System operation reserve requirement keeps going up in the past 3 years to compensate for the variation of wind power. This reduces the efficiency of thermal units by limiting their energy output. Superconducting magnetic energy storage (SMES) as a novel technology was proposed to provide up and down regulation reserve due to its fast response to charge and discharge. However, given the cost and utilization ratio of SMES, an optimal unit commitment (UC) with the integration of SMES is necessary. This paper modifies the traditional UC model into a chance-constrained stochastic problem to realize the optimal schedule objective. To solve this non-convex problem, a Branch/Bound (BB) Technique and Particle Swarm Optimization (PSO) algorithm is introduced, while the initialization of PSO is achieved by the simplex algorithms. Finally, a comparison between the deterministic UC and stochastic UC is given. The result indicates the model in this paper offers independent system operators (ISO) more freedom to balance the system dispatch cost and reliability and it can successfully reduce the SMES costs.