PNNL

Abstract

Big data plays a crucial role in modern power system planning, operation, and control. With the advancement in AI/ML, it has become a valuable tool in power industry. This chapter presents how correlated data streams could be used to develop ML models for predicting the energy requirements and flexibility associated with the assets towards designing and evaluating coordination mechanism. It discusses an application of big data in demand prediction through a microgrid. The chapter presents an urban academic campus, represented as a microgrid, contains photovoltaic generators, energy storage and commercial buildings that can participate in consensus-based transactive energy systems. It elaborates on ML modeling of commercial building connected to distribution feeder in a co-simulation platform. ML model uses DNN to predict power consumption patterns based on historical data. Furthermore, the application of the ML model is demonstrated through a consensus-based transactive mechanism enabling operators to coordinate the flexibility among the resources and thereby achieve efficient operation.