CAREER: Computation-efficient Resolution for Low-Carbon Grids with Renewables and Energy Storage

为了实现无碳清洁电力系统的愿景，可再生能源和储能资源（esr）在可靠和有弹性的电网运行中发挥着至关重要的作用。NSF CAREER项目旨在为可再生能源和esr的低碳电网运行开发新的建模和优化方法。该项目将带来革命性的变化，使电网运营商能够更有效地利用esr，实现可持续和高效的能源未来。这将通过ESR建模、新型配方收紧技术和创新的优化方法来实现。其智力优势包括考虑动态荷电状态（SOC）限制和可再生能源不确定性的新颖ESR市场参与模型；一种创新的基于机器学习的公式收紧方法提高计算效率以及一种基于有序优化的指数复杂度降低优化方法，以有效地管理网格运行中的大量esr。该项目的更广泛影响包括为独立系统运营商（ISO）、区域传输运营商（RTO）和软件开发人员开发培训模块，以及为研究生和本科生开发培训模块，重点是在STEM学科的早期阶段吸引女性和代表性不足的学生；以及向K-12学生提供更广泛的推广活动。该项目解决了可再生能源和ESR在低碳电网运行中的几个技术挑战，包括ESR市场参与者模型、日前调度与实时调度不一致，以及ESR双向充放电和时间耦合SOC等独特特性带来的计算困难。该项目的技术组成部分包括建立各种ESR参与者模型，从自调度到考虑动态SOC限制的iso / rto完全管理；一种面向凸壳的基于深度学习的计算效益公式收紧方法的开发基于有序优化的指数复杂度降低优化方法，有效解决具有大量esr的电网运行问题。由此产生的具有即插即用功能的模型和方法可以集成到iso /TROs供应商开发的现有平台中，以有效利用esr，从而获得经济和环境效益。该项目的成果还将促进与可持续和高效的能源未来有关的教育和外联活动。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。

To realize the vision of carbon-free clean power systems, renewable energy and Energy Storage Resources (ESRs) play critical roles in reliable and resilient grid operation. This NSF CAREER project aims to develop novel modeling and optimization approaches for low-carbon grid operation with renewable energy and ESRs. The project will bring transformative change to enable power grid operators to leverage ESRs more efficiently for a sustainable and efficient energy future. This will be achieved by ESR modeling, novel formulation tightening techniques, and innovative optimization methods. The intellectual merits include novel ESR market participation models considering dynamic State-of-Charge (SOC) limits and renewable uncertainties; an innovative machine learning-based formulation tightening approach to improve computational efficiency; and an Ordinal Optimization-based optimization approach for exponential complexity reduction to efficiently manage a large number of ESRs in grid operations. The broader impacts of the project include the development of a training module for Independent System Operators (ISO), Regional Transmission Operators (RTO), and software developers, and a training module for graduate and undergraduate students, focusing on engaging women and underrepresented students at an early stage in STEM disciplines; and broader outreach activities to K-12 students.The project addresses several technical challenges in low-carbon grid operation with renewable energy and ESRs including ESR market participant models, inconsistency between day-ahead scheduling and real-time dispatch, and computational difficulty caused by unique features of ESRs such as bidirectional discharge and charge operations and time-coupling SOC. The technical components of the project include the establishment of various ESR participant models from self-scheduling to being fully managed by ISOs/RTOs considering dynamic SOC limits; development of a novel convex hull-oriented deep learning-based formulation tightening approach for computational benefits; and an Ordinal Optimization-based optimization approach for exponential complexity reduction to efficiently solve grid operation problems with a large number of ESRs. The resulting models and methods with plug-and-play capabilities can be integrated into ISOs/TROs’ existing platforms developed by vendors for efficient utilization of ESRs, leading to economic and environmental benefits. The results of the project will also facilitate education and outreach activities related to ESRs for a sustainable and efficient energy future.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.