Active distribution network (ADN) mode is a technology to optimize the usage of distributed energy resources. Considering the generation benefit cost of distributed generation and its impact on the power grid when assessing the injected solutions of distributed generation into the active distribution network, and considering the reduction of emission targets of the distributed generation, this paper builds a low-carbon optimization objective function model. To avoid human subjectivity too strong when assigning the index weight, according to the 1-9 scale theory, it forms a judgment matrix through the consistency check and gets the reasonable weighting factor. Secondly, to reflect the trends of each indicator and have a certain degree of distinction, it uses membership function for each index value membership, and weights the effectiveness of each membership value corresponding to the weight. Finally, the optimization studies of IEEE14 system is carried out based on GA algorithm, the results show that the proposed low-carbon optimization model is reasonable and effective.