Optimized Pricing and Rewards in a Vehicle-to-Building Scheme

Introduction

A functional building equipped with a solar photovoltaic system (PV) on its rooftop or parking canopy is one of the most popular applications of the MG concept in urban areas. On the one hand, green energy from the PV systems benefits both building and utility by reducing the building owner’s electricity bill, while destressing the aging distribution network. On the other hand, handling the intermittent nature of PV generation requires energy storage systems (ESS) which are not usually cost-efficient solutions because of the high battery investment. Meanwhile, electric vehicle (EV) fleets are growing and will become a dominant transportation means in the near future. Equipped with a decent battery storage system and parked mostly for 90% of the time makes this a promising energy storage solution for PVMGs.

Fig 1. EVCS integrated in a PVMG

In our study, a vehicle-to-building (V2B) creates a balanced environment in PVMG for the building which indirectly improves grid performance. This encourages self-consumption of the PV generation produced on-site, reduces demand on the distribution network, and increases reliability and stability of both the PVMG and the utility. The highest challenge in adopting V2G (or V2B) is the commitment from EV users. Their interest lies in taking part in the energy trade when their economic benefit is significant. In this paper, we aim to minimize operating costs for PVMG by proposing a dynamic charging price and V2G reward scheme, thus encouraging EV users to sell stored energy from their vehicle back to the PVMG. Two main points are expected to be achieved through this study:

– Formulating a flexible pricing and reward scheme from a PVMG perspective, thus encouraging participation of EVs via V2G to minimize operating cost with a significant PV penetration level.

– Proposing a novel but simple combined particle swarm optimization/mixed-integer linear programming (PSO-MILP) algorithm to schedule optimal dynamic pricing, thereby avoiding a computational complexity.

– Investigating the effectiveness of the pricing/reward scheme in different parking patterns.

Proposed Solution

Fig 2. Combined PSO-MILP algorithm

To find the optimal charging price and reward V2G – minimizing operating costs for PVMG while ensuring EV user satisfaction – we propose a novel yet simple and effective combination between a heuristic particle swarm optimization (PSO) and mixed-integer linear programming (MILP) algorithm. First, a PSO is formulated with the ability to manage constraints in searching for an equilibrium point between charging price and rewarding V2G. Then, the problem can be formulated as a MILP optimization problem, which is effectively solved using the Matlab optimization toolbox to find an optimal set of charging/discharging decisions for EVs at the parking station over a horizon of 24 hours. The PSO-MILP algorithm is illustrated in Fig 2.

The numerical simulations investigate three different scenarios of the EV fleets arrival/departure period to demonstrate the effective impacts of integrating EV into the PVMGs. As a result, the proposed method outperforms when scheduling dynamic pricing and optimal energy trajectory for 24 hours ahead of the 1-hour interval. The simulation results are shown in Fig 3 – Fig 5.

Fig 3. Case No.1 – Peak generation/off-peak demand support – day-time parking

Fig 4. Case No.2 – Peak generation/peak demand support – day-time parking

Fig 5. Case No.3 – Peak demand support – night-time parking

Conclusion

Simulation results show that the flexible pricing/reward scheme encourages EV owners to share their EV battery capacity to support PVMG as a virtual energy storage system. Without the support of EV, all the PV surplus power must be sold back to the grid, up to about 250 kW; this amount can be lower than 150 kW with the participation of V2G. The reduction in reverse power flow potentially avoids network congestion during peak PV generation periods.

The impacts of different EV parking patterns are derived from the simulation results. Since PV generation is only available during the daytime, the parking patterns in commercial and office building is best suited to support PVMG. In comparison, residential parking behavior shows limited benefits for PV regulation purposes.

In future works, there are many opportunities to improve the operational performance of EV-integrated PV-based MG:

– First, the complexity of the optimization problem increases exponentially with the size of EV fleets, which may not be solved in a reasonable duration of time. Therefore, a proper aggregate model of EV fleets is required.

– Second, the stochastic behavior of the aggregate EV fleet instead of individual EVs is worth studying.

– Third, under the highly stochastic behavior of EV fleet and PV generation, the online and real-time EV availability assessment is essential to ensure stability and reliability of the PVMG operation.

Additional Information

For more information on this research, please read the following conference paper:

V. Q. Ngo, K. Khoa Nguyen and K. Al-Haddad, “Optimal Dynamic Pricing and Rewarding for Electric Vehicle Charging Scheme in High Penetration Photovoltaic Microgrid,” IECON 2020 The 46th Annual Conference of the IEEE Industrial Electronics Society, 2020, pp. 3697-3702.