Modeling Of Fuel Cell Multi-field Coupling Mechanism And Its Model Prediction Optimization Algorithm

In order to improve the climate problem,solve the energy crisis and finally realize the goal of "double carbon",developing clean energy and adjusting the energy structure are one of the key themes of sustainable development in the world.Because Proton Exchange Membrane Fuel Cell(PEMFC)has the characteristics of zero emission,high power density and high efficiency,it has quite a wide application prospect.PEMFC system consists of several coupled subsystems,such as air supply system,hydrogen supply system,reactor cathode and thermal management subsystem.Current studies have focused on a single subsystem(such as the air supply system or the thermal management system)and ignored the combined effects of multiple subsystems on the output characteristics of PEMFC systems.In order to improve the reliability of PEMFC system and research innovation,the coupling modeling and lumped control of multiple subsystems are very important.Therefore,this thesis carries out a series of work on mathematical modeling,control strategy development,simulation and experimental verification of PEMFC system:First,taking a high-power water-cooled fuel cell system as a prototype,and fully analyzing the working principle and characteristics of PEMFC system,the whole dynamic modeling work is carried out for the multi-variable coupling system including the air supply subsystem,thermal management subsystem,reactor subsystem,etc.,and the six-order nonlinear simulation model of the fuel cell system is obtained.On this basis,by means of multiple linear regression,each subsystem model is linearized,and finally the discrete linear control model for controller design is derived,and the nonlinear model and the linear control model are simulated and compared,verifying that the linear control model has the dynamic characteristics and variable accuracy close to the original system.To complete the lumped control of PEMFC system,this thesis proposes an optimal Control strategy based on Model Predictive Control(MPC)--Offset-free Explicit Model Predictive Control(Offset-free EMPC),it is designed to solve the model mismatch problem of PEMFC system.On this basis,the operation conditions of each auxiliary equipment are further optimized to make the system work in a safe and reasonable constraint range,and complete the real-time and accurate control of the expected net output power,as well as the optimization of the system efficiency.Finally,the closed-loop simulation of MPC control strategy and Offset-free EMPC control strategy for PEMFC system control were carried out respectively.Compared with MPC,the results show that the Offset-free EMPC has significant advantages in power tracking,system constraints,efficiency optimization and real-time performance.The latter has obvious effect on the performance improvement of PEMFC system.In terms of experimental system construction and verification,this thesis uses a certain type of fuel cell system as the controlled object.Aiming at the problems of temperature control and power tracking in engineering applications,the control strategy based on MPC and Proportional Integral Differential(PID)controllers is designed.Then code generation technology is used.It was downloaded to a Microcontroller Unit(MCU)as the FCU of the system,and finally achieved the temperature control and power tracking targets well in the experiment.