Cellular Computational Network for Distributed Power Flow Inferencing in Electric Distribution Systems

The modern power system is undergoing a rapid transformation from centralized generation to distributed generation. The distributed generation consists of a large number of small distributed energy resource (DER) based generators with stochastic power output connected at different geographic locations throughout the feeder. Therefore, the operation of the future distribution grid requires efficient, scalable, and robust techniques for system analysis and control. This study presents a data driven approach to solve the electric power flow problem based on the framework known as the Cellular Computational Network (CCN). It is scalable since diakoptics resolves the system to computational cells, which are then connected together to form the full system. The results for the IEEE 4 Bus system shows that this approach can generate accurate power flow solutions.