There is a problem that it is difficult to balance the speed and accuracy of fault location in a multi-branch distribution network with a distributed generator. Thus a fault location model based on an improved bald eagle search algorithm is proposed. First, this paper uses the uniform distribution characteristics of sinusoidal mapping to generate the initial population. It also improves the position update method of the bald eagle through a crossover operator, the non-uniform mutation operator and the somersault foraging strategy, and enhances the mining and exploration capabilities of the algorithm. Secondly, the switching function and objective function considering the distributed generator are constructed. An equivalent hierarchical model of a distribution network with a distributed generator is established by referring to the "black box method", and a location correction mechanism is added to ensure the accuracy of location. Finally, a simulation is carried out in the multi-branch distribution network with a distributed generator. Compared with the traditional bald eagle search algorithm zoning location model, solution speed is improved by 30.5% and accuracy is improved by 1.72% on average. In addition, under different fault locations and different distortion points, the improved bald eagle search algorithm zone location model can quickly and accurately locate the fault zone. This work is supported by the National Natural Science Foundation of China (No. 71263043 and No. 61763040).