PaperlayerView exact revision on arXiv

Source-linked AI summary

Flexible Energy Management Protocol for Cooperative EV-to-EV Charging

Rongqing Zhang, Xiang Cheng, Liuqing Yang

arXiv:1802.06059v1eess.SY

TL;DR

The paper addresses the challenge of flexible EV power transfer by introducing cooperative V2V charging and a centralized energy management protocol with multiple matching algorithms. The protocol models EV utilities and matching relationships to optimize social welfare while also producing stable consumer- or provider-optimal matchings, and simulations report improved EV utilities and reduced network energy consumption.

  • Problem

    Effective and efficient online V2V charging strategies for flexible EV power transfer remain an open issue.

  • Method

    The paper combines cooperative V2V charging, utility definitions, bipartite-graph modeling, and max-weight and stable V2V matching algorithms.

  • Results

    Simulations indicate improved EV utilities and network social welfare, reduced network energy consumption, and stable consumer- or provider-optimal matchings from the stable algorithms.

  • Takeaways & Limitations

    Cooperative V2V charging supports flexible and smart EV charging/discharging, with a trade-off between max-weight social welfare and individual-rationality-based stability.

Abstract

from arXiv · show

In this paper, we investigate flexible power transfer among electric vehicles (EVs) from a cooperative perspective in an EV system. First, the concept of cooperative EV-to-EV (V2V) charging is introduced, which enables active cooperation via charging/discharging operations between EVs as energy consumers and EVs as energy providers. Then, based on the cooperative V2V charging concept, a flexible energy management protocol with different V2V matching algorithms is proposed, which can help the EVs achieve more flexible and smarter charging/discharging behaviors. In the proposed energy management protocol, we define the utilities of the EVs based on the cost and profit through cooperative V2V charging and employ the bipartite graph to model the charging/discharging cooperation between EVs as energy consumers and EVs as energy providers. Based on the constructed bipartite graph, a max-weight V2V matching algorithm is proposed in order to optimize the network social welfare. Moreover, taking individual rationality into consideration, we further introduce the stable matching concepts and propose two stable V2V matching algorithms, which can yield the EV-consumer-optimal and EV-provider-optimal stable V2V matchings, respectively. Simulation results verify the efficiency of our proposed cooperative V2V charging based energy management protocol in improving the EV utilities and the network social welfare as well as reducing the energy consumption of the EVs.

I. INTRODUCTION

The paper addresses grid-loading concerns from EV charging by introducing cooperative V2V charging and a flexible energy management protocol for smarter power transfer among EVs.

  • I. INTRODUCTION: EVs create both a significant potential grid load and a flexible resource for reducing peak load through controlled charging and discharging.Uncontrolled charging may overload the grid during peak hours, while smart scheduling can shape the load profile.
  • I. INTRODUCTION: The protocol targets more flexible charging plans while addressing the open problem of effective online V2V charging strategies.The IoE setting allows moving EVs to exchange information about loads and prices and share power when needed.
  • A. System Description: The system includes EVs, smart houses, charging stations, power and communication infrastructures, and a data control center.EVs are categorized as energy consumers, energy providers, or nonparticipants in current energy trading.
  • B. Cooperative V2V Charging: Cooperative V2V charging enables direct power transfer between EVs acting as energy consumers and energy providers.The concept supports active cooperation through charging and discharging operations.
  • cooperative V2V charging based energy management protocol.: The protocol uses real-time EV information and data-center coordination to make charging and discharging decisions through V2V matching.EVs submit requests or surplus-power information through mobile or onboard applications supported by IoV and ITS.

A. EV Utility Definition

The paper defines EV utilities from the costs and profits of cooperative energy trading, including travel to a selected trading location or charging station.

  • 1) EV as an Energy Consumer:: Consumer utility compares cooperative V2V charging costs with the cost of charging at the nearest charging station.The formulation accounts for trading price, requested power, travel distance, and the consumer’s moving energy cost.
  • A. EV Utility Definition: The V2V trading price can lie between the grid’s buying and selling prices, benefiting both consumers and providers when trading is worthwhile.Consumers may buy below the grid selling price, while providers may sell above the grid buying price.
  • 1) EV as an Energy Consumer:: Consumer travel costs are evaluated for reaching the selected parking lot for V2V transfer and the nearest charging station for conventional charging.Driving distance and moving energy cost are explicitly included in the comparison.
  • 1) EV as an Energy Consumer:: The charging-station baseline values consumer charging cost using the grid electricity selling price.The nearest charging station is used as the reference location.
  • 2) EV as an Energy Provider:: Provider utility captures profit from selling surplus power through V2V trading after accounting for transfer-related costs.The provider-side formulation includes travel and other energy-trading costs.

