众多学者对V2G和储能的规划与管理进行了深入研究。电动汽车电池由于其容量大，可以作为储能单元，在电网负荷较低时储存多余的电力，然后在需求高峰时释放，从而保证电网的稳定运行[18]。

V2G技术与储能系统的结合为实现更高效、更灵活、更可靠的电力系统提供了新的机会[19]。电动汽车电池作为一种储能介质，可以减少投资和可再生能源和固定储能的建设，最大限度地减少资源浪费[20];智能电池技术的引入进一步促进了V2G的实现[21]。在经济方面，电池衰退成本、售电价格[22]和基础设施成本[23]影响了电动汽车作为储能单元向电网提供短期运营储备的能力[24];然而，电动汽车仍然有潜力为其他设施提供服务。例如，V2G可以作为储能嵌入到家庭和IES中，为家庭和商业中心提供能源，从而降低能源成本并提高环境效益[25\u26]。Sun 等人研究了电动汽车电池的 SOC 如何影响使用电动汽车作为住宅能源系统一部分的住宅能源系统的总成本。结果表明，电池老化成本仍然是影响电动汽车用户电池放电意愿的重要因素[27]。Wei等人将V2G作为储能嵌入IES中，使两者耦合，以最大限度地提高IES的经济和环境效益[25]。Hipolito、Vandet和Rich估计了电动汽车的储能潜力，并得出结论，只有当电池利用率低于40%时，电动汽车用户的电池放电意愿才会显著增加[28]。

风能和光能的高波动性给并网发电带来了相当大的挑战[29]。由于V2G技术具有响应时间快[30]、灵活的进出机制、相对固定的储能装置等特点，因此不需要额外投资，在清洁能源消耗过剩电力方面发挥着独特且不可替代的作用。

许多研究人员试图用电动汽车消耗这些可再生能源。Gao等人将具有V2G运行能力的风力发电机和电动汽车集成到配电网中，建立了V2G功率控制的数学模型，并部署了电动汽车的动态功率调节[31]。Jin等人提出了一种针对AGC涉及的大规模电动汽车、BESS和传统阻燃资源的协调控制策略，可以提高频率稳定性，促进可再生能源的整合[32]。Fattori、Anglani 和 Muliere 分析了不同渗透情景下光伏和电动汽车对电网的影响，其中 V2G 收益并不乐观，因为电动汽车充电行为与光伏发电的不同步导致对非光伏必须提供的容量的需求增加，并且当时电池成本相对于光伏成本来说很高 [11].Fathabadi等人分析了电源对配电网的影响;这些来源包括传统发电机 （CPG）、DG 可再生能源和电动汽车。实验表明，同时使用CPG、DG可再生能源和电动汽车充电/放电可以降低发电成本和最佳电压曲线;因此，通过在电网中同时使用CPG和DG可以获得最低的功率损耗[33]。Haddadian等人将电动汽车车队部署为分布式储能，并实现了与风能的最佳协调。研究表明，该方法可以降低昼夜运行成本，降低排放，并实现零弃风率的预测风能的彻底消耗[34]。Uddin等人开发了一种电池退化模型，并将其无缝集成到智能电网算法中。这种创新方法表明，将电动汽车连接到这种智能电网系统不仅可以有效地满足电网的需求，注入更多的清洁可再生能源，而且值得注意的是，延长了电动汽车电池的使用寿命[35]。Robledo等人介绍了一个示范项目的结果，包括建筑一体化光伏（BIPV）太阳能电池板，这表明FCEV可以将交通和电力部门整合到可持续能源系统中[36]。Bhatti 和 Salam 考虑通过光伏电网系统在办公室停车场进行日间电动汽车充电，并提出了基于规则的能源管理方案 （REMS）。这项工作可以整合其他稀土来源，如风能、潮汐能和生物质能[37]。Sufyan等人分析了系统成本和能量损失对不同程度的可再生能源集成、电动汽车容量和行驶距离的影响。他们的仿真结果清楚地表明，当可再生能源 （RES） 无缝集成到配电网络中时，运营费用将大幅降低。此外，V2G技术的使用已被证明对电动汽车用户有利，特别是在可再生能源渗透率高的场景中[38]。Noorollahi等人为电动汽车聚合器引入了一个新框架，并为四种不同的电动汽车充电场景设计了模型。他们的方法考虑了广泛的可再生能源，包括风力涡轮机、太阳能光伏电池板和地热装置。他们实施了一个能源调度模型来有效地优化这些场景[39]。Bartolini等人进行了一项分析，以探讨电动汽车车队如何有助于提高地区的自用能力，特别是在不可控可再生能源（RES）大量存在且不断增长的情况下，特别关注光伏系统。此外，他们的研究深入研究了电动汽车的影响以及可再生能源整合对该地区一氧化碳排放的影响₂ [[40].Rahbari等人提出了一种考虑可再生能源消耗和电动汽车作为储能装置的电动汽车智能停车场规模和选址规划方法，并提出了一种适用于V2G和G2V的自适应智能控制策略，以降低系统电压偏差和功率损耗[12]。Shi等人使用V2G技术来稳定可再生能源的间歇性，其中风能的不确定性和电动汽车充电状态使用稳健的最坏情况策略进行建模[41]。Sangswang 和 Konghirun 开发了一种家庭能源管理系统 （HEMS） 的调度策略，该系统集成了太阳能、储能和 V2G 功能，并使用实时定价和紧急减载来优化电动汽车和家用电池的充放电调度 [42]。

