A numerical optimisation method, previously applied to a large industrial cogeneration plant, has been adapted for a small-scale unit that must be driven by the heat demand of a medium-size building located in the north of Portugal. The unit is based on a micro-gas turbine and includes an internal pre-heater (typical of these types of small-scale units) and a water heater. The mathematical model yields a non-linear objective function, subject to physical constraints of system operation, and defined as the maximisation of the annual worth of the CHP system. Seven decision variables were selected for the optimization algorithm, including performance of internal gas turbine components and the electrical and thermal powers. The latter is linked to the estimated thermal load of the building via the number of working hours. A purchase cost equation was used for each major plant component and takes into account size and performance variables. The optimal solution was obtained for a feed-in-tariff and natural gas price reference scenario, and shows higher standards than models currently available in the market A sensitivity analysis was carried out for the two variables that significantly affect the annual worth. Regarding the physical aspects of the major system components, the analysis revealed that the most relevant and susceptible variables are the compressor pressure ratio, the internal pre-heater effectiveness and the turbine inlet temperature. (C) 2012 Elsevier Ltd. All rights reserved.