Fuel Cell Engine Control Based On Reduced Dimension Model

Facing the global challenges of growing energy demand,environmental crisis,increasing population,etc.,alternatives of clean energy utilization are getting more and more attention,amongst which the Proton Exchange Membrane Fuel Cell(PEMFC)is promising with advantages of high efficiency,zero emission,low noise,short start time,etc.,and is one of the next generation technologies for automotive powertrains.In the area of PEMFC engine control,lumped parameter models are often adopted in the design of state estimation and control algorithm.However,the PEMFC is a complex system with coupling of “gas-water-electrical-thermal-force” issues.Control-oriented lumped parameter models are too simplified,neglecting the uneven distribution of the parameters in the fuel cell.Phenomenum eg.local reactant starvation are not quantitatively described.In order to improve the durability and operational statiliby of the fuel cell,it is necessary to focus on the inhomogenities of the parameters,especially the minimum oxygen concentration inside the fuel cell.On the basis of the numerical model,this paper proposes a reduced dimension model for PEMFC,which combines the advantages of numerical model and lumped parameter model.It can reflect the maximum and minimum value of important parameters in the internal of fuel cells.With small computational efforts,this model is suitable for real-time control.For model verification,polarization curve test results are not enough for verifying the description of the internal parameter distribution.In this paper,an on-line gas component sampling and analysis experiment system is designed for the large active area PEMFC single cell.It samples the gas components at different positions inside the flow field in real time during the operation of the fuel cell.Test results are used to calibrate and verify the reduced dimension model.In this paper,the membrane water content state estimation and minimum oxygen concentration estimation algorithms are designed based on the combination of the reduced dimension model and exhaust pipe two-phase flow pressure drop model.The reduced dimension model is further simplified and integrated into the PEMFC engine system model,based on which hydrogen injection control and air supply control algorithms are designed.The control of cathode exhaust recirculation is experimentally explored.Finally the algorithms for each subsystem are integrated and experimentally verified on a fuel cell engine test bench.The innovation of this paper includes the following aspects: A control-oriented reduced dimension model is proposed which can describe the inhomogeneity of component distribution within the fuel cell;an online gas sampling experiment system for fuel cell with large active area and multiple gas channels is designed and the test results are utilized for model calibration and validation;hydrogen injection control,air supply control and cathode exhaust gas recirculation control on a dual-recirculation fuel cell system is realized based on the reduced dimension model.The work in this paper introduces the description of the parameter distribution into the framework of controloriented lumped parameter model,and makes a novel exploration for the modeling and control methodologies of the PEMFC,which has certain reference value to researchers in the field of PEMFC.