For overcoming the disadvantages of wide range wind power perturbation and model parameter uncertainties, a novel maximum power point tracking (MPPT) controller for vertical axis wind power generation system is designed using back-stepping control method. The MPPT controller adopts wind turbine’s optimum rotational speed tracking algorithm, and the wind turbine’s dynamic aerodynamic torque is estimated by a nonlinear disturbance observer (NDO). According to the back-stepping controller design approach, virtual control signals and Lyapunov functions are constructed step by step. By employing Lyapunov stability theorem, back-stepping control law for each subsystem is worked out, thus, the global asymptotical stability for the overall system that all virtual control signals involved are guaranteed. The control system is simulated using Matlab/Simulink software, which is compared with conventional PID control method. The simulation results verify the correctness and effectiveness of the back-stepping controller. This work is supported by National Natural Science Foundation of China (No. 61004053 and No. 61273151), Natural Science Foundation of Jiangsu Province (No. BK20141238), Qing Lan Project of Jiangsu Province, and Graduate Student Scientific Research and Innovation Project of Jiangsu Province (No. CXLX13_681).