Minimum-time MILP-based EV charging scheduling for prosumers

• This work considers complementary charging strategies for prosumers who own an electric vehicle (EV), in addition to rooftop PV and BESS. • Besides the least-cost solution, an original MILP model is proposed to minimize EV charging time for a given cost overrun. • For this purpose, two sets of binary variables are introduced, allowing the on/off plug-in status of the EV to be easily modelled. • Simulations show notable charging time cuts with ∼1% yearly cost overrun vs. the least-cost solution. • The proposed model can be easily extended to workplaces or public charging stations with multiple charging points. With increasing penetration of electric vehicles (EV) and behind-the-meter renewables, strategies that balance cost and user convenience (i.e., EV availability) are essential. This work proposes two sequential optimization models for EV charging in prosumer households equipped with rooftop PV and BESS: first, a linear programming (LP) model for least-cost scheduling; second, an original mixed-integer linear programming (MILP) model for minimum-time charging within an acceptable cost overrun. The two models are rigorously formulated and validated with real-world data. Both charging strategies (least-cost and minimum-time) are simulated throughout a year for prosumers with different combinations of assets. Then, the total annual costs and average charging times are compared with those arising when the owner simply leaves the EV charging while it is plugged in. Simulation results show that the proposed minimum-time charging scheduling, while being computationally efficient, is particularly effective in the presence of both PV and BESS, providing significantly shorter charging times than the least-cost approach, whereas the actual cost overrun is very small (around 1% on average).