2) EV as an Energy Provider:

Provider utility accounts for the costs of traveling to a V2V meeting point, time valuation, transfer speed, and battery degradation when selling surplus energy.

  • 2) EV as an Energy Provider:: The provider model assumes a current surplus-power amount for energy trading and uses transfer efficiency and trading-cost parameters.The cited formulation defines the current trading price, original unit cost, transfer efficiency, and surplus-power-related terms.
  • 2) EV as an Energy Provider:: Battery degradation is represented as an amortized cost associated with each V2V power transfer.The degradation model considers battery replacement cost and wear-out through energy processing.
  • 2) EV as an Energy Provider:: Provider-side costs include travel energy, travel time, and the cost of performing V2V power transfer.The model uses provider energy cost per kilometer, value of time, velocity, and transfer speed.
  • 2) EV as an Energy Provider:: A provider prefers V2V trading over selling surplus power to the grid when V2V trading offers a positive utility.The paper assumes the grid buying price is lower than the cooperative V2V trading price.

B. Flexible Energy Management Protocol Design

The protocol centrally gathers real-time information, performs V2V matching, and filters matches using individual utility rationality before confirming power transfer.

  • B. Flexible Energy Management Protocol Design: The data control center periodically collects and updates real-time information and centrally coordinates the energy management procedure.The collected information includes EV mobility data and nearby infrastructure information.
  • B. Flexible Energy Management Protocol Design: Matching theory reduces frequent EV-to-EV information exchange compared with game-theoretic best-response procedures and supports two-sided stability.The paper motivates matching algorithms as a way to reduce communication overhead while considering both sides’ preferences.
  • B. Flexible Energy Management Protocol Design: Each tentative match is checked against consumer charging-station utility and provider positive-utility requirements before it becomes a confirmed trade.Rejected pairs return to the energy trading buffer for rematching.
  • B. Flexible Energy Management Protocol Design: The control center selects an available parking lot for each potential pair and directs successful pairs to perform power transfer there.Consumers failing to match more than m times receive a notice that a nearby charging station is preferable.

IV. EFFICIENT V2V MATCHING ALGORITHMS

The protocol models consumer-provider cooperation as a weighted bipartite graph and applies three matching algorithms to improve V2V energy-trading efficiency and network social welfare.

  • Three matching algorithms are proposed over a bipartite graph: max-weight, EV-consumer-oriented stable, and EV-provider-oriented stable matching.The graph represents energy consumers and energy providers as two vertex sets.
  • A. Bipartite Graph Construction: The graph connects EV consumers with EV providers, and each edge weight represents the sum utility of the connected EVs.The matching objective is therefore expressed through weighted consumer-provider edges.
  • B. Max-Weight V2V Matching Algorithm: The max-weight algorithm transforms network-social-welfare optimization into finding a maximum-weight matching in the constructed graph.Network social welfare is defined as the sum of edge weights for matched EV pairs.
  • B. Max-Weight V2V Matching Algorithm: The algorithm iteratively augments the matching and updates feasible vertex labels until every vertex is covered by a perfect matching.The final matching identifies the matched EV pairs.
  • B. Max-Weight V2V Matching Algorithm: The output perfect matching is optimal, with maximum weight in the bipartite graph when the two vertex sets have equal size.The algorithm has computational complexity O(max(N, K)^3).

C. Stable V2V Matching Algorithm

The max-weight matching optimizes network social welfare but does not account for individual EV rationality, motivating stable matching with two-sided preferences.

  • The centralized max-weight matching can optimize network social welfare but may be unstable because it ignores each EV’s individual rationality.Ignoring individual rationality may lead to unstable and deviated energy-trading behaviors.
  • The stable V2V matching approach incorporates two-sided preferences to obtain stable matchings while also optimizing network social welfare.

1) Stable Matching:

Stable matching models EV consumers and providers as two sides with preference rankings, requiring individual rationality and no mutually preferred blocking pair.

  • EV consumers and providers are represented as the two sides of a one-to-one marriage model, with complete and transitive preferences over acceptable partners.Preferences are represented by rank-order lists.
  • A matching is individually rational when no EV prefers being unmatched to its assigned partner.
  • A blocking pair consists of two EVs that prefer each other over their assigned partners, creating an incentive to break the current matching.
  • A matching is stable if it is individually rational and contains no blocking pair.The deferred acceptance algorithm guarantees that at least one stable matching exists for every stable marriage problem.

