AitF: Collaborative Research: Algorithms and Mechanisms for the Distribution Grid

配电网-将电力分配给终端消费者的低压线路网络-即将经历重大变革。越来越多的消费者正在成为“产消者”，即同时消费和生产电力，他们在自家屋顶安装太阳能电池板，并购买可用于储能的电动汽车。传统上，配电网中的电力是单向流动的：从变电站到终端消费者。在产消者的新世界里，配电网需要适应两个方向的流量。这对现有的电线和Transformer以可接受的质量水平服务负载的能力提出了挑战。此外，生产者消费者人数的增加导致需求预测的不确定性大大增加，随着生产者消费者人数的进一步增加，这种不确定性可能被证明是不可持续的，并最终威胁到公用事业公司的运营和商业生存能力。这项研究推进了计算和经济学领域以及电力系统领域，通过提供新的算法和机制设计问题配方和技术，以及可以改进配电网规划和运行的解决方案。就更广泛的影响而言，这项研究有可能改善电网的各个方面，例如规划可再生能源资源的整合，减轻与可再生能源变化相关的风险，更好地管理面临战略性生产者的拥堵，并最终提供更可靠，更具成本效益和高效的电网运营。该项目的目标是帮助配电网及其参与者从目前主要为传统消费者服务的功能过渡到需要可持续地整合生产消费者，可再生能源和分布式能源的未来电网：（一）开发简化的数学模型来解决组合和激励问题，使未来的电网能够维持可再生能源和分布式能源的大规模增长能源资源;（2）设计新的算法和机制设计方法，以减少拥塞，提高电网的投资和运营效率，同时进一步提高其可靠性;（3）与配电公司合作，提供真实世界的模型和运营建议，同时促进研究成果转化为实践。 该项目的范围包括：㈠配电网络重新配置; ㈡配电系统升级; ㈢供求平衡的市场机制。 研究方法主要有近似算法和机构设计方法，如子模块优化、随机组合优化、无政府代价分析等。

The electricity distribution grid --the low-voltage line networks that distribute power to end consumers--is about to undergo a major transformation. More and more consumers are becoming "prosumers", namely consumers and producers of power at the same time, by installing solar panels (PVs) on their home roofs and by purchasing electric vehicles, which may be used for energy storage. Traditionally, power in distribution grids has flown one way: from the substation to the end consumer. In the new world of prosumers, distribution grids need to accommodate flow in both directions. This is challenging the existing wire and transformer abilities to serve load at acceptable quality levels. Moreover, the increasing number of prosumers is resulting in dramatically higher uncertainty in demand forecasting, which, with further prosumer increase, may prove unsustainable and ultimately threaten the utility companies' operation and business viability. This research advances the fields of computation and economics as well as the power systems domain, by contributing with novel algorithmic and mechanism design problem formulations and techniques, and with solutions that can enable improved distribution grid planning and operation. In terms of broader impact, this research has the potential to improve aspects of the grid such as planning for increased integration of renewable energy resources, mitigation of risks associated with variability of renewables, better management of congestion in the face of strategic prosumers and ultimately provide for more reliable, cost effective and efficient operation of the grid. The goal of this project is to help the distribution grid and its participants transition from its current functionality of serving mostly traditional consumers, to the future grid that needs to sustainably integrate prosumers, renewables and distributed energy resources, via: (1) Developing simplified mathematical models to solve combinatorial and incentive problems that will enable the future power grid to sustain massive growth in renewables and distributed energy resources; (2) Designing new algorithmic and mechanism design approaches that reduce congestion, and improve the investment and operational efficiency of the grid while further enhancing its reliability; (3) Working with Distribution Utilities for real-world models and operational advice, while facilitating the transfer of research findings to practice. The scope of the project includes (i) distribution network reconfiguration, (ii) distribution system upgrades and (iii) market mechanisms for supply-demand balancing. The research relies on methods from approximation algorithms and mechanism design, such as submodular optimization, stochastic combinatorial optimization, and price of anarchy analysis.