Electric Vehicle Charging Demand Research Articles

A Data-driven Agent-based Planning Strategy of Fast-Charging Stations for Electric Vehicles

Electric vehicles (EVs) are believed to play an important role in mitigating carbon emissions. To accommodate the increasing number of EVs, the co-planning of electrical distribution networks and the charging infrastructure, such as fast-charging stations (FCSs), becomes an emergent task. Both the charging demand of EVs and the stability of the electricity network should be satisfied. In this paper, we proposed a data-driven agent-based planning strategy for FCSs. Different from the conventional planning strategy, we utilized machine learning tools to consider EV behaviors at the microscopic level in a planning problem. First, a Partially Observable Markov Decision Process of EVs is established, and multi-agent deep reinforcement learning is utilized to learn the charging and driving decisions of EVs under different transportation network typologies and FCSs planning schemes. Second, a data-driven agent-based traffic assignment model (DA-TAM) is proposed to aggregate the atomic behaviors of EVs, which can present the sensitivity of the traffic flow and EV charging demand to the FCS planning schemes. Third, the DA-TAM is adapted to the proposed planning model to ensure the quality of service and prevent unbalanced traffic flow. Through the proposed method, microscopic behaviors of EVs can be reflected, and the impact of the planning scheme on the traffic condition can be revealed. The proposed methodologies are verified in the case studies. It can be concluded that compared with the baseline, the presented agent-based planning strategy can serve more EV charging demands (increased by 33%), cause less waiting time (decreased by 59%) in FCSs to enhance the quality of service, and encounter less severe traffic unbalance problems (increased by 17%).

Multistep Multiagent Reinforcement Learning for Optimal Energy Schedule Strategy of Charging Stations in Smart Grid.

An efficient energy scheduling strategy of a charging station is crucial for stabilizing the electricity market and accommodating the charging demand of electric vehicles (EVs). Most of the existing studies on energy scheduling strategies fail to coordinate the process of energy purchasing and distribution and, thus, cannot balance the energy supply and demand. Besides, the existence of multiple charging stations in a complex scenario makes it difficult to develop a unified schedule strategy for different charging stations. In order to solve these problems, we propose a multiagent reinforcement learning (MARL) method to learn the optimal energy purchasing strategy and an online heuristic dispatching scheme to develop a energy distribution strategy in this article. Unlike the traditional scheduling methods, the two proposed strategies are coordinated with each other in both temporal and spatial dimensions to develop the unified energy scheduling strategy for charging stations. Specifically, the proposed MARL method combines the multiagent deep deterministic policy gradient (MADDPG) principles for learning purchasing strategy and a long short-term memory (LSTM) neural network for predicting the charging demand of EVs. Moreover, a multistep reward function is developed to accelerate the learning process. The proposed method is verified by comprehensive simulation experiments based on real data of the electricity market in Chicago. The experiment results show that the proposed method can achieve better performance than other state-of-the-art energy scheduling methods in the charging market in terms of the economic profits and users' satisfaction ratio.

An Efficient and Secure Trading Framework for Shared Charging Service Based on Multiple Consortium Blockchains

While electric vehicles are proliferating rapidly, the number of charging piles is only 1/16 of the number of electric vehicles, showing big tension between small supply of public charging piles and intense charging demand of electric vehicles. To alleviate the tension, it is highly desired to incorporate existing idle private charging piles into a shared charging service. However, existing solutions do not well implement the system due to some non-negligible issues, like poor consensus performance or lack of verification for outputs from blockchain networks. In the paper, we propose an efficient and secure trading framework built atop multiple consortium blockchains. It devises a novel node voting mechanism so as to improve consensus efficiency. Meanwhile, it employs a publicly verifiable design based on the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$(t,n)$</tex-math></inline-formula> Boneh-Lynn-Shacham (BLS) threshold signature technique, so as to enable entities outside the blockchain to verify the consistency between the output results with the valid consensus. Besides these novel system-level designs, it also introduces a fair and robust trading strategy running by smart contracts on the platform. Furthermore, we implemented a proof-of-concept system and conducted extensive experiments. The results show that the overall latency across chains is reduced by about <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$26\%$</tex-math></inline-formula> with the node voting mechanism and the transactions fees are comparable to recent popular cross-chain applications.

