Abstract: The existing intelligent optimization algorithms have many problems in solving the fault location problem of distribution network with distributed generation(DG), such as many iterations, slow convergence speed, easy to fall into the local optimal solution, poor fault tolerance, etc. A cuckoo search algorithm with a new movement strategy(NMS-CS) is proposed. The feeder terminal unit(FTU) state coding with directional is introduced when considering the large capacity DG access to the distribution network, and the switch function and objective function adapted to the DG switching are established. Based on the traditional cuckoo algorithm, the algorithm improves the ability of global optimization in the early stage and the accuracy of the late search by improving the length and proposing new direction parameters. The potential search space parameters are proposed to determine the calculation direction, and then improve the convergence speed of the algorithm, and realize the rapid fault location of active distribution network. Simulation results show that the NMS-CS algorithm can achieve accurate location in single-point, multi-point and information distortion fault location, and has fewer iterations and shorter convergence time compared with other optimization algorithms. Effectively shorting the troubleshooting time.