Individual Electric Vehicles Research Articles

An Approach Towards Extreme Fast Charging Station Power Delivery for Electric Vehicles with Partial Power Processing

This article proposes an approach for realizing the power delivery scheme for an extreme fast charging (XFC) station that is meant to simultaneously charge multiple electric vehicles (EVs). A cascaded H-bridge converter is utilized to directly interface with the medium voltage grid while dual-active-bridge based soft-switched solid-state transformers are used to achieve galvanic isolation. The proposed approach eliminates redundant power conversion by making use of partial power rated dc–dc converters to charge the individual EVs. Partial power processing enables independent charging control over each EV, while processing only a fraction of the total battery charging power. Practical implementation schemes for the partial power charger unit are analyzed. A phase-shifted full-bridge converter-based charger is proposed. Design and control considerations for enabling multiple charging points are elucidated. Experimental results from a down-scaled laboratory test-bed are provided to validate the control aspects, functionality, and effectiveness of the proposed XFC station power delivery scheme. With a down-scaled partial power converter that is rated to handle only 27% of the battery power, an efficiency improvement of 0.6% at full-load and 1.6% at 50% load is demonstrated.

Spatio-Temporal Non-Intrusive Direct V2V Charge Sharing Coordination

Direct vehicle-to-vehicle (V2V) charge sharing system has the potential to provide more flexibility to electric vehicle (EV) charging without depending on the charging station infrastructure or building designated parking lots. It can also provide an opportunity to shift peak time utility load to off-peak times. However, the assignment between the EVs that demand energy and the EVs with surplus energy or existing charging stations is a challenging problem as it has to be performed in real time considering their spatio-temporal distribution, availability, and grid load. In this paper, we study this assignment problem specifically in a supplier non-intrusive scenario (without changing their mobility) and aim to understand the potential benefits of a direct V2V charge sharing system. To this end, we present two new algorithms to match the demander EVs to suppliers for charging. In the first one, the maximum system benefit is targeted under multiple system objectives with different priorities and considering the grid load. In the second one, individual EV priorities are taken into account and a stable matching process depending on predefined user preferences is provided. We conduct extensive simulations using real user commuting patterns and existing charging station locations in three different cities together with a newly developed probabilistic EV charging behavior model and EV mobility. Simulation results show that direct V2V charge sharing can reduce the energy consumption of EVs by 20%–35% by providing a closer charging facility compared to the charging station only case, offload grid charging power in peak times by 35%–55%, and help sustain up to 2.2× EV charging requests without building new charging infrastructure, while causing a marginal increase (i.e., 3%–9%) in the energy cycling of supplier EVs. The results also show that the proposed matching algorithms offer 80%–90% more energy consumption reduction compared to the intrusive V2V charging at designated parking lots.

Eliciting Multi-Dimensional Flexibilities From Electric Vehicles: A Mechanism Design Approach

Electric vehicles (EVs) have been well recognized as a deferrable load with the flexibility to shift their energy demands over time. Although this one-dimensional flexibility has been extensively investigated both by research and industrial implementations, the expanding energy demand and the associated uncertainties still make the integration of a large population of EVs into power system reliably and economically greatly challenging. In this paper, we design an auction scheme via mechanism design to elicit two additional flexibilities from EVs, namely energy flexibility and deadline flexibility. An offline mechanism is firstly designed as a benchmark based on the famous Vickrey–Clark–Groves mechanism. Then based on the primal-dual approach, we propose an online auction, in which all bids are truthful, the loss of social welfare is bounded by competitive ratio, and the mechanism can be implemented in polynomial time. By the numerical results, we quantitatively show that both the power system operators and individual EVs can benefit from the integration of the multi-dimensional flexibilities through our proposed mechanisms.

