PaperlayerView exact revision on arXiv

Source-linked AI summary

Route Optimization of Electric Vehicles based on Dynamic Wireless Charging

Dimitrios Kosmanos, Leandros Maglaras, Michalis Mavrovouniotis, Sotiris Moschoyiannis, Antonios Argyriou, Athanasios Maglaras, Helge Janicke

arXiv:1710.03726v1eess.SYcs.SY

TL;DR

Limited driving range and charging-infrastructure constraints motivate the search for ways to charge EVs on the move. The paper routes EVs to Mobile Energy Disseminators using constraint logic programming and a graph-based shortest-path algorithm. City-condition simulations report about four times better overall travel time with MED-supported charging than with static charging alone.

  • Problem

    Limited driving range and inadequate charging infrastructure constrain EV adoption, while costly road changes and larger batteries are undesirable alternatives.

    Source: §I · p. 1

  • Method

    The paper routes EVs needing charge to mobile MEDs using constraint logic programming and a graph-based shortest-path algorithm.

    Source: §I · p. 2, §II · p. 3

  • Results

    About four times better overall travel time is reported for a MED-supported static charging system than for static charging alone.

    Source: §VI · p. 10

  • Takeaways & Limitations

    Adding a MED to static charging can improve fleet travel time at negligible travel-distance cost, avoiding stops or longer detours to find stationary charging.

    Source: §VI · p. 10

    Evaluation is reported in city conditions; broader assessment with more MEDs and different city areas remains future work.

    Source: §I · p. 2, §VI · p. 10

Abstract

from arXiv · show

One of the barriers to adoption of Electric Vehicles (EVs) is the anxiety around the limited driving range. Recent proposals have explored charging EVs on the move, using dynamic wireless charging which enables power exchange between the vehicle and the grid while the vehicle is moving. In this article, we focus on the intelligent routing of EVs in need of charging so that they can make most efficient use of the so-called {\it Mobile Energy Disseminators} (MEDs) which operates as mobile charging stations. We present a method for routing EVs around MEDs on the road network, which is based on constraint logic programming and optimisation using a graph-based shortest path algorithm. The proposed method exploits Inter-Vehicle (IVC) communications in order to eco-route electric vehicles. We argue that combining modern communications between vehicles and state of the art technologies on energy transfer, the driving range of EVs can be extended without the need for larger batteries or overtly costly infrastructure. We present extensive simulations in city conditions that show the driving range and consequently the overall travel time of electric vehicles is improved with intelligent routing in the presence of MEDs.

I. INTRODUCTION

The introduction frames limited driving range and costly charging infrastructure as barriers to EV adoption, then proposes routing EVs to mobile charging stations. City-condition simulations report improved driving range and overall travel time when MEDs are added to static charging.

I. INTRODUCTION

  • The paper situates connected vehicles within VANETs, where vehicles communicate with one another and with roadside units.Source: §I · p. 1
  • Limited driving range and a lack of charging infrastructure hinder EV adoption, while building infrastructure is costly and slow and larger batteries raise environmental concerns.Source: §I · p. 1
  • Buses and trucks can act as Mobile Energy Disseminators (MEDs), letting EVs meet scheduled routes and charge wirelessly while moving.The buses return to fixed stations after their round trips.Source: §I · p. 2
  • The main contribution is EV rerouting with constraint logic programming and a graph-based shortest-path algorithm, considering multiple criteria and priorities.Source: §I · p. 2
  • About four times better overall travel time is reported for dynamic charging with MEDs than for static charging alone.The introduction also reports improved driving range in city-condition simulations.Source: §I · p. 2

II. RELATED WORK

Earlier work examines static charging, battery swapping, eco-routing, and dynamic charging from fixed sources, with trade-offs including safety concerns and infrastructure cost. This paper focuses on mobile charging stations that meet EVs on the road.

