Collaborative Research: Electronic Analog & Hybrid Computing for Power & Energy Systems

该项目将开发旨在使电网更加可靠、高效和可持续的计算方法。在过去的几十年里，随着太阳能、风能和电动汽车充电等分布式能源的迅速采用，电网发生了巨大的变化。这些趋势在网格的优化、控制和模拟方面引入了大量的新问题，这些问题对传统的数字计算方法提出了挑战。项目团队将开发新的工具，用于建模、分析、设计和对替代计算方法（即电子模拟和混合计算）进行基准测试，最终目的是促进关键算法和工具的实施，这些算法和工具在当前的数字计算范式下是不实际或不可行的。计算技术的潜在进步将使其他各种工业部门受益，例如运输和制造业，并广泛推动计算和半导体技术的进步。我们将探讨与电子模拟和混合计算系统相关的一些基本问题，包括对其设计、稳定性、性能和鲁棒性的统一和系统理论评估，这些问题的动机是它们在未来电力和能源系统应用中的潜在作用。我们为我们的项目提出了三种方法：i)建模和分析，ii)设计自动化，以及iii)性能评估。通过我们的研究计划，我们将开发全面的建模和分析方法，以产生准确、稳定和可预测的电子模拟和混合计算机。我们将利用这些见解来开发一个开源设计自动化库和工具链，使用户能够轻松地将各种控制器，优化器或模拟器合成为模拟或混合计算机。最后，我们将开发一个第一性原理性能评估工具，该工具将阐明利用模拟或混合计算机的好处，确定其好处可以充分实现的特定应用，并使模拟，混合和数字计算机之间的原则和定量比较得以实现。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。

This project will develop computing methods that aim to make the electric grid more reliable, efficient, and sustainable. The grid has evolved tremendously in the past decades with the rapid adoption of distributed energy resources, such as solar and wind, and electric vehicle charging. These trends have introduced a plethora of new problems in optimization, control, and simulation of the grid that have posed challenges to conventional computing approaches, which are primarily digital in nature. The project team will develop new tools for the modeling, analysis, design, and benchmarking of alternative computing approaches, namely electronic analog and hybrid computing, with the ultimate aim of facilitating the implementation of critical algorithms and tools for the future electric grid that are not practical or feasible with the prevailing digital computing paradigm. The potential advancements in computing technologies will benefit a variety of other industrial sectors, such as transportation and manufacturing, and drive advances in computing and semiconductor technology broadly. We will explore a number of fundamental questions related to electronic analog and hybrid computing systems, including a unified and system-theoretic assessment of their design, stability, performance, and robustness, motivated by their potential role in enabling applications for future power and energy systems. We propose three approaches to our project: i) modeling and analysis, ii) design automation, and iii) performance assessment. Through our research program, we will develop comprehensive modeling and analysis methods that will yield accurate, stable, and predictable electronic analog and hybrid computers. We will leverage these insights to develop an open-source design automation library and toolchain that will enable users to easily synthesize various controllers, optimizers, or simulators as an analog or hybrid computer. Finally, we will develop a first-principles performance assessment tool that will clarify the benefits of utilizing analog or hybrid computers, identify the particular applications for which their benefits can be fully realized, and enable principled and quantitative comparisons among analog, hybrid, and digital computers to be made.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.