A Bi-Level Optimization and Scheduling Strategy for Charging Stations Considering Battery Degradation

This paper proposes a bi-level optimization scheduling strategy for integrated photovoltaic (PV) and energy storage systems (ESS) to meet electric vehicle (EV) charging demands while reducing charging costs. First, a battery degradation cost model is developed in order to convert the long-term costs into short-term costs for real-time operation. The upper layer of ESS and power grid operation strategies are obtained by minimizing costs associated with battery degradation and distribution grid costs. The lower layer considers the PV uncertainty and the error caused by the upper layer operation strategy, and obtains the lower layer operation strategy by adding a penalty function to minimize fluctuations in power. Second, the author proposes a global optimization algorithm that combines Particle Swarm Optimization (PSO) and Sequential Quadratic Programming (SQP) in order to solve the above-mentioned models, effectively combining the global search feature of PSO with the local search capability of SQP. Finally, the bi-level optimization scheduling strategy is obtained by solving the model through the algorithm. Simulation results verify the practicality of the scheduling strategy and the effectiveness of the proposed algorithm.

Predicting electric vehicle charging demand using a heterogeneous spatio-temporal graph convolutional network

Short-term Electric Vehicle (EV) charging demand prediction is an essential task in the fields of smart grid and intelligent transportation systems, as understanding the spatiotemporal distribution of charging demand over the next few hours could help operators of charging stations and the grid to take measures (e.g., dynamic pricing) in response to varying charging demand. This study proposed a heterogeneous spatial–temporal graph convolutional network to predict the EV charging demand at different spatial and temporal resolutions. Specifically, we first learned the spatial correlations between charging regions by constructing heterogeneous graphs, i.e., a geographic graph and a demand graph. Then, we used graph convolutional layers and gated recurrent units to extract spatio-temporal features in the observations. Further, we designed a region-specific prediction module that grouped regions based on graph embedding and point of interest (POI) data for prediction. We used a large real-world GPS dataset which contained over 76,000 private EVs in Beijing for model training and validation. The results showed that, compared with recently popular spatio-temporal prediction methods, the proposed model had superior prediction accuracy and steady performance at different scales of regions. In addition, we conducted ablation studies and hyperparameter sensitivity tests. The results suggested that incorporating the demand graph and geographic graph could help improve model performance.

Seasonal variance in electric vehicle charging demand and its impacts on infrastructure deployment: A big data approach

Electric vehicle (EV) charging demand is an essential input of charging facility location models. However, charging demand may vary across seasons. In response, this paper first provided insights into the seasonal variance in charging demand using a unique GPS trajectory dataset which contained travel, parking, and charging information of 2,658 private EVs in Beijing. The dataset was collected in January, April, July, and October 2018, which were representative months in winter, spring, summer, and autumn, respectively. Through statistical and spatiotemporal analyses, we found that in winter, EVs got recharged when their state of charge (SOC) was lower: the average SOCs on working days were 51.96%, 48.39%, 50.86%, and 43.50%, in spring, summer, autumn, and winter, respectively. Furthermore, the central urban areas tended to have a higher charging demand in winter. To further explore how the seasonal variance in charging demand may influence infrastructure deployment, we used the classical p-median model to deploy charging facilities with the charging demands in the four seasons, considering the modifiable areal unit problem (MAUP). The results suggested that the seasonal variance did influence the layout of charging facilities under different spatial analysis units (SAUs). The deployment of charging facilities in the central urban areas and outer suburbs tended to be more sensitive to seasonal variance in charging demand. The findings are expected to be useful for charging infrastructure planning in both the transport and power sectors.

