Integration of Renewable Energy of Solar PV and Wind for the Optimal Charging and Discharging of Plug-in Electric Vehicles

• Developed business-use EV driving, charging and discharging pattern generator • Built an optimization routine for EV demand and renewable energy supply of 1) solar PV only, 2) wind only, and 3) hybrid combined solar PV and wind • Compared different EV charging and discharging strategies • Compared the performance of solar, wind and hybrid systems in satisfying energy requirements of EVs used for business runs. Battery electric vehicles (BEVs) can have a significant impact on the load profile of the distribution grid due to the driving and charging patterns of the owners/operators. At the same time, the drive towards net zero-energy communities and the development of next generation higher performance batteries will offset that effect by making vehicle-to-grid (V2G) technology an attractive solution to reducing the impact to the distribution grid in the near future. This paper proposes a smart charging/discharging algorithm that will control when an EV can be charged and discharged and by how much, to maximize renewable energy usage and reduce grid peak demand. The optimization algorithm utilizes a local controller and a centralized controller to optimize the balance of energy of the whole community. The driving and charging patterns used in the simulation were synthesized from an experiment in Australia made up of a commercial fleet of BEVs. Three technologies were simulated and compared: solar photovoltaics (PV), wind and hybrid system of combined wind and PV. The simulation results show that uncontrolled charging of commercial BEVs changed the load profile, starting the peak demand at 8 am to 5 pm. Numerical simulation results showed that controlled charging of the BEVs resulted in lower peak but no significant shift in the peak was observed due to the commercial application of the BEVs. Despite the intuitive advantage of energy diversity, the hybrid PV-wind system underperformed during summer due to lower wind availability and the reduced capacity allocation to each source. PV-only systems produced more consistent midday generation, aligning well with commercial EV demand. Additionally, hybrid configurations led to higher curtailment and lower effective utilization under fixed capacity constraints. Similar behavior has been observed in recent work by Huang et al. (2022), indicating that seasonal resource alignment is critical to hybrid system success. To improve readability, several overly long paragraphs in the results and assumptions sections have been split into shorter, thematically cohesive units. This improves clarity without altering the content. While the results indicate robust performance under the modeled community conditions, the practical deployment of such algorithms must address real-world factors like charger interoperability, user compliance, communication latency, and the economic viability of bidirectional chargers. Scalability challenges include grid synchronization for large fleets and regulatory approval for dynamic discharging. Simulation comparisons between PV, wind, and hybrid systems reveal distinct behavior. PV systems closely match the business-hour demand of commercial EVs, offering the highest grid savings. Wind energy, which is more stochastic and less available during summer, has had a limited impact. The hybrid system offered time-diversified supply but, when capacity was halved for balancing annual output, it underperformed compared to PV alone in the summer. The main findings showed that uncontrolled charging sharply increased grid demand during morning hours, while controlled schemes significantly reduced peaks. However, due to the rigid nature of commercial operations, shifting the peak load was difficult, limiting the algorithm’s potential for full-tempered load redistribution. Optimization of EV charging and discharging significantly mitigates grid stress by distributing the load more evenly across the day. Controlled charging delayed peak demand from 8 AM to 11 AM and reduced overall grid draw by up to 7% compared to uncontrolled charging. Despite these improvements, accurately aligning EV charging with realistic community loads posed a challenge. Due to the lack of granular empirical driving data, especially on weekends or during off-peak hours, synthetic patterns were assumed using MCMC. This introduced modeling inaccuracies, leading to some under- or overestimation of EV demand, and influenced the ability of the optimization scheme to fully flatten grid demand or ensure renewable absorption. The comparative analysis of PV, wind, and hybrid systems showed that PV alone aligned best with the diurnal charging behavior of commercial EVs, producing the highest grid energy savings. Wind energy was high in variability and often did not coincide with demand, particularly in summer. The hybrid system smoothed energy availability over the day but offered no distinct advantage over PV alone in summer simulations. Uncontrolled charging caused pronounced peaks, whereas controlled schemes lowered but did not shift these peaks due to the fixed commercial operation schedules. Simulation results showed that optimized charging reduced energy import by up to 7%, while controlled discharging increased charging demand due to subsequent SOC replenishment. However, accurately associating EV charging with realistic community load profiles remains a challenge, owing to limited high-resolution empirical data on driver behavior, especially on weekends. This mismatch affects optimization outcomes and necessitates improvements in data acquisition and modeling.