CRII: CPS: Internet-Inspired Autonomous Electric Vehicle (EV) Charging

电动汽车正在改变现代交通和能源系统。随着电动汽车市场的不断发展，电动汽车大量进入电网，由于电力峰值负荷与电动汽车充电的重合，将对电网产生不利影响。这种影响将包括峰值负载增加和电压下降，这将要求对网络进行过度投资，从而导致社会的总体成本高。因此，需要开发先进实用的车辆、充电站和电网之间的交互方式。基于互联网的电动汽车充电控制为电动汽车充电控制研究提供了新的范式转换。本项目旨在建立一种新的自主、公平的电动汽车充电控制技术，并可大规模应用于实践。所提出的想法的意义在于努力将利用端到端互联网网络拥塞控制的现有技术与分散的自动电动汽车充电结合在一起的网络物理系统。这项工作是跨学科的，汇集了计算机科学和电力工程的想法，以支持将电动汽车大规模集成到传统的电力系统中。提出的想法可以影响所有并网电力电子系统的控制方式。在这项工作中开发的算法可用于加速太阳能光伏系统等低碳能源的整合。本研究通过采用互联网中一些部署良好的端到端拥塞控制技术，开发了一系列新的电动汽车充电算法。研究目标包括：(i)通过广泛的现场数据收集、建模和仿真来调查配电网拥堵，以了解和提取电动汽车充电对终端节点电压和频率特征的影响；（ii）设计完全分散和本地化的自动工作的电动汽车充电算法；(3)根据配电网的潮流拥塞情况，研究电动汽车充电率的自适应增减，以实现社区内电动汽车的比例或最大最小公平充电；(4)探索可使用多个电动汽车测量的混合充电架构。该项目将开发的网络物理系统将帮助研究人员了解和解决与大规模电动汽车集成相关的问题。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Electric vehicles (EVs) are transforming the modern transportation and energy systems. With a growing EV market, the mass penetration of EVs into the utility grid will result in detrimental effects due to coincidence between utility peak power loading and EV charging. This impact will include increased peak loading and voltage drops that will call for over-investments in the network resulting in an overall high cost for the society. Therefore, advanced and practical interaction methods between vehicles, charging stations, and the utility grid need to be developed. Internet-inspired EV charge control offers a new paradigm-shift for EV charge control studies. This project aims to build a new autonomous and fair EV charge control technique that can penetrate into practice at large scales. The significance of the proposed idea lies in the effort to bring together a cyber-physical system that utilizes the prior art in end-to-end internet network congestion control with decentralized autonomous EV charging. The effort is inter-disciplinary and brings together ideas from computer science as well as power engineering to support mass integration of EVs into the conventional electric energy system. The proposed idea can impact the way all grid-connected power electronics systems are controlled. The algorithms developed in this work can be used to accelerate the integration of low-carbon energy sources such as solar photovoltaic systems. The proposed research develops a new family of EV charging algorithms by adopting some of the well-deployed end-to-end congestion control techniques in the Internet. The research objectives include (i) investigating distribution grid congestion through extensive data collection from field, modeling, and simulations to understand and extract the impacts of EV charging on voltage and frequency signatures at the end-nodes, (ii) design of entirely decentralized and localized EV charging algorithms that work autonomously, (iii) investigation of adaptive increase and decrease of EV charging rate according to the power flow congestion on the distribution network to attain proportional or max-min fair charging of EVs in a neighborhood, and (iv) exploration of hybrid charging architectures where more than one EV measurement can be utilized. The cyber-physical system that will be developed with this project will help the researchers understand and resolve issues relating to large scale electric vehicle integration.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.