The increasing cost of carbon based fuels pushes the electricity generation market towards more efficient electricity generation technologies. In this context, fuel cells offer the opportunity to increase efficiency and lower CO2 and NOx emissions, in particular with co-generation units. The complexity, as well as the number and different combinations of components involved with a SOFC stack make their analysis impossible without dedicated tools. The simultaneous consideration of economic and thermodynamic considerations is a key challenge for engineers and decision makers. The present work contributes to the development of design tools for fuel cells and energy systems. A detailed model of fuel cell stack and a model of a complete system were developed. Based on a commercial CFD code, additional routines were implemented to model the electro-chemical behavior of a fuel cell as well as the reforming reaction schemes. The model is able to predict cell behavior under various fuel feeds. The detailed 3-D model of stack repeat unit allowed to identify weaknesses of a first stack geometry. With the combination of experimental observations and modeling activities, specific problems of the existing concept were identified and a new geometry designed, together with the corresponding stack concept. A new stack design is presented with successful experiments up to 1 kWe. Besides the stack, a co-generation fuel cell system is composed of multiple components such as heat exchangers, reformer, recycle loops... The number, the type and size of those components influence the system costs but also its efficiencies (thermal and electrical). A multi-objective thermo-economic optimization methodology was defined, allowing to optimize at the same time the configuration and operating parameters of a system. This method was applied to two sizes of co-generators : a 1 kW unit and a 30 kW unit. HTceramix SA (Yverdon, Switzerland) is developing part of a co-generation unit. The work indicates the potential improvement of their prototype. The 1 kW unit was optimized in a first phase on thermal and electrical efficiencies, and in a second phase on electrical efficiency and installed costs. Those two results indicate the influence of the objective function selection on the results. The high operating temperature of Solid Oxide Fuel Cells (SOFC) offers good opportunity for coupling with a gas turbine. A simple concept of pressurized SOFC-µGT was studied and optimized, for an net electrical output of 30 kW. The simple layout described in this work allows present technical feasibility of the concept. Moreover, the costs of electricity, lower than 0.20$/kWh, makes it economically viable solutions. The methodology and the tools developed allow the engineer to explore the solution space through optimal configurations, in order to select those fitting with its priorities and technical constraints.