An Architecture Based on Digital Twins for Smart Power Distribution System

Constructing an effective architecture based on digital twins using advanced artificial intelligent technologies remains a key challenge in smart power distribution system. Despite recent advances in important domains such as device health maintenance and manufacturing process, the conventional architecture does not offer a satisfactory solution for rapidly providing data models, a set of analytics or algorithms, and knowledge. In this work, we employ ideas from distributed network based on digital twins and from recent advances that augment artificial intelligent with big data. Our architecture provides a paradigm that uses a distributed network makes computer programming, software, and data spread out across more than one terminal, but communicate complex messages through terminal nodes, and are dependent upon each other, obviating the pressure from master station. We demonstrate the usefulness of digital twins that are used to track the performance of devices such as transformer during normal operation to detect possible defects and fix them before they result in failure. We also conclude that using digital twin technology will understand, predict, and optimize performance in order to achieve improved business outcomes, and apply machine learning and knowledge to predict the future.