Charging management and pricing strategy of electric vehicle charging station based on mean field game theory

AbstractAs an intermediary between the power grid and the electric vehicles (EVs) in the charging station, the charging pile promotes the exchange of electric energy between the power grid and EV group and also brings benefits to the charging station. It is difficult to accurately analyze the detailed energy trading behavior of a large number of charging piles with the power grid and EV group. Based on the theoretical framework of mean field game (MFG), this paper considers the battery degradation and charging efficiency taking into account the charging demand of EVs, the charging control problem of energy storage charging piles is proposed to achieve the goal of minimizing the cost of the charging station. In addition to modeling the interaction between the charging station and power grid and EVs as a finite‐time dynamic game problem, optimal decentralized energy scheduling control strategies are formulated for charging piles, and by introducing the mean field term, the optimal pricing strategy for power trading between the charging station and EVs is obtained under the mean field equilibrium condition. An iterative algorithm is proposed to solve the charging control problem, and the rationality of the problem and the effectiveness of the algorithm are verified by numerical simulation.

Mind the gap: Modelling difference between censored and uncensored electric vehicle charging demand

Electric vehicle charging demand models, with charging records as input, will inherently be biased toward the supply of available chargers. These models often fail to account for demand lost from occupied charging stations and competitors. The lost demand suggests that the actual demand is likely higher than the charging records reflect, i.e., the true demand is latent (unobserved), and the observations are censored. As a result, machine learning models that rely on these observed records for forecasting charging demand may be limited in their application in future infrastructure expansion and supply management, as they do not estimate the true demand for charging. We propose using censorship-aware models to model charging demand to address this limitation. These models incorporate censorship in their loss functions and learn the true latent demand distribution from observed charging records. We study how occupied charging stations and competing services censor demand using GPS trajectories from cars in Copenhagen, Denmark. We find that censorship occurs up to 61% of the time in some areas of the city. We use the observed charging demand from our study to estimate the true demand and find that censorship-aware models provide better prediction and uncertainty estimation of actual demand than censorship-unaware models. We suggest that future charging models based on charging records should account for censoring to expand the application areas of machine learning models in supply management and infrastructure expansion.

Optimal Electric Vehicle Fleet Charging Management with a Frequency Regulation Service

Electric vehicles are able to provide immediate power through the vehicle-to-grid function, and they can adjust their charging power level when in the grid-to-vehicle mode. This allows them to provide ancillary services such as frequency control. Their batteries differ from conventional energy storage systems in that the owner’s energy requirement constraint must be met when the vehicles participate in a frequency control system. An optimization problem was defined by considering both the owner satisfaction and frequency control performance. The main contribution of the proposed paper, compared to the literature, are (1) to keep the total available energy stored in the batteries connected to a charging station in an optimal region that favors the frequency regulation capability of the station and the proposed QoS and (2) to consider the optimal region bounded by the efficiency thresholds of the charger to allow for maximum regulation power. The problem is expressed as a multi-criteria optimization problem with time-dependent references. The paper presents an energy management strategy for frequency control, describes a concept of an optimal time-dependent state of charge for electric vehicle charging demands, and considers the power dependence of the electric vehicle charger efficiency. Finally, the simulation results are presented via Matlab/Simulink to prove the effectiveness of the proposed algorithm.

Estimating the electric vehicle charging demand of multi-unit dwelling residents in the United States

