With the evolution of smart distribution grids, industrial parks are progressively advancing towards intelligence. Addressing the challenges in smart industrial parks where the response capability of demand-side users is underutilized, leading to suboptimal energy use and economic inefficiencies, an energy optimization strategy for smart industrial parks that considers user satisfaction is proposed. Firstly, the load characteristics of users within the park are extracted, and the responsive resources in the park are assessed based on users' electricity consumption behavior, which allows an analysis of the potential for demand response from industrial park users, as well as an evaluation of their willingness to respond. Secondly, the distributed photovoltaic resources and user-side demand response resources are introduced to establish an industrial parks' energy optimization model with the optimization objective of minimizing the sum of demand response cost, grid electricity purchasing cost, carbon emission cost, and equipment operation and maintenance costs. In addition, the non-dominated sorting genetic algorithm is employed to solve this model. Lastly, the effectiveness of the proposed approach is verified through a case study of an industrial park in a city in North China.