Distribution network planning with DG is a complex combinatorial optimization problem. Along with the development of smart distribution network and fluctuant renewable energy access, it puts forward higher requirements on the efficiency of optimization model. This paper proposed COA-EO algorithm which combined chaos optimization algorithm （COA） and extreme dynamics optimization algorithm （EO） to solve the multi-objective optimization problem. The example verification results show that COA-EO optimization algorithm can take advantage of both COA and EO and manage to avoid the shortcomings, so that it can make ordinary EO escape from local optimal solution, avoid the premature phenomenon of the algorithm, and eventually obtain the globally optimal solution. In addition, in order to get a better multi-objective optimization result, this paper introduced the Pareto optimal solution, and used the proposed COA-EO algorithm to solve the Pareto optimal solution. The calculation results show that the optimization performance of COA-EO algorithm is superior to EO, genetic algorithm （GA）, ant colony optimization （ACO）, ACO-EO algorithm and GA-EO algorithm, which indicates that COA-EO algorithm is effective for distribution network planning with DG.