Aggregator Zone Selection for EV Smart Controls Based-On ML Clustering of Grid Strength, Distance, and Charging Homogeneity

Smart electric vehicle (EV) charging control methods from a central utility hub often require communication infrastructure over a large service area of electric power distribution systems with a large number of nodes. Industry standards such as Open Charge Point Protocol (OCPP) 2.1 have evolved to include topologies for local controllers to the individual chargers, i.e. EV aggregator zones. A machine learning (ML) application of k-means clustering is proposed to establish zones for coordination of EV charging based on grid strength and EV owner decision-making to charge per day. Very large-scale distribution networks including the IEEE 123 and 8500 benchmark feeders are characterized by their distances from the substation and line reactance to resistance ratios (X/R). Then, a sensitivity study is performed with six different statistical distributions of “daily homogeneity”, i.e. overlapping EV owner decision to charge that day between all homes on each line. Following increased voltage violations in scenarios with high homogeneity, a case study under the uniform statistical distribution shows how the development of the EV zone selection process based on stochastic inputs such as decision homogeneity, line $X / R$ ratio, and house quantity may intelligently identify groups of EVs for localized controls.