PEM Fuel Cell System Identification and Control

  • Pinagapani Arun Kumar
  • Mani Geetha
  • K.R. Chandran
  • P. Sanjeevikumar
Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 435)


A model is an input–output mapping that suitably explains the behavior of a system. Model helps to analyze the functionality of the system and to design suitable controllers. System identification builds model from experimental data obtained by exciting the process with an input and observing its response at regular interval (Wibowo et al. in System identification of an interacting series process for real-time model predictive control, American Control Conference, pp. 4384–4389, 2009). Fuel cells (FC) systems are a potentially good clean energy conversion technology, and they have wide range of power generation applications. Classification of fuel cells is based on the fuel and the electrolyte type used. The proton exchange membrane fuel cells (PEMFC) are portable devices with superior performance and longer life. They act as a good source for ground vehicle applications. They also possess high power density and fast start-up time. In this work, mathematical model of a real-time PEMFC is obtained and its quality is assessed using various validation techniques. The model is obtained using system identification tool in MATLAB, and validation procedures like recursive least square algorithm, ARX and ARMAX were employed to assess the model. Controllers such as PI and PID were employed in order to achieve the desired load current by controlling the hydrogen flow rate. The values of the gain constant, integral time and derivative time were obtained using Cohen-Coon method. PI and PID control schemes were implemented using SIMULINK in MATLAB environment, and the system response was observed.


Proton exchange membrane fuel cell (PEMFC) Recursive least square algorithm ARX ARMAX Cohen-Coon PI PID 


  1. 1.
    Wibowo, T.C.S., Saad, N., Karsiti, M.N.: System identification of an interacting series process for real-time model predictive control, American Control Conference, pp. 4384–4389 (2009) Google Scholar
  2. 2.
    Spiegel, C.: PEM fuel cell modeling and simulation using MATLAB. Academic Press, London (2011)Google Scholar
  3. 3.
    Ljung, L.: System identification—theory for the user, 2nd edn. Prentice-Hall, New Jersey (1999)Google Scholar
  4. 4.
    Astrom, K.J., Wittenmark, B.: Adaptive control, 2nd edn. Dover Publications, New York (2008)Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2018

Authors and Affiliations

  • Pinagapani Arun Kumar
    • 1
  • Mani Geetha
    • 2
  • K.R. Chandran
    • 3
  • P. Sanjeevikumar
    • 4
  1. 1.Department of Instrumentation & Control Systems EngineeringPSG College of TechnologyCoimbatoreIndia
  2. 2.School of Electrical EngineeringVellore Institute of Technology (VIT) UniversityVelloreIndia
  3. 3.Department of Information TechnologyPSG College of TechnologyCoimbatoreIndia
  4. 4.Department of Electrical and Electronics EngineeringUniversity of JohannesburgJohannesburgSouth Africa

Personalised recommendations