Abstract: Microgrid will become the main mode of power supply for the community in the future, and the development of information technology will make the data of the community microgrid more real-time and transparent, which help improve the demand-side management level of the microgrid. With the cyber-physical-social system（CPSS） architecture, the community microgrid was divided into a three-layer architecture consisting of physical equipment, cyberspace and social roles and a new type of optimization approach was proposed based on guiding cues and parallel control that enables community residents to actively participate in the energy management process of the microgrid. Guiding cues are a series of messages issued by the microgrid operator on the Internet to guide community power users to use renewable energy as much as possible, which make the microgrid more environmentally friendly. Based on the open data in the microgrid, parallel control is a data-driven approach that help the operator reduce the costs of this method and achieve the maximum benefit. The community microgrid operator and power users achieve a win-win situation in this approach, and at the same time the microgrid becomes greener. This paper described the CPSS architecture and operation process of a community microgrid and then expounds on the concept of guiding cues and the application of parallel control, as well as some necessary algorithms. Simulation results that are based on a grid-connected community microgrid demonstrate the feasibility and effectiveness of the approach proposed.