Unit commitment problem is one of the major problems in power system operation and control. The determination of time intervals at which a particular unit to be on and off, satisfying various constraints is a multi constrained complex optimization problem. In this paper we have used Particle swarm optimization technique which is population based global searching optimization technique is applied to solve unit commitment problem, for optimum unit commitment schedule. With the application of PSO algorithm it is easy to update the lagrangian multipliers, useful to sub divide the main problem in to a number of sub problem. Single unit dynamic programming is used to solve for each unit so that the total cost can be minimized over a scheduling period of time.