Collaborative Research: CISE-MSI: DP: CCF: SHF: MSI/HSI Research Capacity Building via Secure and Efficient Hardware Implementation of Cellular Computational Networks

该奖项全部或部分由《2021年美国救援计划法案》（公法117-2）资助。随着风能和太阳能等可再生能源的使用不断增加，需要分布式人工智能（AI）来综合大量预测使用指标，如天气数据和物联网（IoT）传感器数据，以使电网能够在与可再生能源相关的高水平可变性和不确定性下继续可靠运行。由于这需要实时处理，因此需要一个安全高效的硬件平台；仅靠人工智能软件是不够的。细胞计算网络（CCN）是一种分布式人工智能框架，具有大脑启发的神经网络架构，适用于关键的网络系统，如电网。因此，利用安全高效的CCN硬件实现电力系统应用将加速运行，在不影响准确性的情况下实现具有代表性的大规模网络的实时性能保证，并同时提供对网络物理系统攻击的弹性，从而增强电力系统的可持续和安全运行。该项目使用可重构的现场可编程门阵列开发CCN单元和整体CCN系统的同步逻辑和异步逻辑硬件实现，并探索应用于CCN的近似计算机会。由此产生的CCN硬件系统将通过集成到克莱姆森大学的各种实时电力和智能系统（RTPIS）实验室测试平台进行测试，包括电力系统变量的广域预测状态估计，解决动态功率流，以及时空风速/功率，太阳辐照度/功率和建筑物/房间能耗的预测。此外，该项目还与德克萨斯州农工大学金斯维尔分校（TAMUK）的少数民族/西班牙裔服务机构合作，与克莱姆森大学合作，让更多的TAMUK计算机科学教师参与RTPIS实验室的相关研究，并为TAMUK的优秀西班牙裔学生建立攻读计算机工程或计算机科学博士学位的渠道。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 award is funded in whole or in part under the American Rescue Plan Act of 2021 (Public Law 117-2).As use of renewable energy sources, such as wind and solar power, continue to increase, distributed Artificial Intelligence (AI) is needed to synthesize the large amounts of predictive use indicators, such as weather data and Internet of Things (IoT) sensor data, in order to allow the electric power grid to continue to operate reliably with the high levels of variability and uncertainty associated with renewable energy sources. Since this requires real-time processing, a secure and efficient hardware platform is needed; AI software alone is not sufficient. The Cellular Computational Network (CCN) is a distributed AI framework, with a brain-inspired neural network architecture, which is suitable for critical networked systems, such as the electric power grid. Hence, utilizing secure and efficient CCN hardware implementations for power system applications will accelerate operations to achieve a real-time performance guarantee on representative large-scale networks without compromising accuracy, and will simultaneously provide resiliency to cyber-physical system attacks, thus enhancing sustainable and secure power system operation.This project develops both synchronous logic and asynchronous logic hardware implementations of CCN cells and overall CCN systems using reconfigurable Field Programmable Gate Arrays, and explores approximate computing opportunities for application to CCNs. The resulting CCN hardware systems will be tested via integration into Clemson University’s various Real-Time Power and Intelligent Systems (RTPIS) Laboratory testbeds, including for wide area predictive state estimation of power system variables, solving dynamic power flows, and predictions of spatial-temporal wind speed/power, solar irradiance/power, and energy consumption of buildings/rooms. Furthermore, this project partners a Minority/Hispanic Serving Institution, Texas A&M University – Kingsville (TAMUK), with Clemson University to involve many more TAMUK Computer Science faculty with RTPIS Lab related research, and establishes a pipeline of high-performing Hispanic students from TAMUK to pursue Computer Engineering or Computer Science PhD degrees.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.