Control-oriented modeling analysis and optimization of planar solid oxide fuel cell system

Because of multi-timescale characteristics and gas transmission delays, the thermal electrical cooperative control of solid oxide fuel cells is a complex and difficult issue. We have used modeling, analysis and optimization of a solid oxide fuel cell system control to guarantee a high efficiency and temperature safety during steady-state and power switch transients. An analysis-based optimization method that applies to discrete optimization problem with constraints is proposed to obtain optimal operating points with a maximum efficiency and to satisfy temperature constraints. Artificial neural-network models that were identified from a validated two-dimensional differential model were used to analyze the temperature constraints and system efficiency. Compared with traversal optimization in literature, the proposed optimization method could save more than 90% of the workload and allow for a more reasonable choice of variable discrete precision. The optimization results were proven feasible by the proposed new system structure and stack inlet condition-based control strategy. The simulation results show that the key problems of the solid oxide fuel cell system control, which include temperature safety, fast load following, fuel starvation and efficiency promotion, could be solved by the proposed strategy.