Premium
Temperature Control for Proton Exchange Membrane Fuel Cell based on Current Constraint with Consideration of Limited Cooling Capacity
Author(s) -
Chen F.,
Yu Y.,
Gao Y.
Publication year - 2017
Publication title -
fuel cells
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.485
H-Index - 69
eISSN - 1615-6854
pISSN - 1615-6846
DOI - 10.1002/fuce.201700001
Subject(s) - control theory (sociology) , current (fluid) , proton exchange membrane fuel cell , model predictive control , cooling capacity , constraint (computer aided design) , actuator , temperature control , duty cycle , computer science , power (physics) , automotive engineering , control (management) , engineering , fuel cells , control engineering , mechanical engineering , thermodynamics , electrical engineering , physics , artificial intelligence , chemical engineering
Temperature is one of the key factors to ensure reliable and efficient performance of proton exchange membrane fuel cells. Whereas, due to the limited cooling capacity in automotive application, conventional strategies may fail to remove the excessive heat when the cooling actuator saturates. Hence, three temperature control strategies based on current constraint are developed in this study. Firstly, a dynamic thermal model based on energy conservation is presented. Then a duty ratio split control strategy is proposed to calculate the reduced current from the excess of duty ratio of radiator, yet spike behavior occurs. Hence, a current governor control strategy and a constrained model predictive control strategy are derived to trade off the power demand and cooling capacity, where the maximum load current is calculated by current governor in real time and optimized under the physical constraints of duty ratio and current change rate, respectively. The performances of all the strategies are demonstrated. The current governor control strategy is easy to implement with less computational burden. The constrained model predictive control strategy has fast response and can take into account various constraints explicitly. Both of them have satisfying control performances regardless of limited cooling capacity.