RIPS Type 2: Collaborative Research: Towards resilient computational models of electricity-gas ICI

通过使用天然气发电厂，电力行业是天然气(NG)消费增长最快的行业之一。天然气发电已经在美国许多地区的电力组合中占据了很大比例，由于前者具有快速的按需响应能力，而后者具有间歇性和不确定性的特点，随着可再生能源的日益普及，天然气发电正在进一步增长。因此，天然气和电力网络正变得越来越相互联系和相互依赖。因此，电网运营商和天然气供应商之间加强协调和信息共享，对于拥有一个弹性的、相互依赖的电力和天然气关键基础设施变得非常必要。该提案旨在创建此基础设施的计算模型，获得对现有基础设施的分析洞察，以响应系统和政策的变化，并为具有更高弹性的协调基础设施设计指导方针。这些模型尽可能基于工业合作伙伴的真实数据，并纳入了新英格兰地区电力和天然气网络的比例版本。实现这些目标的具体工具包括安全受限的机组组合，这是一种强大的工具，可以为给定的网络拓扑和策略提供最佳的发电调度；动态的市场机制和混合网络科学，允许分析基础设施的不确定性；以及基于网络安全的恢复能力工具，用于恶意和非恶意事件。

One of the fastest growing consumers of Natural Gas (NG) is the electricity sector, through the use of NG-fired power plants. Already a large portion of the electricity portfolio mix in many regions in the US, NG-fired power generation is increasing even further with growing penetration of renewable energy due to the former's fast, on-demand response capabilities, and latter's characteristics of intermittency and uncertainty. As a result, NG and electricity networks are getting increasingly linked and interdependent. Tighter coordination and information sharing between electric grid operators and NG suppliers is therefore becoming highly necessary to have a resilient interdependent critical infrastructure of electricity and NG. This proposal seeks to create computational models of this infrastructure, gain analytical insight into the existing infrastructure in response to system and policy changes, and design guidelines for a coordinated infrastructure with improved resilience. These models are based, wherever possible, on real data from industrial partners and incorporate scaled versions of the electrical and NG networks of the New England area. Specific tools to reach these objectives include security constrained unit commitment, a powerful tool that provides optimal dispatch of power generation for a given network topology and policies, dynamic market mechanisms and hybrid network science, that permit analysis of uncertainty on the infrastructure, and cyber-security based resiliency tools for malicious and non-malicious events.