The integration of electric vehicles (EVs) into distribution networks poses challenges for peak demand management and voltage regulation. This paper presents a hierarchical framework combining deep reinforcement learning (DRL) and model predictive control (MPC) for voltage-aware EV charging coordination. A Proximal Policy Optimization (PPO) agent learns strategic power allocation policies from historical data, while a lower-level MPC layer enforces network constraints using linearized power flow (LinDistFlow) equations. Validation on 873,286 smart meter measurements from 25 Pecan Street households over a full calendar year demonstrates that the proposed method achieves a 6.8% additional peak demand reduction (95% CI: [5.2%, 8.4%]) compared to standalone MPC, while maintaining zero voltage violations under LinDistFlow-enforced constraints, within the validated operating regime. PV self-consumption reaches 72.1% versus 61.2% for MPC-only (p < 0.001, Cohen’s d = 2.00). Comparison against six baselines confirms consistent improvements, though gains are incremental rather than transformative. Computational analysis demonstrates within-interval computational feasibility (211±50 ms) and scalability up to 50 EVs. The results suggest that hierarchical learning–optimization architectures offer modest but statistically significant benefits for network-constrained EV charging coordination.

The gradual substitution of internal combustion engines with electric vehicles (EVs) gives a lot of flexibility potential to the power systems; however, this potential is not fully utilized because there are no utility-deployable, data-driven approaches to the measurement of EV charging flexibility based on already accessible infrastructure of utilities. This paper presents a data-driven framework for constructing flexibility envelopes for pilot-scale residential EV fleet aggregations using smart-meter data without the need for EV-level submetering, direct user input, or vehicle-specific data. The proposed framework integrates non-invasive load monitoring (NILM) and multi-feature clustering to identify EV charging sessions using aggregate whole-home smart-meter readings and to identify prevailing charging behavior patterns. Based on these behaviorally defined clusters, time-dependent power and energy flexibility sheaths are built, therefore defining practicable charging paths, adjustable energy capacity, and peak-reduction potential both at the cluster level and at the fleet level. One year of residential smart-meter data from ten EV-equipped households of the Pecan Street Dataport is used to validate. The findings indicate that EV charging episodes could be determined from 15-minute smart-meter records with high recall (0.83-0.91) and moderate-to-high precision (0.77-0.85), and that the charging behavior will naturally fall into a few explainable patterns with specific flexibility properties. At the fleet scale, the derived envelopes show that about 60% of total EV charging energy could be shifted over time, temporarily allowing up to a 50% of coincident peak charging demand, which allows meeting historical user charging demands. Unlike forecasting-based techniques, the current technique provides operationally significant flexibility limits that can be effectively fitted in demand-response, aggregation, and grid-optimization applications. The results highlight the feasibility of using the available smart-meter infrastructure to describe the flexibility of the EV fleet in residential aggregation settings on a pilot basis.

This paper proposes an advanced metering infrastructure (AMI)-informed hierarchical energy management framework for coordinated operation of electric vehicles (EVs), photovoltaic (PV) systems, and battery energy storage systems (BESS) in campus microgrids. The proposed two-layer architecture integrates a soft actor–critic (SAC) deep reinforcement learning (DRL) agent in the upper layer with a receding horizon model predictive control (MPC) optimizer in the lower layer. The key novelty is an AMI-to-control pipeline that transforms historical 15 min smart-meter measurements into operational flexibility features and embeds them into a hierarchical SAC–MPC architecture, where the DRL layer provides adaptive coordination and the MPC layer enforces grid, storage, and EV-service constraints. The proposed framework using the real-world Pecan Street data (15 min resolution) of 73 homes across Austin, Texas and California (2014–2019) achieves a 53.1% cost reduction and a 25.7% peak demand reduction when compared with uncontrolled charging, and the proposed framework outperforms MPC-only (50.9%), DRL-only (−5.2%), and rule-based (5.1%) baselines. The statistically significant contributions of network-aware constraints, demand-response activation, and predictive look-ahead horizon are statistically significant (n = 10 independent runs) contributions (p = 0.001). The state representation informed by AMI offers directional cost improvement (+8.4%, p = 0.055) with 11% faster convergence of training. The zero network constraint violation is observed in all evaluation scenarios and the average MPC solve time is around 150 ms, which is much less than the 15 min sampling period. Sensitivity analyses show that the hierarchical DRL–MPC architecture remains computationally feasible across EV penetration, seasonal, and forecast-uncertainty scenarios. However, BESS provided no net economic benefit under the evaluated energy-only TOU tariff, increasing weekly cost by $15.25 and peak grid demand by 14.2 kW. Break-even analysis indicates that demand charges of approximately $9.9/kW per month are required for BESS to become cost-effective in the proxy system, highlighting that storage value depends strongly on tariff design and peak-demand objective formulation.