Early battery electric vehicle (EV) adopters can access home chargers for reliable charging. As the EV market grows, residents of multi-unit dwellings (MUDs) may face barriers in owning EVs and charging them without garage or parking availability. To investigate the mechanisms that can bridge existing disparities in home charging and station deployment, we characterized the travel behavior of MUD residents and estimated their EV residential charging demand. This study classifies the travel patterns of MUD residents by fusing trip diary data from the National Household Travel Survey and housing features from the American Housing Survey. A hierarchical agglomerative clustering method was used to cluster apartment complex residents’ travel profiles, considering attributes such as dwell time, daily vehicle miles traveled (VMT), income, and their residences’ US census division. We propose a charging decision model to determine the charging station placement demand in MUDs and the charging energy volume expected to be consumed, assuming that MUD drivers universally operate EVs in urban communities. Numerical experiments were conducted to gain insight into the charging demand of MUD residents in the US. We found that charging availability is indispensable for households that set out to meet 80% state of charge by the end of the day. When maintaining a 20% comfortable state of charge the entire day, the higher the VMT are, the greater the share of charging demand and the greater the energy use in MUD chargers. The upper-income group requires a greater share of MUD charging and greater daily kWh charged because of more VMT.

Energy management of green charging station integrated with photovoltaics and energy storage system based on electric vehicles classification

As the number of electric vehicles (EVs) increases, EV charging demand is also growing rapidly. In the smart grid environment, there is an urgent need for green charging stations (GCS) to effectively manage the internal photovoltaic (PV), energy storage system (ESS), charging behaviors of EVs and energy transactions with entities. In this paper, a novel EV classification approach was proposed for GCS, of which the objective is to minimize the total cost of energy trading between charging station and entities. Based on EV charging requirements and risk preference for participating in demand response (DR), EVs are classified into three categories: (1) regular; (2) conservative; and (3) vehicle to grid(V2G). For different types of EVs, charging price schemes are proposed for GCS to stimulate more EVs to interact with the grid and duly dis-charging their power in the capacity of V2G runners. In dealing with the uncertainty issues caused by photovoltaic generation, chance-constrained and probabilistic sequence methods are employed to solve the proposed framework with satisfactory computational efficiency. Case studies based on genuine PV generation, load and pricing data are carried out to verify the proposed framework. It is demonstrated V2G EVs can dramatically reduce the total cost to the GCS. Besides, V2G EVs also can greatly reduce the charging cost compared to regular and conservative EVs.

Electric vehicle charging and discharging scheduling strategy based on dynamic electricity price

The rapid growth in the number of electric vehicles (EVs) has significantly increased the demand for electricity for residents. In addition, because the charging time of EVs highly coincides with the peak period of user electricity consumption, the disorderly charging of EVs will lead to the overload of the power grid transformer. Traditional control methods lack certain robustness and do not fully consider the uncertainty of EVs. As a result, the V2G participation rate of electric vehicles cannot be determined, and the control reliability is low. To solve the above problems, this paper designs a reinforcement learning framework of Long Short-Term Memory network and Improved Linear programming algorithm (LSTM-ILP) to control the V2G of EVs.This paper comprehensively considers the overall electric vehicle charging demand, discharge potential, large grid electricity price, aggregator, and users’ interests demands. Firstly, aiming to minimize the charging and discharging fee of EVs and the load peak-to-valley difference of the power grid, a dynamic electricity price based on Long Short-Term Memory neural network (LSTM) is established. Then, the improved linear programming algorithm (ILP) is used to solve the charging and discharging optimization problem of EV, and the results are fed back to the input of the next iterative update of the LSTM, and finally, the optimal electricity price and EV charging and discharging schedule are achieved. The simulation results show that the LSTM-ILP framework can not only reduce the charging fee of electric vehicles, but also achieve the Peak and valley trimming of the grid load. Charging costs for EV users were reduced by 42.1% compared with unordered charging, and by 22% percent compared with orderly charging.

Dynamic planning and energy management strategy of integrated charging and hydrogen refueling at highway energy supply stations considering on-site green hydrogen production

The layout of electric vehicles charging stations and hydrogen refueling stations (HRSs) is more and more necessary with the development of electric vehicles (EVs) and progress in hydrogen energy storage technology. Due to the high costs of HRSs and the low demand for hydrogen, it is difficult for independent HRSs to make a profit. This study focuses on the dynamic planning of energy supply stations on highways in the medium and long term, considering the growth of EV charging demand and the change in the proportion of hydrogen fuel cell vehicles (HFCVs). Based on the perspective of renewable energy generators (REGs), this study seeks the dynamic optimal configuration and comprehensive benefits of adding HRS and battery to existing EVCS considering the travel rules of new energy vehicles (NEVs). The results show that (1) It is profitable for REGs to invest in HRSs; (2) The economy of investment in batteries by REGs depends on the source-load matching. It is feasible only when the output of renewable energy is difficult to meet the demand. (3) The business model of REGs producing hydrogen on-site and supplying both electricity and hydrogen is feasible.

