The unit commitment problem of a large power system with the popularity of large‐scale wind power is complex and difficult to solve. Vector ordinal optimization (VOO) is a new method to solve multi‐objective optimization problems with complex calculation burdens which can obtain a satisfactory solution at a high enough probability. Our objective is to develop a new technique for optimal design with stochastic wind power. In this paper, we select coal consumption, power purchase costs and SO2emissions as the objective functions. Based on the scenario method, the stochastic wind power generation is taken into account. A real provincial power system with a large‐scale wind farm is taken as a numerical example, in which hydropower, nuclear power, biomass energy, gas turbines, thermal power and other complex power structures are considered. The results show that the VOO based method proposed in this paper is effective in solving large‐scale multi‐objective unit commitment problems considering stochastic wind power. © 2017 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.