Solar–EV Logistics Energy Dataset

This dataset contains a rich multi-modal collection of hourly solar-energy, microgrid, battery-storage, and electric-vehicle (EV) charging measurements obtained from the Royal Berkshire Smart Hospital Network (UK). The data is publicly available via the Zenodo Open Research Repository and was originally produced as part of the hospital group’s smart energy and mobility modernization program. It provides an authentic view of how large healthcare facilities operate under both grid-connected and off-grid conditions while integrating renewable energy, storage assets, and electric logistics fleets. The dataset spans the period January 2021 to August 2025, with all systems—rooftop photovoltaic arrays, microgrid controllers, battery storage banks, and EV charging stations—recording synchronized telemetry at hourly intervals. These measurements reflect genuine variability caused by weather dynamics, seasonal changes, transformer loading, EV charging patterns, and operational fluctuations within the logistics and facility-management environment. Additional data is sourced from nearby logistics depots associated with the hospital’s transport and delivery operations, providing a combined view of medical logistics and renewable-energy scheduling. To ensure privacy and regulatory compliance, all personal or identifiable information was removed prior to publication, and only aggregated operational statistics were retained. The dataset therefore offers a secure, ethical, and research-ready foundation for developing forecasting models, demand-response controllers, grid-aware EV-charging strategies, and data-driven optimization frameworks. The dataset includes six major feature groups: Solar and weather conditions: irradiance components, temperature measurements, wind parameters, humidity, cloud cover, and solar geometry. PV and battery system indicators: state-of-charge (SOC), state-of-health (SOH), module temperature, genset activity, and stored-energy behavior. Grid and market signals: day-ahead and real-time price fluctuations, transformer utilization, voltage levels, and frequency deviations. EV charging and fleet dynamics: arrival and departure events, requested energy, charger limits, connector type, route information, payload mass, and charging priorities. Logistics and operational context: hub type, parcel activity, delivery windows, vehicle category, shift type, and facility load demand. Temporal descriptors: hour of day, weekday index, month, seasonal grouping, and holiday flags. These comprehensive attributes make the dataset highly suitable for research in solar forecasting, microgrid optimization, predictive maintenance, EV charging coordination, operational analytics, and intelligent energy-management systems within modern industrial and healthcare environments.