Measured-PV Converter-Loss-Aware Model Predictive Smart Charging and Vehicle-to-Grid Control for Electric Vehicle Fleets Under Forecast Uncertainty
Variable solar availability and bidirectional-converter losses complicate the coordination of electric-vehicle (EV) charging and vehicle-to-grid (V2G) services. This study computationally assesses a receding-horizon model predictive control (MPC) strategy using irradiance measurements collected at the Microgrid Laboratory of Universidad de Cuenca, Ecuador. The laboratory dataset provides minute-resolution irradiance records; photovoltaic (PV) power is estimated using a 35 kWp reference system rather than measured at an operating charging site. A 24 h simulation with two bidirectional EV chargers compares uncontrolled, solar-smart, rule-based V2G and MPC-V2G scheduling under the same scenario-based facility load and EV sessions. In the deterministic case, MPC reduces peak grid import from 24.51 to 12.77 kW (47.9%) and grid import from 184.36 to 147.88 kWh (19.8%) relative to uncontrolled charging. Thirty paired stochastic runs achieve 100% modeled departure-target satisfaction and a lower mean peak with MPC. The controller uses nominal conversion efficiencies; a load-dependent map quantifies conversion losses after dispatch. Peak mitigation entails increased battery throughput, and the reported results are computational rather than hardware or field validation.
Authors
- Edisson Villa‐Ávila (ORCID: https://orcid.org/0000-0002-2766-5913)
- Paul Arévalo-Cordero (ORCID: https://orcid.org/0000-0002-6721-1326)
- Michael Villa-Ávila
- Emilia Sempértegui-Moscoso (ORCID: https://orcid.org/0000-0002-1178-1928)
- Esteban Albornoz-Vintimilla (ORCID: https://orcid.org/0009-0002-4811-2020)
Institutions
- University of Cuenca (EC)
- Universidad de Jaén (ES)
Publication Details
- Journal
- Vehicles
- Published
- 2026-11-17
- DOI
- https://doi.org/10.3390/vehicles8110302
- Primary Topic
- Electric Vehicles and Infrastructure
- Type
- article
- Field-Weighted Citation Impact
- 0.00