II. RELATED WORK

  • Earlier approaches include static stations, swappable batteries, eco-routing, and dynamic charging from fixed sources, with safety and infrastructure costs among the trade-offs.Source: §II · p. 2
  • Prior dynamic charging work considered fixed sources, while moving charging stations serving EVs also in motion remained unaddressed.The paper identifies this as a gap in efforts to reduce range anxiety and improve EV reliability.Source: §II · p. 2
  • Telewatt reuses public-lighting power but raises magnetic-flux health concerns; an ORNL inductive-coupling prototype reached nearly 90% efficiency at 3 kW, while grid coupling may cause pulsations.The passage links those pulsations to possible battery-service-life deterioration and lower grid power quality.Source: §II · p. 2
  • The proposed approach uses buses or trucks on predefined routes as MEDs, rather than relying on vehicle-to-vehicle charging schemes.The passage contrasts this with pitfalls associated with static charging stations.Source: §II · p. 3
  • EVs attach to one or more MEDs until they can reach their destination or the nearest static station, without significant road-network changes or the stated health hazards of other proposals.Source: §II · p. 3

III. DYNAMIC CHARGING AND MOBILE ENERGY DISSEMINATORS

Dynamic wireless charging combines inter-vehicle communication with inductive power transfer, using buses or trucks as mobile energy sources. The section describes the charging mechanism, the MED architecture, and its practical safety considerations.

A. Energy transfer via IPT

  • A 10-minute IPT charge supplies 3–8 kWh, equivalent to about 9–23 miles, at a 20–50 kW charging rate.The passage says this distance is 30–78% of drivers’ average daily travel; at 50 kW it estimates 78% for typical urban American drivers.Source: A. Energy transfer via IPT · p. 3
  • A 30 amp public wired station provides 3.7 miles per 10-minute charge, while about an hour is needed for 22 miles.Table I is identified as a comparison of miles per 10-minute charge for electric cars.Source: A. Energy transfer via IPT · p. 3, Table I · p. 3

B. EVs and MEDs in a VANET architecture

  • Buses and trucks serve as MEDs: scheduled routes let EVs meet them, communicate, form a platoon, and recharge through IPT while moving.The section describes buses for urban routes and trucks mainly for highways; Figure 2 depicts bus charging, MED–EV communication, and EV recharging from a bus.Source: B. EVs and MEDs in a VANET architecture · p. 3, §III · p. 3, Figure 2 · p. 4
  • The proposed coil-based wireless transfer is described as exceeding 90% efficiency when coils are very close.The coils may be plain or conical, and shielding is available on both vehicles.Source: B. EVs and MEDs in a VANET architecture · p. 4
  • The coils are activated only when a compatible EV is over the charging pad; foreign-object detection is proposed for future deployment to avoid hazards.The passage identifies possible effects of strong magnetic fields on living organisms and metal objects near charging pads as safety concerns.Source: B. EVs and MEDs in a VANET architecture · p. 4

C. Communication among entities

The system broadcasts charging-station and MED information, then models EV routing as a travel-time minimization problem constrained by energy feasibility and charging availability.

C. Communication among entities

  • MEDs and static charging stations periodically broadcast CAMs with vehicle identity, location, scheduled trip, charging capability, energy, and waiting time.Charging capability is the energy a mobile station can provide without jeopardizing its own needs; vehicles relay these messages.Source: C. Communication among entities · p. 4

A. Problem Formulation: Constrained Shortest Path

  • MEDs visit predefined cyclic routes, and EVs may attach at a MED point where charging rate exceeds consumption, subject to availability and waiting time.MED and static-station demands may be accepted or rejected to minimize vehicle routes or distribute energy.Source: A. Problem Formulation: Constrained… · p. 4
  • EV routing is modeled on a directed weighted graph, minimizing driving, charging, and waiting time across routes for multiple vehicles.Arcs carry travel-time and energy costs, while each EV has battery capacity, start and destination points, and route and charging decision variables.Source: A. Problem Formulation: Constrained… · pp. 4–5
  • Feasible routes must preserve nonnegative energy and reach a recharging point; multiple shortest paths and charging queues mean one EV’s route can affect others.The constraints also bound energy by battery capacity and require enough energy to reach the next point or a recharging location.Source: A. Problem Formulation: Constrained… · p. 5

B. Solution Method

The method uses Dijkstra’s algorithm to build energy-feasible routes, adding charging stops when a direct route fails the energy constraint and selecting among reachable energy points.

