This study aims to propose a solving approach for the thermal unit commitment (UC) problem using the mutated particle swarm optimization (MPSO) combined with a novel encoding scheme. Unlike traditional straightforward encoding arrangements, the proposed encoding method applies the load demand and spinning reserve constraints to construct a small searching space, and then put the constraints of minimum up and down-time into the encoding structure so as to shorten the searching time effectively. This novel coding scheme could effectively prevent obtaining infeasible solutions through the application of stochastic search methods, thereby dramatically improving search efficiency and solution quality. Many nonlinear characteristics of power generators, and their operational constraints, such as minimum up and down-time, spinning reserve, generation limitations, ramp rate limits, prohibited operating zones, transmission loss, and nonlinear cost functions were all considered for practical operation. The effectiveness and feasibility of the proposed approach were demonstrated by three system case studies and compared with previous literature in terms of solution quality. The simulation results reveal that the proposed approach was capable of efficiently determining higher quality solutions in resolving the thermal unit commitment problems. (C) 2010 Elsevier Ltd. All rights reserved.