Strategic Pricing of Electric Vehicle Charging Service Providers in Coupled Power-Transportation Networks

Electric vehicles (EVs) are developing at a rapid pace. The EV charging station (EVCS) plays an indispensable role in supplying EV charging demand and promoting the transportation electrification. This work proposes a pricing method to maximize the profit of EVCS owners, charging service providers. First, the tri-level pricing framework is established where EVCSs set charging prices at the upper level, EVs make route and charging options at the middle level, and the electricity price is cleared in power distribution network at the lower level. Second, the singlelevel optimal pricing model is established with traffic flow assigned, power generation scheduled and electricity price explicitly contained via KKT conditions. The proposed model considers the impact of EVCS configuration on traffic assignment and pricing strategy. The operation risk brought by uncertainties is managed via Conditional value at risk. Third, a decomposition solution algorithm is proposed, which tackles the computational burdens and overcomes the infeasibility of existing algorithms. Finally, the real-world urban network-based case studies verify the effectiveness and merits of the proposed method.

Optimization of charging stations integrated with dynamic transportation systems in metropolises

The development of electric vehicles (EVs) is expected to play an important role in achieving the emission reduction targets in the transportation sector by many countries and regions, including China’s dual carbon goals. A bottleneck to the mass roll-out of EVs is the limited charging facilities. This paper considers the planning of charging facilities for a new developing area in a metropolis, and an optimization model for charging station planning based on the dynamic transportation system is proposed. The proposed model is developed using an objective framework that considers the spatio-temporal characteristics of EV charging demand in order to minimize the overall cost while the constraints from suppliers and drivers are met. The Voronoi diagram is used to determine the final service boundary of each charging station, and the effect is verified in the planning of charge stations for the Yizhuang new town in Beijing. The case study confirms that the proposed method can optimize the charging facilities that fit well with the traffic network conditions. Furthermore, it is shown that the charging demand varies in accordance with the population density and regional functionality in different areas.

A data-driven framework for medium-term electric vehicle charging demand forecasting

The rapid phase-in of electric vehicles (EV) will cause unprecedented issues with managing the supply of electricity and charging stations. It is in the interest of utility providers and everyday consumers to be able to plan for peak charging times, and related congestion. While past work has been done for localized, short-term forecasting, it has not included longer term forecasting, or considered the relationships between multiple stations. Importantly, past work has also not offered a framework for dataset construction and evaluated different dataset features. We propose a methodology to forecast demand at public EV charging stations, and use it to explore the potential of data-driven models to predict demand up to one week in advance. Our strategy includes selecting parameters for formatting a dataset given a list of charging events, a way to consider flexible prediction horizons, and deployment of deep and supervised learning-based models. To the best of our knowledge, ours is the first study to propose machine learning to forecast medium-term public EV charging demand, to exploit weather and other features at public charging stations, and to forecast demand at multiple stations and the entire network. We validated our approach using data from eleven stations over three years from Scotland, UK. Our method outperforms the benchmark time series method, and predicts network demand with a symmetric mean absolute percentage error (SMAPE) of 5.9% and a mean absolute error (MAE) of 124.7 kWh, or less than twelve percent of average daily demand.

