In order to cope with increasing energy demand and environmental pollution, the development of distributed generation (DG) and demand response (DR) has become important in promoting the development of the global energy system. This paper proposes an electricity consumption behavior portrait method based on improved k-means clustering analysis and optimal feature collection to achieve the demarcation of demand response schemes for differentiated user electricity consumption characteristics. Then a distribution network expansion planning model with a differentiated demand response scheme is constructed. Annual comprehensive cost and DG utilization are added as optimization objectives in the decision-making objectives. Normalized normal constraint (NNC) is used to obtain the Pareto optimal solution set. Finally, the validity of the demand response scheme definition method and the extended planning model is verified on the IEEE33-bus system. This work is supported by the Project of Sichuan Science and Technology Department (No. 2021YFSY0053).