User behaviour and electric vehicle charging infrastructure: An agent-based model assessment

The transition to electric mobility is accelerating, and, thus it is increasingly important to be able to anticipate and adapt future development of the electric vehicle charging infrastructure. A novel agent-based simulation framework coupled with a detailed geo-referenced digital model of the built infrastructure is developed and applied. The charging behaviour of individual electric vehicle users as well as the spatial distributions of electric vehicles are accounted for in the simulation framework. More than 2500 scenarios of the transition to electric mobility in a mid-size city in Switzerland are assessed. The time to break-even of the electric vehicle charging infrastructure is up to 50% shorter when users are charged on the basis of parking fees rather than power sales. However, the revenues from parking fees are shown to be more sensitive to the behaviours and preferences of the users. At today’s low penetrations of electric vehicles, the profitability of the charging infrastructure is very uncertain, and thus entrants into the marketplace will have substantial financial exposure until the penetrations are of order 10%. Additionally, it is shown that, at specific transformers, public charging considerably increases grid loads by up to 78% during peak hours; these local increases, rather than the average city-wide increase in load, are the critical determinant of the required upgrades to the distribution grid. Overall, this novel simulation framework facilitates the planning of electric vehicle charging infrastructure that will support a successful transition to electric mobility.

Definition and Evaluation of Model-Free Coordination of Electrical Vehicle Charging With Reinforcement Learning

Demand response (DR) becomes critical to manage the charging load of a growing electric vehicle (EV) deployment. Initial DR studies mainly adopt model predictive control, but models are largely uncertain for the EV scenario (e.g., customer behavior). Model-free approaches, based on reinforcement learning (RL), are an attractive alternative. We propose a new Markov decision process (MDP) formulation in the RL framework, to jointly coordinate a set of charging stations. State-of-the-art algorithms either focus on a single EV, or control an aggregate of EVs in multiple steps (e.g., 1) make aggregate load decisions and 2) translate the aggregate decision to individual EVs). In contrast, our RL approach jointly controls the whole set of EVs at once. We contribute a new MDP formulation with a scalable state representation independent of the number of charging stations. Using a batch RL algorithm, fitted $Q$ -iteration, we learn an optimal charging policy. With simulations using real-world data, we: 1) differentiate settings in training the RL policy (e.g., the time span covered by training data); 2) compare its performance to an oracle all-knowing benchmark (providing an upper performance bound); 3) analyze performance fluctuations throughout a full year; and 4) demonstrate generalization capacity to larger sets of charging stations.

State Space Model of Aggregated Electric Vehicles for Frequency Regulation

Existing models featuring numerous electric vehicles (EVs) for centralized frequency regulation achieved high accuracy at the expense of heavy computational workloads and a high real-time communication requirement. This paper develops a state space model (SSM) that provides a probability to realize the real-time power control of aggregated EVs with high accuracy and computational efficiency but a low real-time communication requirement. The SSM, a reduced model based on the state space method, accurately describes aggregated EVs with different connecting states and various state-of-charge (SOC) states. Considering heterogeneous charging characteristics and random traveling behaviors of EVs, the SSM realizes the state transition prediction and the regulation capacity estimation with the Markov state transition method, which has a much higher computational efficiency than the existing models. The SSM is used for the frequency regulation, and the SOC adaptive coefficient is implemented to derive the identical control signal and improve the prediction accuracy. The SSM lowers the real-time communication requirement by replacing some real-time processes with offline processes. Meanwhile, the identical control signal is more suitable for real-time dispatching because it broadcasts the control signal to individual EVs globally. Simulation results indicate that the SSM achieves the high prediction accuracy with much higher computational efficiency. Comparison results are conducted to validate the effectiveness of SSM for real-time frequency regulation.

Enhanced primary frequency control from EVs: a fleet management strategy to mitigate effects of response discreteness

Electric vehicle (EV) chargers can be controlled to support the grid frequency by implementing a standard-compliant fast primary frequency control (PFC). This study addresses potential effects on power systems due to control discreteness in aggregated EVs when providing frequency regulation. Possible consequences of a discrete response, as reserve provision error and induced grid frequency oscillations, are first identified by a theoretical analysis both for large power systems and for microgrids. Thus, an EV fleet management solution relying on shifting the droop characteristic for the individual EVs is proposed. The PFC is implemented in a microgrid with a power-hardware-in-the-loop approach to complement the investigation with experimental validation. Both the analytical and the experimental results demonstrate how the controller performance is influenced by the response granularity, and that related oscillations can be prevented either by reducing the response granularity or by applying appropriate shifts on the droop characteristics for individual EVs.

