Distributed genenration in distribution network has some influence on the system voltage, network loss etc. In the planning of distributed generation in distribution network, A multi-objective Optimization Model is proposed in this paper, which including the minimum investment cost, minimum power loss and the quality of voltage is the best. Using PSAT to build numerical model for power flow calculation. And in order to overcome the premature phenomena and convergence problem of genetic algorithms, a new improved self-adaptive genetic algorithm is employed to optimize the location and sizing of distributed genenration in distribution network. The effectiveness and feasibility for improving the quality of voltage and reducing the power loss are finally verified by the simulation of the IEEE 33 bus system example．