As the front end for users, the stable and safe operation of distribution networks are directly related to the reliable power supplies. However, the large amount of branches in 10 kV distribution networks makes the transmission conditions more complicated. When a fault occurs, it is difficult to determine its specific location, which will affect the power supply. In order to solve this, this paper uses particle swarm optimization (PSO) to improve artificial fish swarm optimization (AFSA). The feasibility of AFSA?PSO algorithm is tested by a case study, where it is applied in a standard distribution network model to verify its practicality and superiority. The results show that the AFSA?PSO algorithm can accurately reflect the single and multiple point failures in distribution networks, and this algorithm is better in both iteration times and convergence speed.