Assessment of Distribution System Margins Considering Battery Swapping Stations

Penetration of electric vehicles (EVs) into the market is expected to be significant in the near future, leading to an significant increase in EV charging demand, and that will create a surge in the demand for electrical energy. In this context, there is a need to find intelligent and cost effective means to make better use of electricity resources, improve the system flexibility, and slow the growth in demand. Therefore, swapping EV batteries rather than traditionally charging them can serve as flexible sources to provide capacity support for the power distribution grid when they are charged during off-peak periods prior to their swapping at the station. This paper presents a novel mathematical optimization model to assess distribution system margins considering different EV charging infrastructures. The proposed model maximizes the distribution system margins while considering the flexibility of battery swapping station loads and distribution grid limitations. To demonstrate the effectiveness of the proposed model, simulation results that consider the National Household Travel Survey data and a 32-bus distribution system are reported and discussed. Unlike charging EV batteries, swapping them would not affect system margins during the peak hours.

Short-term electric vehicle charging demand prediction: A deep learning approach

Short-term prediction of the Electric Vehicle (EV) charging demand is of great importance to the operation of EV fleets and charging stations. This paper develops a Long Short-Term Memory (LSTM) neural network to predict the EV charging demand at the station level for the next few hours (e.g., 1–5 h), using a unique trajectory dataset containing over 76,000 private EVs in Beijing in January 2018. To explore the performance of the LSTM model, we set up four scenarios by 1) comparing LSTM against two typical time series prediction models, i.e., the Auto-Regressive Moving Average model (ARIMA), and the Multiple Layer Perceptron model (MLP), 2) and investigating how different input data structures, sample sizes, and time spans and intervals would influence model accuracy. The results suggest that the LSTM model outperformed the ARIMA, and MLP models, and their MAPE1 values are 6.83 %, 21.58 %, and 18.31 %, respectively. In addition, we find that the time span and interval tend to be more influential to the LSTM model’s prediction accuracy than input data structures, and sample sizes. In general, the LSTM model with a shorter time span or interval (e.g., 1 h) would perform better.

Electric vehicle behavior modeling and applications in vehicle-grid integration: An overview

The increasing electric vehicle (EV) adoption in the context of transportation electrification and carbon neutrality pursuit brings both new challenges and opportunities for all the stakeholders in EV-grid integration. To fully unleash the potentials of EVs as flexible distributed energy storage to facilitate efficient EV-grid interactions, it is imperative to predicting spatio-temporal distributions of EV charging demand, optimizing charging infrastructure layout and implementing smart charging scheduling schemes. Appropriate EV behavior modeling plays a fundamental role to realize these targets. This paper aims to provide a comprehensive review on EV behavior modeling and its applications in EV-grid integration algorithm development. Various models have been developed to describe EV usage patterns, charging decision making process and response to smart charging schemes. In particular, the existing usage pattern models including temporal, spatial and energy sub-models are expounded, and different sub-models of charging choice and response to smart charging are also presented. An EV behavior modeling paradigm is proposed to provide guidance for EV behavior model selection in different application scenarios by developing different portfolios of temporal, spatial, energy usage, charging choice and response models. Accordingly, enabling EV behavior modeling for EV charging demands prediction and smart charging scheduling is covered in details. This study provides in-depth behavioural insights and viable approaches to developing efficient EV behavior models for advancing EV-grid integration and provides perspectives towards future research directions.

Design and Implementation of Renewable Based EV Charging Station

A significant transformation occurs globally as transportation switches from fossil fuel-powered to zero emission E-Vehicles. The transition to the electric vehicle requires an infrastructure of E-Vehicle charging stations everywhere. The rapid adoption of electric vehicles (EVs) calls for the widespread installation of EV charging stations. To maximize the profitability of charging stations, intelligent controllers that provide both charging and electric grid services are in great need. However, it is challenging to determine the optimal charging schedule due to the uncertain arrival time and charging demands of EVs. In this paper, we propose solar energy based three phase voltage source inverter fed grid. In this research solar energy is maintain a consistent output voltage to the load or grid. In this research PV system with a NeuroFuzzy-based MPPT algorithm and a Switched Z source Boost converte .