Abstract: When the distributed generation is connected to distribution network, the applicability of conventional fault location method decreases. To this end, this paper puts forward a sub-region fault location method of distribution network based on improved grey wolf optimizer and implicit enumeration method. The grey wolf optimization for binary processing and introduces four improvement strategies of dynamic range search, Levy flight, local search and nonlinear convergence factor, which improves the quickness and accuracy of the algorithm. The coding scheme, switching function and target function are established which can meet the demand of distributed power distribution network. By analyzing the fault point and the corresponding mechanism of switching function, the sub-region location model is put forward combing the regional location and orientation of section, and the improved grey wolf optimizer and implicit enumeration method are use for solution. Finally, the IEEE 33 nodes distribution network is taken as an example to execute the simulation test. The consequence reveals that the raised algorithm and sub-region location method can locate the fault area precisely and promptly with high fault tolerance and stability.