2) EV-Consumer-Oriented and EV-Provider-Oriented V2V Matching Algorithms:

Two deferred-acceptance-style algorithms generate stable V2V matchings by alternating which side proposes, producing distributed procedures with linear complexity.

  • The EV-consumer-oriented algorithm has consumers propose, while providers retain their preferred offer and reject the others.
  • The EV-provider-oriented algorithm swaps the roles so providers propose and consumers hold or reject received proposals.
  • Both algorithms operate independently and in a distributed manner with computational complexity O(N + K).
  • The consumer-oriented procedure yields an EV-consumer-optimal stable matching, whereas the provider-oriented procedure yields an EV-provider-optimal stable matching.

V. SIMULATIONS AND DISCUSSIONS

The simulations evaluate the proposed protocol against traditional EV charging using utility, welfare, energy-consumption, and computation-time comparisons under a modeled urban EV scenario.

  • Simulation Setup: The simulations compare the proposed cooperative V2V protocol with traditional charging, using randomly initialized EVs in a 20 km × 20 km urban network.The scenario includes 50 EVs, two charging stations, and 25 parking lots.
  • Matching Algorithms: The EV-consumer-oriented matching algorithm is implemented through preference lists, proposal exchanges, temporary holds, rejection updates, and final pair formation.The algorithm operates on a constructed bipartite graph and outputs matched consumer-provider pairs.

B. EV Utility Comparison

Cooperative V2V charging improves utilities for participating EVs, while stable matching preserves side-specific optimality and allows EVs with unfavorable trades to remain unmatched.

  • EV Utility Comparison: EV_C6 chooses the charging station because cooperative V2V charging cannot provide it with a better utility.This produces the same utility as the traditional charging protocol for that consumer.
  • EV Utility Comparison: Most EVs acting as energy providers achieve positive utility, creating incentives for EVs with extra power to participate in cooperative V2V charging.The provider-oriented stable matching algorithm never gives providers lower utilities than the consumer-oriented stable matching algorithm.
  • EV Utility Comparison: EV_P7 has zero utility because it finds no effective energy-trading partner and prefers to remain unmatched.Unmatching can result when neither side can obtain a positive or improved utility through the proposed trade.
  • EV Utility Comparison: The max-weight matching can produce higher individual utilities than stable matchings, although stable optimality is guaranteed only among stable matchings.The max-weight solution optimizes network welfare without imposing stability or individual rationality.

C. Network Social Welfare Comparison

The proposed protocol improves network social welfare over traditional charging, with max-weight matching highest overall and stable matching performance depending on the provider-to-consumer balance.

  • Network Social Welfare Comparison: Fig. 7 compares network social welfare across different numbers of EVs and matching protocols.The comparison includes max-weight, consumer-oriented stable, provider-oriented stable, and traditional charging protocols.
  • Network Social Welfare Comparison: The proposed protocol achieves an obvious network social welfare improvement over traditional EV charging across the simulated comparisons.The welfare values are averages over 10,000 randomly initialized simulations with small standard deviations.
  • Network Social Welfare Comparison: Max-weight V2V matching always achieves the highest network social welfare among the three proposed matching algorithms.Its matching is unstable because it does not account for individual rationality.
  • Network Social Welfare Comparison: When the number of energy providers exceeds the number of energy consumers, consumer-oriented stable matching outperforms provider-oriented stable matching.The consumer-oriented algorithm better exploits the larger set of provider candidates available to consumers.

D. Energy Consumption Reduction

Cooperative V2V charging reduces network energy consumption across all proposed matching algorithms, while matching choices trade welfare or stability against computational cost.

  • Energy Consumption Reduction: Energy consumption of the involved EVs is effectively reduced by the proposed protocol with all three V2V matching algorithms.Reduction is measured as the network energy-cost difference from traditional charging at the nearest station.
  • Energy Consumption Reduction: The energy-consumption reduction results are limited to the simulated scenario.The paper identifies broader energy-Internet settings as future work, including charging stations, smart houses, and EVs.
  • Energy Consumption Reduction: Fig. 9 compares computation time for the three matching algorithms using 1,000 randomized realizations on a logarithmic axis.The logarithmic scale accommodates the large computation-time difference involving max-weight matching.
  • Energy Consumption Reduction: The max-weight algorithm provides the best network performance but may yield unstable matchings, whereas both stable algorithms provide side-optimal stable outputs with low complexity.This summarizes the protocol’s welfare, stability, and computational trade-offs.
Loading 1802.06059v1…