With the rapid growth of wind power penetration, system operators are required to enforce grid codes that implement frequency regulation of wind farms. This paper presents a distributed cooperation framework for a doubly fed induction generator-based wind farm to participate in primary frequency regulation, including imitated inertia and droop characteristics similar to conventional plants. Compared with centralized schemes, the control efficiency is significantly improved. To realize fast distributed coordination and optimal power distribution among wind turbines (WTs), a distributed Newton method is developed, which only requires WTs to exchange limited information with their neighbors over a sparse communication network, and has a super-linear convergence rate. By introducing an index of state of energy, this method can adequately exploit the kinetic energy of all WTs without loss of security. The simulation results indicate that the method exhibits satisfying dynamic performance and reliability, and the system frequency can be stabilized faster when the wind farm controlled by the proposed method participates in frequency control.