Grid Loading Due to EV Charging Profiles Based on Pseudo-Real Driving Pattern and User Behavior

This paper defines a method for generating individual electric vehicle (EV) charging patterns, and it intends to quantify the realistic loading impact on distribution grid feeders. The inputs are based on the historical driving characteristics of private conventional vehicles from Denmark and home plug-in behavior of EVs from Japan. The first input is used to define properties, such as the daily driven distance and the expected departure and arrival time, which determines the possible home charging window. The second input is used to quantify the probability of performing a domestic charge every day. Because most of the EVs do not need to charge every day, even when considering a 100% EV penetration scenario, the amount of simultaneous charging with domestic single-phase charging power (3.7 kW) determines a coincidence factor lower than 45%. When considering three-phase charging (11 kW), the combined power of the EV population increases only to 50% because of shorter charging sessions. Although the power increase, due to 11-kW charging, is likely to trigger grid components overloading, it is highlighted how uncontrolled distribution of single-phase charging could be responsible for local voltage unbalances.

Real-time adjustment of load frequency control based on controllable energy of electric vehicles

The time-varying characteristics of electric vehicle (EV) controllable energy and the rationality of frequency regulation (FR) demand power allocation have significant influences on participating in system FR. Combined with the state transition characteristics of EVs, the calculation models of real-time controllable quantities and real-time controllable energy of EVs are established. Then, considering the dynamic changes of EVs’ controllable energy, the system FR strategy with real-time adjusting scheme of FR coefficients is put forward. Finally, based on the unit participation time contribution, the selecting strategy for individual EVs to participate in FR is proposed. The simulation results show that based on the calculation of EVs’ real-time controllable energy, the proposed load frequency control model with real-time allocation of FR demand power suppresses the frequency deviation effectively, and the private electric car is found to have the most potential for the FR system.

Real-time frequency regulation using aggregated electric vehicles in smart grid

The electric vehicle (EV) market has witnessed a continuous and steady increase in the past few years. The benefits of lower energy costs and less greenhouse gas (GHG) emission have been widely recognized by customers. In addition to these benefits for the transportation sector, EVs are also considered a critical supplementary resource for building a sustainable energy system in a smart grid environment. The applications of EVs in a smart grid have attracted wide attention in recent years. One appealing application is to use aggregated EVs as either energy sources or sinks to provide a service of frequency regulation for the request signals from the grid. A real-time decision-making model is proposed in this paper for the EV aggregator to dynamically control the energy flow between the grid and each individual EV in the aggregated group as an effective response to the signals of frequency regulation issued by the grid using Markov Decision Process. The aggregator’s benefit is maximized through the identification of a set of optimal charging/discharging decisions for the aggregated EVs. A semi-online solution strategy is also proposed to find the near-optimal decisions on a real-time basis. A numerical case study is used to illustrate the effectiveness of the proposed model.

Controllability Evaluation of EV Charging Infrastructure Transformed from Gas Stations in Distribution Networks with Renewables

A considerable market share of electric vehicles (EVs) is expected in the near future, which leads to a transformation from gas stations to EV charging infrastructure for automobiles. EV charging stations will be integrated with the power grid to replace the fuel consumption at the gas stations for the same mobile needs. In order to evaluate the impact on distribution networks and the controllability of the charging load, the temporal and spatial distribution of the charging power is calculated by establishing mapping the relation between gas stations and charging facilities. Firstly, the arrival and parking period is quantified by applying queuing theory and defining membership function between EVs to parking lots. Secondly, the operational model of charging stations connected to the power distribution network is formulated, and the control variables and their boundaries are identified. Thirdly, an optimal control algorithm is proposed, which combines the configuration of charging stations and charging power regulation during the parking period of each individual EV. A two-stage hybrid optimization algorithm is developed to solve the reliability constrained optimal dispatch problem for EVs, with an EV aggregator installed at each charging station. Simulation results validate the proposed method in evaluating the controllability of EV charging infrastructure and the synergy effects between EV and renewable integration.

Retail and Wholesale Electricity Pricing Considering Electric Vehicle Mobility

We consider the operation of multiple charging network operators (CNOs), each serving a population of electric vehicle (EV) drivers who subscribe to them for charging services. The CNOs set retail prices for the charging stations they operate, which effectively control the electricity demand and queues formed at their charging stations. The stations’ electricity demand also ties the CNOs’ retail optimization problem to their position in the wholesale market of electricity, creating a coupling between the electric transportation systems and the bulk power grid, which is managed by an independent system operator (ISO). In this paper, we provide retail pricing mechanisms and iterative wholesale market-clearing algorithms that together maximize social welfare in this coupled system. Specifically, our algorithm would enable the CNO to reduce its electricity costs and the customers’ travel delays (including charging station queueing times) and the ISO to reduce societal electricity costs by tapping into the EVs’ flexibility as mobile electric loads. Our theoretical analysis relies on the application of congestion pricing techniques to a capacitated user equilibrium problem on an energy-and-time expanded transportation network model. This expanded network allows us to track the travel costs of individual EVs, including their limited battery charge, the cost of waiting in queues at public charging stations, as well as the temporally variant nature of travel times.

