Constructing a Locational Marginal Emission Model Framework for Electrical Distribution Grids: Application to the Province of Alberta Grid for Dual Cost-Emissions Optimization.

广泛的电气化在加拿大“2030年减排计划”的成功中发挥着至关重要的作用。然而，为了实现电气化的潜力，必须在保持能源供应的稳定性和可靠性的同时，将电力部门的温室气体（GHG）排放尽可能减少到接近零。这笔赠款属于后者。电网运行中温室气体的成功减排取决于几个因素。电网运行是一个复杂的异构过程组合，其时间跨度从几年到几毫秒不等。从政策制定者到最终消费者，许多利益相关者影响着运营。虽然可以找到大量的现有技术来讨论网格的最佳操作，但一个主要挑战仍然几乎没有触及；目前还不存在将系统的实时边际排放纳入利益相关者决策的框架。发电机的本地直接测量无法提供最佳实时运行所必需的位置边缘（LM）信息。对“实时”的强调是针对可再生能源和电力运输等随机因素的高度渗透所导致的现代电网快速变化的运行轨迹。LM的重点是使分布式能源和需求的积极参与，以尽量减少温室气体排放。这笔拨款将提出一个模拟和估计地区边际排放的框架，并将其应用于加拿大阿尔伯塔省。这项多学科研究将使用机器学习方法来绘制实时和提前一天的LM排放图。该框架将用于在考虑电网运行约束的情况下开发双重成本排放最小化的最佳调度工具。目标是让分布式能源和最终用户了解其对温室气体排放的影响，并积极参与减排。这项工作将与工业界合作完成，并为电网运营商提供指导，以尽量减少温室气体排放。人们高度期待这些发现的商业化。

Widespread electrification plays a vital role in Canada's "2030 Emissions Reduction Plan" success. However, for electrification to deliver its potential, it must be followed by a reduction in the electricity sector's greenhouse gas (GHG) emissions to as close to zero as possible while maintaining the stability and reliability of the energy supply. This grant pertains to the latter. Successful GHG reduction in electrical grid operation depends on several factors. Electricity grid operation is a complex combination of heterogenous processes spanning a time frame of years ahead to a few milliseconds. Many stakeholders influence the operation ranging from policymakers to end consumers. While a plethora of prior art can be found discussing the grids' optimal operation, one main challenge remains barely touched; there exists no framework to incorporate real-time marginal emissions of the system into decision-making by stakeholders. Local direct measurements from generators provide no locational marginal (LM) information that is necessary for optimal real-time operation. The emphasis on "real-time" deals with the fast varying operation trajectory of modern electrical grids, caused by high penetration of stochastic agents such as renewable energies and electric transportation. The emphasis on LM is to enable active participation of distributed energy resources and demand to minimize GHG emissions.This grant will propose a framework for modeling and estimating the locational marginal emissions, and apply it to the province of Alberta, Canada. The multidisciplinary research will use machine learning methods to map real-time and day-ahead LM emissions. The framework will be used to develop an optimal dispatch tool for dual cost-emission minimization while considering grid operational constraints. The objective is for distributed energy resources and end users to understand their impact on GHG emissions and to actively participate in emissions reductions. The work will be done in partnership with industry and provides guidance to grid operators for minimizing GHG emissions. Commercialization of the findings is highly anticipated.