The molten carbonate fuel cell (MCFC) is a promising device for stationary power and heat supply. In this work, the temperature control in MCFC stack is studied. The operating temperature of the stack is an important controlled variable, which impacts the performance of the MCFC due to thermal cycling. In order to improve the generating performance of the MCFC, prolong its life and guarantee safety, credibility and low degradation of the fuel cell, the temperature of MCFC must be controlled efficiently. Based on analyzing the temperature dynamic model of MCFC, a model-based variable structure control (VSC) controller is designed for MCFC in this paper. The optimal parameters of VSC are selected by genetic algorithm (GA). Numerical results demonstrate the effectiveness and advantages of this approach. (c) 2006 Elsevier B.V. All rights reserved.