| Nowadays,the new energy industry,especially the hydrogen energy and fuel cell industry,is developing rapidly.Fuel cells are widely used in various fields due to their advantageous characteristics such as high power density,high energy conversion efficiency,cleanness and pollution-free.The development of fuel cell technology in the transportation field is particularly rapid.The fuel cell engine system with multiple subsystems is the core component of the fuel cell vehicles.In order to achieve the optimal overall performance,modeling and control strategy research should be carried out on each subsystem.Focusing on the anode hydrogen supply subsystem,researches on the hydrogen recirculation and purge strategy of fuel cell engines are presented in this thesis.The work is summarized as follows:1.The hydrogen circulation pump test platform was designed and built,the existing fuel cell engine platform was improved and the characterization experiments on the hydrogen recirculation and purge loop were performed on the platforms.The experiment design and the result analysis were based on how the recirculation ratio and purge strategy affected the anode hydrogen loop.The experimental results show that the hydrogen recirculation has a good effect on maintaining the stability of the anode hydrogen pressure,improving the hydrophobic performance of the system,and maintaining the voltage and its uniformity.Especially the effect is better when the recirculation ratio is larger.Purge strategy has a significant impact on the anode hydrogen pressure,the output capability of system and the hydrogen utilization.At last,the optimal recirculation ratio and purge strategy under the operating conditions were obtained.2.The modeling research on hydrogen recirculation and purge process was carried out,and by combining mechanism model and neural network model,the control-oriented model of anode loop of fuel cell engine was established.The neural network model was trained with the experimental data of the hydrogen circulation pump,and the flow characteristic model of the pump based on neural network regression prediction was obtained.It was verified that the model had good generalization ability.Meanwhile,the mechanism model was established based on the working principle and mechanism equations of each component.Finally,combining the above-mentioned models,the control-oriented model of anode loop of fuel cell engine was obtained.3.The study on the effect of the purge strategy on the nitrogen concentration in the anode channel was completed with the obtained model,and focused on two control objects,the corresponding control strategy was designed and realized.The lumped model of the anode hydrogen loop system was used to analyze the influence of the purge strategy on the nitrogen concentration of the anode channel.This thesis established a control-oriented model in terms of the pressure of anode channel and the stoichiometric ratio of hydrogen.A PI controller and a fuzzy-PI controller were designed based on the model,and the control strategy was verified by simulation analysis on MATLAB.The simulation results show that the PI controller has a good control effect on the pressure of anode channel,and the pressure was remained within the target pressure ± 1.5 k Pa,but for the stoichiometric ratio of hydrogen,the effect of fuzzy-PI control is better than the effect of PI control. |