Construction of Knowledge Graph for Terminal Electricity Consumption Policies and Simulation of Typical Application Scenarios

For developing and making decisions on terminal electricity consumption policies, one must systematically understand and predict the dynamic changes of terminal electricity consumption. Building a knowledge graph terminal electricity consumption policies and applying simulation and reasoning is an important way and can provide a scientific basis. In the research project, we aimed to use the knowledge graph to understand the terminal electricity consumption policy. We used information such as policy, regulations, market data, regulators, and terminal electric consumption behavior to construct a knowledge graph in the field of terminal electrical consumption. In the construction of the knowledge graph, we hope to systematically integrate different types of data, establish the relationship and knowledge nodes between policies, data, and perform a comparison of the performance of simulation and reasoning based on three algorithms, which are RNN, CRF, SVM. Through the simulation test, the RNN can achieve better simulation and prediction, and comparison performance. RNN has achieved a significantly improved accuracy compared to CRF and SVM, up to 94–98%. RNN is appropriate for the performance evaluation of policy effectiveness in reducing energy consumption, energy consumption trajectory forecasting, as well as the identification and mitigation of risk that is potentially encountered. If this study is successfully implemented or widely applied, it can be used to support the realization of sustainable energy development projects, as well as a decision making system.