V2G调度受几个关键因素的影响，包括用户对里程的焦虑、电动汽车电池的磨损和状况[43]、用户收入预期以及用户的地理位置。确保从电动汽车到电网的电力传输得到有效控制和管理至关重要。

Wang和Wang研究了联网电动汽车的数量和电动汽车电池组特性对电网调峰和低谷调节的影响[10]。Farzin、Fotuhi-Firuzabad 和 Moeini-Aghtaie 考虑了电动汽车电池磨损对 V2G 调度的影响，并开发了电池退化成本模型来准确评估 V2G 的经济成本 [44]。Huang、Yang和Li介绍了电动汽车用户与电网之间交易价格波动的期权定价模型，并推导了V2G储备合作系数、交易保证金和电力合约价格之间的均衡关系[45]。埃尔多安、额尔登和基萨奇科格鲁提出了两级V2G放电调峰控制方案;它基于预测需求和电动汽车移动性模型确定V2G服务的调峰和持续时间，然后通过考虑实际电网负荷和并网电动汽车特性来动态调整电动汽车放电率[46]。Maeng等分析了电动汽车能源（混合动力或纯电动汽车）对电动汽车用户放电偏好的影响，研究了电动汽车剩余电量对用户放电意愿的影响，这与电动汽车用户里程焦虑对放电意愿的影响相似[47]。Jiao等人设计了一个模型来研究电动汽车用户的里程焦虑对V2G调度的影响，结果表明，通过提高平均运营成本和放电功率，可以缓解电动汽车用户的里程焦虑，进而影响充放电调度计划[48]。Wei、Yi 和 Yun 使用 Q-learning 来预测风力涡轮机的功率，并提出了一种基于强化学习的智能电网优化能量管理方法，以消散 RE，同时提供最佳电力调度并降低电网成本 [49]。Wang、Gao 和 Tang 创建了一个闪电般的规模转换模型来描述电动汽车用户的需求响应行为，以支持负载聚合器做出调度决策。该模型通过演化为混合双向方法，增强了初始单向尺度变换方法，包括深度尺度变换模式和广义尺度变换模式。这种创新方法促进了空间和时间观测尺度的同步和协同转换[50]。能源调度不仅仅发生在电动汽车和电网之间;许多研究深入研究了各种能源的相互调度[36,51,52,53,54]。

总体而言，未来V2G调度将更加智能，更加以用户为导向，专注于电池管理、定价策略和可再生能源集成，以满足不断变化的电网需求，提高电力系统的效率和可靠性。

为了便于电网功能，有功功率和无功功率的产生和消耗应不断平衡，以确保频率和电压的幅值接近其额定值。通过调节系统频率可以实现有功功率的有效控制，而无功功率的管理则取决于系统电压的控制。因此，频率和电压是衡量电网电能质量的重要因素。通过V2G模式实现整体电压和频率管理，将进一步提高有功和无功电能质量。

Frequency control is essential for power systems to maintain stable operation; it involves the synergistic action of all the generators and loads in the power system and can keep the grid frequency within the normal range. In this context, V2G technology has emerged as a promising resource that can be used to assist in the frequency regulation of power systems. EVs, when utilized as distributed storage devices, hold the potential to offer frequency regulation services owing to their ability to swiftly adjust their charging and discharging power.

