Model predictive control for regulating fuel cell stack temperature and air flow rate

Stack temperature and airflow rate are vital control problems for Proton Exchange Membrane fuel cell (PEMFC). Two separate Model Predictive Controllers (MPC) have been employed to regulate these problems. The controllers utilized Laguerre and exponential weight functions to reduce its numerical inst...

Full description

Saved in:
Bibliographic Details
Main Authors: Abdullah, Muhammad, Idres, Moumen, Mohammed Sapardi, Mohd Azan
Format: Article
Language:English
English
English
Published: SEMARAK ILMU SDN BHD 2022
Subjects:
Online Access:http://irep.iium.edu.my/115716/12/Model%20Predictive%20Control%20for%20Regulating%20Fuel%20Cell.pdf
http://irep.iium.edu.my/115716/2/Acceptance.png
http://irep.iium.edu.my/115716/18/Model%20Predictive%20Control%20for%20Regulatingscopus_new.pdf
http://irep.iium.edu.my/115716/
https://semarakilmu.com.my/journals/index.php/fluid_mechanics_thermal_sciences/article/view/422
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Stack temperature and airflow rate are vital control problems for Proton Exchange Membrane fuel cell (PEMFC). Two separate Model Predictive Controllers (MPC) have been employed to regulate these problems. The controllers utilized Laguerre and exponential weight functions to reduce its numerical instability and computational time. The temperature MPC considered delayed and constrained coolant pump voltage as manipulated input and stack temperature as the desired output. While airflow MPC manipulated compressor motor voltage to maintain the desired level of oxygen excess ratio subjected to starvation, surge, and choke constraints. Results showed that both controllers worked well together. The desired temperature and oxygen excess ratio were maintained subjected to all the constraints, even with the presence of external disturbances. This study highlights that MPC manages to handle both control problems without any conflict. Yet, it also proves that MPC can handle a large time delay process.