Research Of Performance Evaluation,health Management And Control For SOFC System

With the increasing demand for new energy in modern society,solid oxide fuel cell(SOFC)system has been increasingly applied in the field of new energy vehicles,distributed power generation and military equipment due to the advantages of high co-generation efficiency,no mechanical movement,low noise,low emission and favorable environmental protection.It is necessary to design a method that can identify the evolution of system performance in a timely and accurate manner,and formulate corresponding control strategies for achieving the stable,high-efficiency and long-life operation of the SOFC system.Starting from building the SOFC system multi-mode(including normal mode,transition mode and failure mode)mechanism model,combining full-condition system experiment and historical operating data,this dissertation studies the performance evaluation and health control strategy of SOFC system and key components based on the k W-scale SOFC independent generation system as the research object.The main research contents are as follows:First of all,building the SOFC system model which can reflect the multimodal response of four key hotbox components(reformer,heat exchanger,afterburner and SOFC stack)based on description of the multi-modal dynamic behavior of the SOFC system.Four typical faults at the unit level in the SOFC system are embedded in the model: performance degradation of the reformer,breakage of the heat exchanger,imbalance of the gas flow rate in the afterburner and degradation of the electrical characteristics in the SOFC stack.The modeling of transition mode and failure mode is further completed,and the evolution mechanism of system performance is obtained through simulation and analysis.Secondly,based on the qualitative evaluation of the mechanism model,the threedimensional simulation analysis of the SOFC system high temperature heat source — the afterburner is carried out to obtain the temperature distribution of the afterburner which cannot be accurately measured because of the small number of sensors.On this basis,a fullcondition experiment(including start-stop,standby and long-term operation)is carried out,and combined with data-driven method(empirical mode decomposition and Elman neural network)to realize the forecast of the system operating state.Then,on the basis of the overall system operation state forecast,a fault detection method based on principal component analysis and exponentially weighted moving average control chart is proposed for the gas supply problem caused by the heat exchanger and the reformer fault,as well as the thermoelectric buffeting fault of the afterburner and SOFC stack.The effectiveness of the proposed method is verified by online testing of multi-class units,and the corresponding fault repair scheme is put forward for the key components such as heat exchanger,reformer,afterburner and SOFC stack.Finally,this dissertation analyzes the typical thermoelectric buffeting faults at the SOFC system level,and identifies the natural gas SOFC system model.At the same time,the system model is used for dynamic analysis,and it is found that the thermoelectric buffeting of the system is highly correlated with steam to carbon ratio and other parameters.Based on this,a predictive control algorithm with reference output is designed to meet the system temperature safety by controlling the parameters such as fuel,air flow rate and so on.In addition,the repair scheme for thermoelectric buffeting is put forward from the perspective of SOFC system transformation.For the purpose of the stable,efficient,and long-life operation of the SOFC system and solving the unit-level and system-level typical key issues such as the safety of main components and frequent buffeting of thermoelectric characteristics in the operation of the SOFC k W-scale independent generation system,this dissertation proposes a method based on the mechanism model,supplemented by data driving,and in accordance with the research and strategy design process of “first the overall system operation state forecast,then the system components separately evaluation,finally the system problem health control”,which effectively realizes the performance evaluation and health control of the system and its key components,and broadens the thinking for the further development of the SOFC system.