Abstract: In distributed power distribution networks, due to the differences in power supply status among different zones, there is often inconsistency in node operation status, which leads to distortion in the logical relationship between switch equipment status and section fault current, limiting the effectiveness of fault localization in zones. To this end, a fault location method for distribution networks with distributed power sources based on swarm intelligence algorithm is proposed. By constructing a switch function, the status information of the switchgear is transformed into the status information of the feeder section, and the degree of impact on each feeder section is clarified. By combining the information self correction method, the distortion information in the logical relationship between the status of switchgear and section fault current is corrected. Transform the problem of fault location in different zones into a problem of finding the optimal path information, and introduce the weight allocation method to expand the utilization of information for determining the operation status of the distribution network. By iteratively updating pheromones, the algorithm can search for the optimal partition fault location solution and achieve precise localization of partition fault nodes. The test results of the example show that the method has fewer information distortion bits and a section discrimination accuracy of over 99% in fault location of distribution networks with distributed power sources, demonstrating a relatively ideal fault location effect.