A perspective cut neighborhood search (PC-NS) method is proposed for unit commitment (UC) with volatile wind power. Firstly, a mixed integer linear programming formulation based on PC is presented for UC with wind power. Then, a neighborhood is built on the solution of UC without considering wind power to find the feasible solution of UC with wind power. After that, based on the feasible solution, another neighborhood is constructed to get high-quality sub-optimal solutions for UC with wind power. At last, the proposed method is tested and the numerical results indicate that the PC-NS is efficient for solving UC with wind power. When compared with other methods, the presented PC-NS can yield better sub-optimal solutions and the running time can be greatly reduced. This work is supported by National Natural Science Foundation of China (No. 71201049 and No. 51407037).