Integrated configuration and charging optimization of aggregated electric vehicles with renewable energy sources

The commercialization and wide application of electric vehicles (EVs) rely on the development of EV charging infrastructure and its coordinated operation in the electric power grid with renewable energy sources. As the EV aggregator is introduced as intermediate control entity to form a multi-layer control framework, the optimal configuration of EV aggregator and the charging power regulation are closely related to each other. This paper proposes an integrated optimization to solve the mixed configuration and operation problem of EV aggregator. A hybrid optimization algorithm based on partial swarm optimization (PSO) and sequential quadratic programming (SQP) is formulated to find the best solution for the location and size of EV aggregator in the distribution network and the charging scheduling of each individual EV. The master problem in the coupled optimization algorithm is to calculate the optimal configuration of the EV aggregator based on the estimation of the EV charging load, and based on the setting of EV aggregator the subproblem is to obtain the charging plan for individual EV under the day-ahead optimal dispatching of charging demand in the distribution grid. Finally, 123-bus distribution system is adopted to analyze the characteristics of the proposed model for concurrent optimization of EV aggregator and its operation in the test distribution network. The simulation results validate the hybrid optimization algorithm in integrated configuration and operation of EV charging infrastructure.

A Hierarchical Governor/Turbine and Electric Vehicles Optimal Control Framework for Primary Frequency Support in Power Systems

This paper aims to develop a hierarchical framework for control of governor/turbine (G/T) and electric vehicles (EVs) as primary reserves to provide primary frequency support. The framework involves two decision layers. The top layer based on EV utilization cost dispatches the primary reserve references that balances load and generation following a generation drop. Subsequently, control commands of this layer transmit to G/T and the lower layer. The lower layer dispatches the aggregated EV power change that commanded from the top layer among large number of EVs. Therefore, in this layer, each individual EV will solve an optimal EV dispatch control in a decentralized manner while meeting its own constraints. To draw further insights into the performance of our proposed scheme, we investigate the impact of the proposed control framework on a simplified Great Britain power system model. The listed simulation results show that the proposed scheme restores the power system frequency within a short period of time upon a sudden supply demand imbalance along with minimizing the aggregated and individual EV utilization and G/T generation change for participating in primary frequency support.

Vehicle grid integration for demand response with mixture user model and decentralized optimization

With the rapidly growing electric vehicle adoption rate and increasing number of public electric vehicle charging stations in recent years, electric vehicle becomes more and more critical in the demand response programs. Development in Vehicle to Grid technology has converted electric vehicle to distributed energy resources. Using the Electric Vehicle Smart Charging Infrastructures on UCLA campus and city of Santa Monica as testbeds, we have collected real-world datasets of electric vehicle usage, based on which, we proposed optimal bi-directional charging control strategies to integrate electric vehicle in commercial and public parking facilities into the power grid as distributed energy resources for demand response programs by two-stage distributed optimization and water-filling algorithm. Driver behavioral uncertainties have been considered in our approach. Specifically, electric vehicle users are clustered by their behavioral patterns using a modified Latent Semantic Analysis. The first-stage optimization is performed to minimize energy cost using day-ahead wholesale energy price with predictions on energy demand and electric vehicle availability which generated by a mixture user model. Decentralized optimization (second-stage) is carried out on the next day in real-time to control individual electric vehicle so that the aggregated load can follow the first-stage optimal profile. As an alternative, a fast converging water filling algorithm is proposed and compared with two-stage optimization. Extensive simulation results show that proposed charging controls can utilize electric vehicle as distributed energy resource to accommodate demand response program while satisfying electric vehicle energy demands and providing significant energy cost savings.

