The gray wolf algorithm has the advantages of simple structure, clear concept, easy implementation, good global performance, etc., but it has the disadvantages of slow convergence and weak local search ability in the later period. In order to improve the accuracy and rapidity of the fault location of the distribution network, the grey wolf optimization algorithm is improved. First of all, because the network fault location method can represent the problem as 0~1 integer programming problem, the conversion function is introduced to solve the problem of location update in binary space. Then the dynamic weight strategy is introduced to balance the global search capability and local search capability of the algorithm and accelerate the convergence speed of the algorithm. Finally, a probabilistic perturbation strategy is added to avoid premature local convergence of the algorithm. In this paper, the feasibility and efficiency of the improved binary grey wolf optimization algorithm are verified by an example.