| Currently,due to the over-exploitation and consumption of fossil energy,the development and utilization of renewable energy sources(RES)has become a global consensus to deal with the problems of resource depletion and environmental pollution,and the distributed generation(DG)technology based on RES is developing rapidly.In2020,a double carbon target was proposed in China,highlighting the importance of building a new type of intelligent power system,while microgrids,as relatively independent integrated intelligent power systems,can promote RES consumption in close proximity and local areas,and effectively achieve optimal dispatch and autonomous management of energy.With rapid socio-economic development,the electricity demand of community customers is increasing,which means there is significant potential for optimizing community customer-side energy dispatch and guiding residents to participate in autonomous energy management.At the same time,the frequent two-way interaction between users and the grid also puts forward new requirements for intelligent,personalized and socialized energy management on the user side of community microgrids.This paper focuses on the electricity consumption characteristics on the user side of community microgrids,considers the social cognitive factors in information-physical systems,and elaborates autonomous energy management strategies from both individual households and community groups,mainly accomplishing the following research work.For the autonomous energy management decision of community microgrid,the cyber-physical-social systems(CPSS)is built to realize the autonomous individual energy management on the customer side by analyzing the potential factors influencing the electricity consumption behavior and integrating behavioral science and cognitive psychology.Firstly,the energy consumption behavior of electricity users is modeled to obtain their energy consumption plans.Secondly,a gray correlation analysis is conducted based on physical environment factors to collaboratively filter historical energy use data with the help of the correlation degree of each factor under the multi-factor change trend.Then,the prospect theory is introduced to characterize the special cognitive states and processes influenced by various factors in the energy use decision for the limited rationality property of power users,so as to carry out the collaborative filtering of cognitive psychological factors.Finally,an energy recommendation model for individual customers was built in a laboratory environment to demonstrate that the results show that the energy recommendation plan under the double-layer filtering mechanism can meet the personalized needs of customers,optimize energy scheduling and reduce electricity costs,thus stimulating customers to actively participate in energy management.For the autonomous energy management and regulation of community microgrid group,we automate the decision knowledge based on knowledge graph and build an automated decision model for group energy management.Firstly,based on knowledge mapping technology,we abstract external public data and build community microgrid knowledge map;abstract personalized data of power users and build digital mirror of power users.Then,in the context of market-oriented electricity trading,we break the data barriers between users and the system with the help of knowledge graph in the field of electricity consumption,propose a direct trading mechanism based on bilateral matching model,make group energy use decision based on matching satisfaction index,and improve the satisfaction of users.Finally,simulations are conducted in a laboratory environment to verify the results,which show that the proposed method can promote the consumption of electricity in close proximity and reduce the number of interactions with the large grid while satisfying customer demand and improving customer satisfaction. |