Liu et al. proposed a V2G control strategy for EV aggregators based on frequency regulation capacity (FRC) and expected V2G (EV2G) power; this control strategy can shift the regulation task from the EV aggregator to the EV charging station to satisfy both the frequency regulation and the charging demand of the EV charging station [55]. Lam, Leung, and Li designed a queuing network-based EV aggregation model that can be used to estimate regulation-up and regulation-down capacity to help establish a regulation contract between the aggregator and the grid operator [56]. Chen et al. proposed a hierarchical V2G system communication architecture containing a smart V2G aggregator (SVA), and they designed a multilevel online V2G (MLOV) algorithm for hierarchical V2G scheduling to achieve a balance between service quality and computation time [57]. Wang, Wang, and Liu proposed a dynamic scheduling strategy for V2G frequency regulation capacity based on deep Q-learning to evaluate the hourly regulation capacity in real time for maximizing the revenue of frequency regulation services provided by battery swapping stations [58]. Alfaverh, Denaï, and Sun scheduled EV battery charging and discharging using the deep deterministic policy gradient (DDPG) to meet driving needs and to participate in frequency regulation, as well as to meet the driving needs of car owners and the interests of aggregators [59].

Frequency regulation of the grid by V2G is related to the state of EVs, the charging demand, and the changing grid frequency, which requires understanding the state of EVs on the grid; however, accessing these states involves issues such as communication security, user privacy security and control security between the information network and the energy network [60].

Voltage control is one of the key elements in the stable operation of power systems, and large generators and transformers are conventional means of controlling voltage. However, the operation and control of power systems have become more complex with the popularization of distributed energy resources [61,62].

In this context, V2G technology offers new possibilities for voltage control. With V2G, EVs can act as distributed energy storage devices to provide the necessary reactive power support to the grid [63], thus helping to maintain voltage stability [64]. García-Villalobos et al. solved voltage problems and load balancing issues in low-voltage distribution networks by optimizing the charging behavior of PEVs. This was conducted by executing intelligent charging algorithms through the processor unit built into the PEV and by developing an optimal charging strategy based on factors such as the price of electricity and battery life [65]. An intelligent sag control method that is independent of line parameters, and it devises a new strategy to enable PHEVs to participate in voltage and frequency control of islanded microgrids (MGs) without a communication link, thus maintaining MG stability more efficiently [66]. Huang proposed a day-ahead optimal control model based on three-phase power flow and sensitivity methods that could solve the overrun voltage problem and mitigate the neutral potential rise, thus improving the voltage regulation of residential grids [67].

As with the application of V2G and smart grid technology [69], EV grid technology is an important part of “smart grid technology” in the solving of the problems related to the large-scale development of EVs brought about by the grid load pressure [70]. Moreover, EVs, as mobile distributed energy storage units, are connected to the power grid for peak shaving, valley filling, and rotating standby in order to improve the flexibility of the power supply and the reliability and efficiency of energy utilization and to slow down investment in power grid construction.

Waraich et al. proposed an agent-based transport simulation framework tailored for PHEVs. Furthermore, they expanded the framework to encompass V2G capabilities and decentralized smart grid functionalities [71]. In a separate study, Jian et al. explored the concept of a household smart microgrid. They investigated how to potentially reduce load variance within the household microgrid by adjusting the charging patterns of family PHEVs. Their findings revealed a significant reduction in the variance of load power when such regulations were implemented [72]. Kennel, Goerges, and Liu proposed an energy management system designed for smart grids. This system incorporates load frequency control (LFC), economic efficiency optimization, and the integration of electric vehicles through a HiMPC approach. The simulation findings underscore the fact that electric vehicles can play a significant role in mitigating fluctuations in renewable energy generation, thereby contributing to grid stability [60]. Vachirasricirikul and Ngamroo presented the new coordinated V2G control and frequency controller for robust LFC in the smart grid system with wind power penetration. The simulation results clearly showcase the resilience and effectiveness of the suggested V2G control strategy and proportional–integral (PI) controllers in LFC, even when subjected to altered system parameters and diverse operating conditions [53]. Morais et al. considered the evaluation of the EV impact on the power demand curve under a smart grid environment and further addressed the impact of EVs on system operation costs and on the power demand curve for a distribution network with the deep penetration of distributed generation (DG) units [73]. In another study by Jian et al., they discussed a V2G implementation scenario within regional smart grids. They introduced a double-layer optimal charging (DLOC) strategy to address the computational challenges posed by large-scale PEVs and charging stations [74]. Liang and Zhuang believe that the forthcoming smart grid is anticipated to take the form of an interconnected network comprising small-scale, self-sustained microgrids. They also provided an overview of the latest advancements in stochastic modeling and optimization tools which was aimed at facilitating the planning, operation, and control of these microgrids [75].

