Fully exploring the value of user data to identify user energy quality needs and promoting differentiated value-added services are of great value for improving user satisfaction and the competitiveness of power grid companies. Firstly, based on multidimensional user data, user profile theory to establish a multi-dimensional high- quality power user profile label system to identify users’power quality needs is introduced. Secondly, K-means clustering algorithm is used to profile high- quality power user groups,achieve user group division, quantify user power quality requirements based on prospect theory, and based on this differentiated value-added service package design is carried out. Finally, the effectiveness of the proposed method is verified through case analysis.