electrons reach the cathode through an external circuit to generate electric current. At the cathode, electrons and protons combine with oxygen to generate water and heat as by- product [1–3]. However for a PEMFC to achieve the desired perfor- mance and for utilizing in any applications, the required output voltage has to be maintained to a predefined desired proportional under sudden load variations. Hence the requirement of an intelligent controller is inevitable and mandatory in a PEMFC to control and maintain the required output voltage as per the varying load requirements. To design an intelligent Controller the knowledge of the input/output variables in a fuel cell system is essential. The relationship between the different critical variables required to run the PEMFC system are important and the same is attained by modeling a fuel cell system. In this paper the accurate model of the PEMFC is described by considering the State space Model of the fuel cell [19]. Neural Networks based Controllers [4] are widely used to control the terminal voltage of a dynamic fuel cell model and the simulation results registered improved performance in tracking the voltage [5–9]. Fuzzy controls are also applicable to the fuel cell system to track the output voltage since fuzzy linguistics helps to evidently define the rela- tionship between the different variables of fuel cell [10]. PID controllers are also widely used to control fuel cell output voltage due to their simple structure and reliability. & R. Vinu vinur81@gmail.com Varghese Paul vp.itcusat@gmail.com 1 Anna University, Chennai, Tamil Nadu, India 2 Toc H Institute of Science and Technology, Cochin, Kerala, India 123 Cluster Computing https://doi.org/10.1007/s10586-018-1992-7