CPS: Small: Fusion of Sensory Data and Expansivity of System Dynamics for Detection and Separation of Signature Anomaly in Energy CPS Wide-Area Monitoring and Control

提出的研究重点是用于网络物理系统（CPS）实时监测和控制的传感数据异常检测和分离领域的创新。本文以大型电网为例，对我国重要的基础设施进行了研究。来自网络攻击的传感器数据的虚假或恶意注入问题非常重要，因为它会严重危及电网的监测和稳定控制。这可能导致全系统停电，并使我们的经济损失数十亿美元。该项目旨在通过利用系统的扩展动态行为来区分干扰和数据异常来解决这个问题。为此，本研究的目的是弥合奇异值摄动理论和主成分分析（PCA）领域发展之间的差距——传统上侧重于CPS的“信号”方面，与CPS的“系统”方面的内在属性。感官数据与物理系统动态特性的融合将有助于获得对网络和物理层之间耦合的基本洞察，并利用这些知识来检测和分离源自网络攻击的虚假信号或恶意数据操纵。该项目的成功结论将量化相量测量单元特征异常对高维和低维子空间主成分的影响。此外，将探讨鲁棒PCA的新兴概念，以分离异常签名并校正数据以用于实时应用。在当今的“工业互联网”世界中，传感数据被用于系统健康监测和控制，这项工作有可能在理解数据质量和异常方面取得重大进展。因此，该项目的成功完成可以帮助保护关键基础设施免受网络攻击，并促进改进的系统诊断，减少停机时间，提供更好的服务和更高的弹性。拟议的研究可能会使CPS广泛受益，包括过程控制、石油和天然气、能源，以及可能涉及机器人的其他系统，甚至是未来使用自动驾驶汽车的运输系统。本项目将把其研究成果与电力系统课程模块相结合，以扩大大学生对能源CPS新发展的经验。此外，该项目将把拟议的研究纳入iTech：青少年暑期科技夏令营，这是一个为期一周的免费互动日营，旨在向高中生（9 -12年级）介绍信息技术。在这个项目中创建和开发的所有材料，包括软件数据、主要研究数据和元数据，以及教学材料，将在以下网站上以开源的方式公开提供：https://sites.psu.edu/nilanjan/。为加快研究成果的传播，已发表的论文将以PDF格式在本网站提供。

The proposed research focuses on innovation in the area of detection and separation of anomaly in sensory data used for real time monitoring and control of cyber-physical systems (CPS). In this work, bulk power grid is considered as an example CPS, which is a critical infrastructure of our nation. The problem of spurious or maliciously injected sensor data originating from cyber-attacks is very important, because it can seriously jeopardize the monitoring and stabilization controls of power grids. This can lead to system-wide blackouts and cost our economy billions of dollars. This project aims to solve this problem by leveraging the system's expansive dynamic behavior to distinguish disturbances from data anomalies. To that end, the aim of this research is to bridge the gap between the developments in the area of singular value perturbation theory and Principal Component Analysis (PCA) -- traditionally focused on the 'signals' side of the CPS, with the intrinsic properties from the 'systems' side of the CPS. Fusion of sensory data with the dynamic properties of the physical system will help in gaining fundamental insight into the coupling between the cyber and the physical layer and use this knowledge to detect and separate spurious signals or malicious data manipulations originating from cyber-attacks. A successful conclusion of this project will quantify the effect of Phasor Measurement Unit signature anomalies on Principal Components in the higher versus lower dimensional subspaces. In addition, emerging concepts of Robust PCA will be explored to separate the anomalous signatures and correct the data for real-time application. In today's world of 'Industrial Internet' where sensory data is being utilized for system health monitoring and controls, this work has the potential to make significant strides in understanding data quality and anomaly. Therefore, a successful completion of this project can help protect critical infrastructures from cyber-attacks and facilitate improved system diagnosis, lower downtime, better service, and higher resiliency. The proposed research can potentially benefit a wide range of CPS including process control, oil and gas, energy, and probably other systems involving robots or even future transportation systems employing autonomous vehicles. This project will integrate its research outcome with power systems course modules to broaden the undergraduate experience on the newer developments in energy CPS. In addition, this project will integrate the proposed research into iTech: Summer Technology Camp for Teens, which is a free week-long interactive day camp, designed to introduce high school students (9th -12th grades) to information technology. All materials created and developed in this project including software data, primary research data and metadata, and teaching material will be publicly available on an open source basis at the following site: https://sites.psu.edu/nilanjan/. For faster dissemination of the research results, published papers will be made available in PDF format in this site.