Optimised fast-charging service to allow long-distance trips with electric vehicles.

Mots-clés:

Résumé:

Gerpisa proposal – Optimised fast-charging service to allow long-distance trips with electric vehicles.

 Purpose

The Covid crisis might lead to an increase in private car use to avoid health concerns, but environmental needs to decrease GES emissions remain a critical issue. The development of electric vehicles appears to be one of the solutions to mitigate those problems. However, even if improvements have been done in the electric vehicle field, there are still some drawbacks to EVs’ acceptability, such as limited range and lack of fast-charging infrastructure that impede electric mobility on long-distance (>500 km). This paper proposes a solution based on an optimised fast-charging service that make long-distance trips possible with electric vehicles.

State of the art:

• According to [1], it is more cost-efficient to invest in fast-charging infrastructure (>22 kWh) than in battery with a higher range;
• The acceptability of EV can be improved by optimising the charging stations’ location and size [2], [3];
• It appears that dynamically scheduling EVs’ charging plans on the highway can also enhance EVs’ acceptability by reducing the traveling time [4].

Methodology

To face the actual lack of charging infrastructure for long-distance trip that may cause long waiting times at charging stations, this paper choses to focus on a charging coordination service between all-electric vehicles (Battery electric vehicle – BEV) on a highway.

We used the framework developed by Jean Hassler [5] further evaluate the interest of a communication model. The framework consists in:

• A highway (>500 km) represented by its entrances/exits and fast-charging stations (CSs). The CSs are described by their position on the highway, their charging power, and their number of sockets (see Figure 1 attached);
• A fleet of EVs traveling on the highway. Each EV has its intrinsic parameters (battery capacity, maximum charging power, consumption…) and trip characteristics (entrance and exit number, entry time, State of Charge – SoC– at entrance…).
• A model of charging schedule communication between EVs and CSs. The framework simulates offline the flow of EVs in real-time (the time step can be set as little as possible). Every time step, each EV communicates their charging plan (stations where they plan to charge, amount of energy they want to store, arrival time at each station…) to the stations. Then, the stations compute the waiting time for each EV and send the result back to them. According to the result, the EV adapts its charging plan if necessary.

The charging schedule of an EV is set in order to minimize its traveling time by choosing a combination of charging stops and the amount of energy stored that minimizes its global waiting and/or charging time. The framework also enables us to consider the charging price at each station when setting the EV’s strategy, but we only focus on optimizing the trip duration for this study.

To better evaluate the interest of communication between EVs, we suggest comparing the communication scenario with two other scenarios: one without communication and another with communication and reservation possibility.

The scenarios can be described as follow:

• No communication: the reference scenario. The EVs do not communicate during the whole trip and calculate their charging plan to minimize their traveling time, only knowing the position and the charging power of the CSs. They do not know the strategy of other vehicles.
• Communication: as previously described in the framework, the EVs communicate their charging plan every time step, and determine their charging plan to minimize their traveling time in accordance with other EVs’ plans. The rule at a station is first arrived, first served.
• Communication and reservation: the EVs communicate their charging plan every time step, but they also book the charging time slots that will better minimize their trip duration. This solution mainly benefits slower EVs since they can book time slots before faster EVs’ arrival to a station and thus, reduce their waiting time.

Expected results

The study tested the different scenarios on a fleet of 100 EVs . The average trip time gain compared to the scenario without communication was 16 % for the communication scenario and 6% for the reservation scenario.

We notice that even if the communication reduces the average trip time for the 100-EV fleet, some EVs from the fleet see their traveling time increase when they communicate their charging plan in comparison with when they do not.

The paper will test the different scenarios on a significant number of fleets to quantify the quality of the charging service with communication. 

 Practical and theoretical implications

As communication can increase the traveling time for some EVs (or for the whole fleet), test on a high number of fleets would probably help dimensioning the fast-charging infrastructure (adding/removing charging points) to avoid that increase in traveling time.

References

[1] S. Funke, P. Plötz et M. Wietschel, «Invest in fast-charging infrastructure or in longer battery ranges? A cost efficiency comparison for Germany,» Applied Energy, n° 1235, p. 888–899, 2019.
[2] Y. He, K. M. Kockelman et K. A. Perrine, «optimal locations of U.S. fast charging stations for long-distance trips by battery electric vehicles,» Journal of Cleaner Production, n° 1214, pp. 452-461, 2019.
[3] W. Kong, Y. Luo, G. Feng, K. Li et H. Peng, «Optimal location planning method of fast charging station for electric vehicles considering operators, drivers, vehicles, traffic flow and power grid,» Energy 186, 2019.
[4] V. d. Razo et H.-A. Jacobsen, «Smart charging schedules for highaway travel with electric vehicles,» IEEE, vol. 2, n° 12, pp. 160-173, 2016.
[5] J. Hassler, Z. Dimitrova, M. Petit et P. Dessante, «Service for optimization of charging stations selection for electric vehicles users during long distances drives: time-cost tradeoff,» IOP Conference Ser.: Mater. Sci. Eng., vol. 1002, 2020.

  940 words