B. Solution Method

  • A Dijkstra route to the destination may be infeasible because it does not satisfy the EV’s energy constraint.The method first checks whether the direct route has enough energy to reach the destination.Source: B. Solution Method · p. 5
  • When direct travel is infeasible, the method evaluates reachable static or MED energy points and selects the feasible point with the lowest total travel time.The time criterion includes driving, waiting, and charging; points the EV cannot reach with its current energy are discarded.Source: B. Solution Method · pp. 5–6
  • After selecting an energy point, the method joins shortest paths to and from it, repeating the search if the onward path remains energy-infeasible.The resulting route and its travel time are output for the EV.Source: B. Solution Method · p. 6

V. EVALUATION

The evaluation simulates dynamic wireless charging in the city of Erlangen.

V. EVALUATION

  • The evaluation uses city-condition simulations conducted in Erlangen to assess dynamic wireless charging.The passage specifies the simulation location but gives no outcome or further setup details.Source: §V · p. 6

A. Evaluation Setup

The city simulations compare a static charging station (SCS) with a system combining an SCS and a mobile energy disseminator (MED). EVs start from side roads, with variable initial energy and communication conditions.

A. Evaluation Setup

  • The Erlangen setup compares an SCS at a fixed roadside point with a MED following a bus route, using side roads as EV starting points.The map marks the MED route in yellow, the SCS in green, and the EV starting points with brown arrows.Source: A. Evaluation Setup · p. 6, Figure 3 · p. 7
  • Table III is identified as the source of evaluation parameters for the simulations.Source: Table III · p. 7
  • The simulations include between 0 and 100 EVs, each entering with starting energy drawn uniformly from 1–6 kWh.Source: A. Evaluation Setup · p. 7
  • Communication uses only the ad-hoc network; a 130–300 m transmission range can leave some EVs unconnected and excluded from charging.The stated range follows from antenna sensitivity settings; building attenuation can reduce SINR below 10 dB.Source: A. Evaluation Setup · p. 7

B. Implementation of the Dynamic Charging System

The implementation estimates travel and charging times for SCS and MED options, then uses shortest-path routing to choose a charging plan. Its energy model accounts for road-segment consumption and induction while an EV follows a MED.

B. Implementation of the Dynamic Charging System

  • The vehicle minimizes travel time from its starting point to destination, comparing an SCS route with a MED route using Dijkstra’s algorithm for path segments.The SCS estimate includes travel, charging, and queue waiting; the MED estimate includes meeting, waiting, following the MED, and onward travel.Source: B. Implementation of the Dynamic… · p. 7
  • Vehicles periodically receive station wait information, while MED position and bookings inform meeting-point and waiting-time estimates.A booked MED segment or a missed arrival cycle can require waiting for another cycle; the model specifies no upper limit on this wait.Source: B. Implementation of the Dynamic… · p. 7
  • Road-segment energy cost is calculated from the vehicle’s power consumption and time spent on the segment.The consumption model relates steady-speed power to rolling friction and air resistance; acceleration and deceleration are ignored.Source: B. Implementation of the Dynamic… · p. 8
  • When an EV travels on a MED route, induction adds energy in proportion to the time the EV and MED remain connected.Contact time depends on the meeting point, segment length, and MED availability; induced energy is modeled using contact time, induction coefficient, and MED power.Source: B. Implementation of the Dynamic… · p. 8

C. Starting Energy vs Power Consumption levels

The simulations vary how many EVs lack enough starting energy to reach their destinations, using three levels of charging need. The proportion of such drivers changes with the vehicles’ energy needs.