The integration of EVs and smart grid technology continues to evolve, and numerous studies have been conducted to explore the potential benefits and challenges of this integration. These studies encompass various aspects, from enhancing the self-consumption of PV power to the development of comprehensive battery degradation models. Van der Kam and van Sark presented a model designed to investigate the augmentation of the self-consumption of PV power through the utilization of smart charging EVs with smart grid technology. The outcomes of their study conspicuously illustrate the advantages of employing smart charging and V2G technologies within a microgrid context [5]. López et al. proposed an optimization-driven model for executing load shifting within the framework of smart grids. They presented findings based on a test system derived from the IEEE 37-bus distribution grid. These results not only demonstrate the efficacy of their approach but also highlight the impact of hourly energy prices on the smoothing out of the load curve [76]. Xing et al. concentrated on acquiring load shifting services through the optimal scheduling of PEVs for charging and discharging within smart grids, adopting a decentralized approach. Their research culminated in the development of a decentralized algorithm rooted in iterative water-filling techniques [77]. Soares et al. presented a decision-making framework designed to aid virtual power plants (VPPs) in the management of smart grids characterized by a substantial presence of sensitive electricity loads. They further devised a two-stage optimization algorithm, leveraging the weighted sum methodology, to facilitate this management process [78]. Uddin et al. developed a comprehensive battery degradation model based on long-term ageing data and integrated a comprehensive battery ageing model into a smart grid algorithm. The result demonstrated that linking an EV to this smart grid system can effectively meet the power network’s demand, particularly as the proportion of clean renewable energy in the system increases [79]. Dileep gave an overview of the evolution of the smart grid and explained various smart grid technologies, like smart meters, smart sensors, V2G, and PHEV, and their application in the smart grid [35]. The integration of EVs and smart grid technology holds significant promise for the future of sustainable energy. Through various models, algorithms, and frameworks, researchers are paving the way for a more efficient, reliable, and environmentally friendly power grid system.

The integration of communication networks with battery-operated vehicles and the electrical grid is a transformative step in the evolution of energy systems. This integration streamlines the coordination of the electric load, but it introduces challenges, especially with regard to security and privacy.

Communication networks bridge the gap between user battery-operated vehicles and the electrical grid, enhance the coordination of electric loads, and improve energy efficiency and reliability. Zhang et al. discussed the V2G network architecture and described the different security challenges during V2G power and communication interactions. A new context-aware authentication solution was also proposed for this problem [80]. Saxena et al. described the challenges of the security and privacy in smart V2G networks and proposed a cybersecurity architecture that attempted to address the problem; this architecture encompassed anonymous authentication, blind signatures, fine-grained access control, and payment system security [81]. Tao et al. proposed a hybrid computing model based on fog and cloud computing for V2G networks in 5G networks. This hybrid computing model can respond more flexibly and in a more timely manner to EV mobility; it can provide auxiliary power services for renewable energy sources in V2G systems, as well as manage and monitor power usage [82]. A privacy-friendly and efficient secure communication (PESC) framework was proposed by He, Chan, and Guizani [83]. It used group signatures and public key cryptography for access control from each EV to the aggregator and established shared keys (with the Diffie–Hellman key exchange algorithm), which could reduce the communication and computation overhead while ensuring privacy and security. Ahmed et al. designed a signature-encrypted, privacy-preserving authentication key agreement scheme that enables all participants to verify each other; it uses a one-way hash function to improve verification efficiency [84]. Umoren, Shakir, and Tabassum proposed a strategy to balance the efficiency of both spectral considerations and cost by expressing resource efficiency as a weighting factor, which improved the efficiency of V2G wireless communication networks [85]. Pokhrel and Hossain developed a novel, adaptive demand-side energy management framework for the wireless charging of V2G systems that employed privacy-preserving techniques based on federated learning [86]. Hossain, Pokhrel, and Vu proposed a demand-side energy management approach based on reinforcement learning using rechargeable batteries for V2G cost-friendly privacy, efficient scheduling, and accurate billing [87].