Research On Optimal And Control Of Solid Oxide Fuel Cells Based On Multi-Physics Coupling Model

Solid oxide fuel cell(Solid Oxide Fuel Cell,SOFC)is a power generation device that converts chemical energy in hydrocarbons into electrical energy under medium and high temperature conditions.It has high-efficiency energy conversion,wide fuel range,low emission and low noise and so on,which has important development potential and application prospects in the fields of distributed power generation and cogeneration.Optimizing fuel cell structure and operating conditions to improve cell performance,and implementing thermal management to ensure safe and stable operation of the system are current research hotspots.However,SOFC experiments need high cost,long period,and have few measurable parameters.It is difficult to analyze the mutual coupling mechanism of various internal physical and chemical processes as well as operation mechanism comprehensively through experiments.Therefore,this thesis combines low-cost and highefficiency simulation calculation methods to analyze the internal mechanism of the fuel cells,optimize the cell performance,and conduct thermal management research.The main research contents are as follows:Firstly,based on the fundamental of electrochemistry,fluid mechanics,thermodynamics and solid mechanics,a SOFC electric-flow-thermal-mechanical multiphysics coupling model is established.The relationship between the electrochemical active area and the electrode microstructure in the reaction zone is considered.The actual fuel cell parameters including structure size,material parameters,operating parameters,etc.are also used.With the numerical simulation software named COMSOL,a threedimensional model of the flat anode-supported SOFC is estiablished,which is verified by the voltage-current density and power density-current density curves using the voltage and power experimental data of the same size cell.Secondly,for the power generation characteristics and structural stability of the SOFC,a steady-state multi-objective optimization for output power density and maximum first principal stress is carried out,which can ensure the increase of cell output power while reducing the stress that may cause damage to the cell structure.Then,The influence of operating parameters or structural parameters on the fuel cell’s outputs is analyzed to make sure the optimization objectives.The global sensitivity analysis method of Sobol’ parameters is used for sensitivity analysis to make sure the optimization variables.The mathematical mapping relationship between output power density as well as the maximum first principal stress and parameters are obtained by combing multiphysics simulation(MPS)of SOFC with the artificial neural network algorithm.Based on parameter sensitivity analysis results of the proposed Sobol’ parameters’ sensitivity analysis,sensitive parameters are selected.The fast non-dominated sorting multiobjective optimization algorithm is used.Maximize the output power and minimize the maximum first principal stress are set as the optimization objectives.After obtaining the optimal Pareto solution set,the ideal evaluation method is used to choose the best point.The corresponding optimal parameters are input into the multi-physical field model,and the output values of the MPS model are used to verify the proposed optimization strategy,and the temperature distribution under the optimal condition is analyzed,providing the optimal reference trajectory for the next step of temperature control.Finally,as for the temperature control problem of solid oxide fuel cells,a temperature model predictive control(MPC)strategy is designed based on the Spatiotemporal reduced order model.Through the combination of Karhunen-Loeve decomposition and least squares support vector machine(LS-SVM)algorithm,the order reduction model of SOFC temperature distribution is realized,and the order reduction model is used as the prediction model of MPC.Both optimization index functions and constraints are estiablished to achieve temperature distribution control.Under transient operating conditions,the temperature distribution under the optimial operation conditions obtained in third chapter is applied as the reference temperature.A joint simulation platform conbining COMSOL and MATLAB is constructed to verify the proposed temperature control scheme and realize the tracking control of the reference temperature.