Unit commitment has been solved with many techniques viz., genetic algorithms evolution ary programming, simulated annealing, optimization and tab along with the combination of dynamic programming. This paper proposes Particle swarm Optimization combined with Lagrange Relaxation method (LR) for solving Unit Commitment (UC). The results from the test samples are compared with those obtained by Particle swarm optimization for solving unit commitment, Genetic algorithm and LR. The shortcoming of branch-and-bound is the exponential growth in the execution time with the size of UC problem. The integer and mixed integer methods adopt linear programming technique to solve and check for an integer solution. These methods have only been applied to small UC problems and have required major assumptions which limit the solution space. Lagrange relaxation for UC problem was superior to dynamic programming due to its faster computational time. However, it suffers from numerical convergence and solution quality problems. Furthermore, solution quality of LR depends on the method to update Lagrange multipliers. This paper proposes a new hybrid method for solving UC problem. The proposed method is developed in such way that a particle swarm optimization technique is applied to update Lagrange multipliers and improves the performance of LR method. To illustrate the effective of the proposed method, it is tested and compared to the conventional LR [69], GA [69], and HPSO [79] on 4 units test system and 10 units test system, respectively.