C. Starting Energy vs Power Consumption levels

D. SCS vs. SCS + MED

The SCS-plus-MED system has lower travel time than the SCS-only system across the evaluated charging-need levels and anxious-driver counts. Its reported advantage is largest at the highest charging-need level, although MED bookings can add waiting.

D. SCS vs. SCS + MED

  • Travel time increases for both charging systems as vehicle charging needs increase.The comparison covers three charging-system scenarios, with travel-time subfigures corresponding to the charging-need levels.Source: D. SCS vs. SCS + MED · p. 8
  • Travel time for SCS + 1 MED is about 2 times better at Level 1, about 3 times better at Level 2, and about 4 times smaller at Level 3 than for SCS.Source: D. SCS vs. SCS + MED · p. 8
  • SCS + MED has less travel time than SCS for all evaluated anxious-driver counts from 0 to 100 and all charging-need levels.At Level 3, the difference diminishes above 50 anxious drivers because vehicles wait through MED cycles due to preceding bookings.Source: D. SCS vs. SCS + MED · p. 8

E. SCS + MED system evaluation

The evaluation compares MED waiting time with SCS queue time and tracks charging choices as anxious-driver numbers rise. At Level 3, greater MED use and switching between stations accompany a wide travel-time difference between SCS and SCS+1 MED.

E. SCS + MED system evaluation

  • As anxious-driver numbers rise, more EVs select the MED, while MED waiting time and SCS queue time fluctuate as drivers switch between them.At Level 3, above 80 anxious drivers, the increased MED waiting time prompts some drivers to choose the SCS, and the changing SCS queue can reverse that choice.Source: E. SCS + MED system evaluation · pp. 8–9
  • Figure 4 covers travel time across all energy re-charging levels, while the evaluation compares the SCS+1 MED system with SCS.The cited discussion describes a reduced travel-time difference between the systems as anxious-driver numbers increase.Source: Figure 4 · p. 9, E. SCS + MED system evaluation · p. 9
  • When the MED serves more EVs at Level 3, MED waiting time and SCS queue time rise less steeply; greater MED use and switching accompany the wide travel-time difference.The passage attributes the Level 3 travel-time difference to greater MED use and more interchanges between MED waiting time and SCS queue time.Source: E. SCS + MED system evaluation · pp. 9–10, Figure 5 · p. 10

F. Cost Benefit Analysis

The cost-benefit discussion outlines possible revenue and entrepreneurial opportunities around MEDs. The conclusions report improved fleet travel time with MEDs, at negligible travel-distance cost, and identify broader evaluation as future work.

F. Cost Benefit Analysis

  • Utilities and governments may support converting trucks and buses into MEDs through revenue sharing or tax incentives.The passage also describes potential revenue-sharing partnerships between utilities and MED owners.Source: F. Cost Benefit Analysis · p. 10
  • MED conversion is estimated to cost around 26,000 for a bus or truck and about 1,500 to add the technology to a passenger vehicle.The discussion also identifies software for platooning, appointments, and billing, and magnetic subsystems as entrepreneurial opportunities.Source: F. Cost Benefit Analysis · p. 10

VI. CONCLUSIONS

  • The proposed approach uses city buses on scheduled routes for dynamic charging, rather than vehicle-to-vehicle charging, and does not require road-infrastructure changes.The paper presents the approach as complementing, rather than excluding, larger batteries.Source: §VI · p. 10
  • Inductive power transfer enables efficient, real-time energy exchange in which vehicles can play an active role.The conclusions connect this energy exchange to extending vehicle travel range.Source: §VI · p. 10
  • Using one MED alongside an SCS improves fleet travel time about four times compared with using only an SCS, with negligible travel-distance cost.Vehicles otherwise might stop or take longer reroutes to reach a stationary charging station.Source: §VI · p. 10
  • Future evaluation will examine more MEDs with existing SCSs, different city areas, and a wider diversity of evaluation parameters.Source: §VI · p. 10
Loading 1710.03726v1…