This paper presents a methodology for finding the optimal output power from a PEM fuel cell power plant (FCPP). The FCPP is used to supply power to a small micro-grid community. The technique used is based on evolutionary programming (EP) to find a near-optimal solution of the problem. The method incorporates the Hill-Climbing technique (HCT) to maintain feasibility during the solution process. An economic model of the FCPP is used. The model considers the production cost of energy and the possibility of selling and buying electrical energy from the local grid. In addition, the model takes into account the thermal energy output from the FCPP and the thermal energy requirement for the micro-grid community. The results obtained are compared against a solution based on genetic algorithms. Results are encouraging and indicate viability of the proposed technique. (C) 2004 Elsevier B.V. All rights reserved.