Modeling And Control Strategy Study Of The Solid Oxide Fuel Cell-Gas Turbine System With Ejector Recirculation

It has generated a lot of realization on the significance of the sustainable development with the developing energy-hungry civilization and environment issues.The establishment of an efficient,clean,economic energy structure has become the consensus of the world.Furthermore,the stringent regulations of restricting emissions,such as CO₂ and NO_X,have accelerated the search for a clean energy.Fuel cell-gas turbine hybrid system is an attractive energy conversion technology due to its high efficiency combined with low emissions.However,the thermodynamic performance of the hybrid system is not understood completely due to its highly nonlinear,multi-variable,and strongly coupled characteristics.Furthermore,the research of its control strategy is still in the early stage.There are several serious problems,such as low efficiency and safety issues,during variable load operation.The research of modeling method,thermodynamic characteristic and control strategy design can provide necessary technical reserves and guide the practical applications for the hybrid system.Therefore,this thesis focuses on these key technologies for a novel SOFC-GT hybrid system with anode and cathode ejectors.First of all,the ejector technology is introduced to perform the anode and cathode recirculation loops due to its lower maintenance costs and higher reliability.A novel SOFC-GT hybrid system with anode and cathode ejectors is designed.A performance simulation model of the novel hybrid system is established via modular modeling method.The simulation errors of the SOFC and ejector models are both less than 1%.The design performance,part-load performance,and dynamic performance are investigated respectively.Firstly,the performance of the hybrid system is analyzed at the design condition,and the system output power and efficiency are 328kW and 61.88%respectively.Then,considering the system efficiency and safety under part-load operating conditions,four different operation strategies are proposed and their part-load operation characteristics are compared.The results show that the part-load performance of case 4 is better than the other cases.Finally,the dynamic performance of the novel hybrid system is investigated and summarized to guide the selection of the control system parameters.Moreover,two further researches are performed which are effect of the nonlinear characteristic in the fuel system and control strategy of the novel hybrid system.The object of analyzing the effect of the fuel system is to informationize the influence of the servo actuator nonlinear characteristics on the hybrid system.Firstly,a fuel system model considering the nonlinear characteristic is added to the SOFC-GT system model,and the influence of dead zone-relay nonlinear characteristic is investigated.The hybrid system will be oscillating when the width of the dead zone-relay nonlinear characteristic is too small.Then a two dimensional fuzzy compensator is proposed and designed to restrain the oscillating phenomenon.Finally,a servo actuator experiment platform is designed and built to validate the effectiveness of the designed fuzzy compensator.The object of researching the control strategy is to design an effective control system for the novel hybrid system.In the SOFC thermal management system,a concept of monitoring the temperature difference between anode channel and cathode channel is proposed,and the anode inlet temperature and cathode inlet temperature are both controlled by designing two control loops.The control loops including output power,fuel utilization,turbine inlet temperature,rotational speed are also performed in the control strategy.The control performance shows that both anode and cathode inlet temperature loops are necessary.It can not only keep high efficiency but also avoid latent dangers of the novel hybrid system at a part-load condition.Finally,combining with the advantages of NARX in nonlinear system identification and LS-SVR in nonlinear function regression,a NARX modeling method with LS-SVR is proposed to describe the dynamic performance of the novel SOFC-GT system.A training framework is also proposed to ensure the model accuracy.The fuel utilization dynamic model is developed according to above data-driven modeling methodology,and the data driven model is effectively applied to researching the performance of fuel utilization control loop.