An improved quantum-behaved particle swarm optimization based on Lagrange multiplier was given to solve the constraint optimization problem in this paper. It is very difficult to decide the penalty function properly in the common methods of Sequential Unconstrained Minimization Technique (SUMT) such as interior or exterior point method. In order to overcome this problem, the Lagrange multiplier was combined in the Quantum-behaved Particle Swarm Optimization (QPSO) method to handle the constraint optimization problem. In order to validate the accuracy and convergence speed of the new method, the simulation results of five benchmark functions were compared with two other methods which dealing with the constraint functions differently. The comparison results demonstrate the accuracy and robustness of the proposed method.