Abstract: The optimal allocation of distributed generation is an important part of distribution network planning and multiobjective optimization algorithm is the key factor of this link.Most of the algorithms currently used to solve such problems convert multiple goals into single goals,and to some extent,the correlation between the targets is ignored.To solve this problem,an improved artificial searching swarm algorithm is proposed in this paper.Combined with the Pareto theory,parallel search optimization is optimized for each optimization target,and the external archive set is constructed to coordinate the relationship between the targets.The defect that the manual search group algorithm cannot handle the multi-objective optimization problem is solved.The introduction of dynamic parameters is to improve the poor convergence performance and the situation which is easy to fall into premature.Finally,taking the 30-node system as an example,a multi-objective optimization model considering DG investment operation cost,active network loss and load node voltage offset is established.The improved algorithm is adopted to complete the optimization.The effectiveness of the algorithm is verified by comparison with the particle swarm optimization algorithm.