Optimal Day-Ahead Charging Scheduling of Electric Vehicles Through an Aggregative Game Model

The electric vehicle (EV) market has been growing rapidly around the world. With large scale deployment of EVs in power systems, both the grid and EV owners will benefit if the flexible demand of EV charging is properly managed through the electricity market. When EV charging demand is considerable in a grid, it will impact spot prices in the electricity market and consequently influence the charging scheduling itself. The interaction between the spot prices and the EV demand needs to be considered in the EV charging scheduling, otherwise it will lead to a higher charging cost. A day-ahead EV charging scheduling based on an aggregative game model is proposed in this paper. The impacts of the EV demand on the electricity prices are formulated with the game model in the scheduling considering possible actions of other EVs. The existence and uniqueness of the pure strategy Nash equilibrium are proved for the game. An optimization method is developed to calculate the equilibrium of the game model through quadratic programming. The optimal scheduling of the individual EV controller considering the actions of other EVs in the game is developed with the EV driving pattern distribution. Case studies with the proposed game model were carried out using real world driving data from the Danish National Travel Surveys. The impacts of the EV driving patterns and price forecasts on the EV demand with the proposed game model were also analysed.

Decentralized charging control strategy of the electric vehicle aggregator based on augmented Lagrangian method

Uncoordinated charging of large-scale electric vehicles can affect the safe and economic operation of distribution power system. As an intermediary between power grid and individual electric vehicle, electric vehicle aggregator can play an important role in the coordination and management of electric vehicle charging. In this paper, we develop a decentralized EV charging control strategy of the electric vehicle aggregator for scheduling the flexible charging demand of plug-in electric vehicles in residential distribution networks. First, the centralized charging control model of the electric vehicle aggregator aiming at the maximization of its revenue under the background of time of use price mechanism is established, meanwhile meeting individual charging requirements and satisfying distribution network constraints. Then a decentralized charging control strategy of the electric vehicle aggregator based on the augmented Lagrangian method and the alternating direction multiplier method is proposed. The electric vehicles can decide their charging plan locally with the decentralized strategy, which can eliminate the problem of high communication cost, low computing efficiency and privacy protection caused by the centralized charging control model in practical application. The effectiveness of the proposed strategy is evaluated with a representative distribution network model.

A Comprehensive Model to Integrate Emerging Resources From Supply and Demand Sides

This paper extensively models the interactions of emerging players in future power systems to analyze their impacts on electricity markets. To this end, renewable energy resources are modeled in such a way that wind power poses uncertainty on the supply side, and rooftop photovoltaics add uncertainty to the demand side. Moreover, both uncontrolled and controlled behaviors of individual electric vehicles (EV) in electricity markets are addressed through a new EV model. Further, a comprehensive demand response model considering several customer-driven constraints is developed to undertake the practical constraints of customers. A stochastic market clearing formulation is presented to comprehensively account for the unique features of the given resources while evaluating their impacts. The numerical results clearly show the importance of such modeling in electricity markets to investigate the mutual impacts of emerging resources.

Three-Stage Distribution Feeder Control Considering Four-Quadrant EV Chargers

With the increased penetration of electric vehicle (EV) chargers in distribution systems, there is a need to understand and minimize their impact on medium-voltage (MV) and low-voltage (LV) networks. Thus, this paper proposes a three-stage algorithm to coordinate the operation of four-quadrant EV chargers with other volt/var control devices in MV and LV distribution feeders. The first stage operates on a day-ahead basis and defines the load tap changer and capacitor schedules while minimizing the peak load associated with EVs in the distribution system. The second and third stages update their operation every five minutes, to fairly allocate the aggregated and individual EV loads in the MV an LV feeders, respectively, while minimizing active power losses and voltage deviations. The proposed technique is applied to CIGRE’s North-American MV and LV benchmark systems to demonstrate its ability to properly allocate EV loads, and improve distribution system performance in terms of losses and voltage deviations.

Estimating the Charging Profile of Individual Charge Sessions of Electric Vehicles in The Netherlands

The mass adoption of Electric Vehicles (EVs) might raise pressure on the power system, especially during peak hours. Therefore, there is a need for delayed charging. However, to optimize the charging system, the progression of charging from an empty battery to a full battery of the EVs, based on real-world data, needs to be analyzed. Currently, many researchers view this charging profile as a static load and ignore the actual charging behavior during the charging session. However, this study investigates how different factors influence the charging profile of individual EVs based on real-world data of charging sessions in The Netherlands, and thereby enable optimization analysis of EV smart charging schemes.