PAPMSC: Power-Aware Performance Management Approach for Virtualized Web Servers via Stochastic Control

As green computing is becoming a popular computing paradigm, the performance of energy-efficient data center becomes increasingly important. This paper proposes power-aware performance management via stochastic control method (PAPMSC), a novel stochastic control approach for virtualized web servers. It addresses the instability and inefficiency issues due to dynamic web workloads. It features a coordinated control architecture that optimizes the resource allocation and minimizes the overall power consumption while guaranteeing the service level agreements (SLAs). More specifically, due to the interference effect among the co-located virtualized web servers and time-varying workloads, the relationship between the hardware resource assignment to different virtual servers and the web applications’ performance is considered as a coupled Multi-Input-Multi-Output (MIMO) system and formulated as a robust optimization problem. We propose a constrained stochastic linear-quadratic controller (cSLQC) to solve the problem by minimizing the quadratic cost function subject to constraints on resource allocation and applications’ performance. Furthermore, a proportional controller is integrated to enhance system stability. In the second layer, we dynamically manipulate the physical frequency for power efficiency using an adaptive linear quadratic regulator (ALQR). Experiments on our testbed server with a variety of workload patterns demonstrate that the proposed control solution significantly outperforms existing solutions in terms of effectiveness and robustness.     

基于蠕虫传播和FDI的电力信息物理协同攻击策略EI北大核心CSCD

随着信息技术与现代电力系统的结合日趋紧密,通信系统异常和网络攻击均可能影响到电力系统的安全稳定运行.为了研究工控蠕虫病毒对电网带来的安全隐患,本文首次建立了基于马尔科夫决策过程(Markov decision process,MDP)的电力信息物理系统跨空间协同攻击模型,该模型同时考虑通信设备漏洞被利用的难易程度为代价以及对电力网络的破坏程度为收益两方面因素,能够更有效地识别系统潜在风险.其次,采用Q学习算法求解在该模型下的最优攻击策略,并依据电力系统状态估计的误差值来评定该攻击行为对电力系统造成的破坏程度.最后,本文在通信8节点-电力14节点的耦合系统上进行联合仿真,对比结果表明相较单一攻击方式,协同攻击对电网的破坏程度更大.与传统的不考虑通信网络的电力层攻击研究相比,本模型辨识出的薄弱节点也考虑了信息层的关键节点的影响,对防御资源的分配有指导作用.     

具有随机网络攻击的切换信息物理系统的事件触发控制（英文）

本文研究了随机网络攻击下切换信息物理系统的事件触发控制问题.将信息物理系统描述为一种切换线性系统形式.引入事件触发机制来节省系统资源和减轻网络负载,当误差超过给定阈值时传感器中的采样数据才通过通信网络传输到控制器中.考虑在传感器与控制器的通信网络中受到两种不同特征的随机网络攻击.在网络攻击和所设计的事件触发控制器下,建立了切换随机信息物理系统模型.利用模态依赖平均驻留时间方法构建了相应的切换信号.在设计的事件触发控制器和模态依赖平均驻留时间切换信号下实现了系统的均方指数稳定性,并给出了控制器增益.最后,通过实例验证了所得理论结果的有效性.     

分布式电网CPS 系统数据攻击下的状态估计

电力物理网络通过构建信息网络进行优化调控并构成信息物理融合系统, 实现大规模分布式系统的优化控制, 随之而来的问题是病毒、黑客入侵、拒绝服务等来自信息网络的威胁, 导致物理系统恶意破坏. 鉴于此, 以攻击可检测为前提, 建立攻击信号下的电力系统分布式动态模型, 设计动态状态估计器检测受攻击的信号, 并估计其原始信号. 最后通过3 机9 节点分布式电网系统仿真实验验证了所设计的状态估计器对于数据攻击检测的有效性.

     

面向电力信息物理系统的虚假数据注入攻击研究综述

随着电力信息通信技术的发展与应用,电力流与信息流深度融合,共同实现对系统的全景状态感知与控制决策,电力系统转变成典型的信息物理系统（Cyber physical system,CPS）.开放的通信环境与复杂的信息物理耦合交互过程,使得信息安全风险成为影响电力系统安全稳定运行的重要因素.其中,虚假数据注入攻击（False data injection attack,FDIA）通过破坏网络数据完整性以干扰控制决策,是一种典型的网络攻击方式.本文针对面向电力CPS的虚假数据注入的攻击过程和防御手段进行了分析与总结.从攻击者视角分析了FDIA的攻击目标、策略及后果;从防御者视角总结了保护与检测环节中的各类方法;最后基于联合仿真技术,提出了针对虚假数据攻防过程建模和评估的电力CPS联合攻防平台.     

MAS Based Distributed Automatic Generation Control for Cyber-Physical Microgrid System

The microgrid is a typical cyber-physical microgrid system (CPMS). The physical unconventional distributed generators (DGs) are intermittent and inverter-interfaced which makes them very different to control. The cyber components, such as the embedded computer and communication network, are equipped with DGs, to process and transmit the necessary information for the controllers. In order to ensure system-wide observability, controllability and stabilization for the microgrid, the cyber and physical component need to be integrated. For the physical component of CPMS, the droop-control method is popular as it can be applied in both modes of operation to improve the grid transient performance. Traditional droop control methods have the drawback of the inherent trade-off between power sharing and voltage and frequency regulation. In this paper, the global information (such as the average voltage and the output active power of the microgrid and so on) are acquired distributedly based on multi-agent system (MAS). Based on the global information from cyber components of CPMS, automatic generation control (AGC) and automatic voltage control (AVC) are proposed to deal with the drawback of traditional droop control. Simulation studies in PSCAD demonstrate the effectiveness of the proposed control methods.      

Active Disturbance Rejection Control in fully distributed Automatic Generation Control with co-simulation of communication delay

This paper compares Active Disturbance Rejection Control (ADRC), Proportional–Integral (PI) and ultra-local adaptive controller (also labeled intelligent-PID) when applied to Automatic Generation Control (AGC) for a multi-area power system with co-simulation of the communication system. The co-simulation platform PiccSIM integrates Simulink/Matlab and Network Simulator Ver. 2 (NS2) with real-time simulation data exchanged through Local Area Network (LAN). Instead of the traditional control center, automatic generation controller is embedded in each participating generation unit locally to avoid communication delay for the control signals. This new control scheme is referred to as fully distributed AGC. The performance of the proposed scheme is tested on a nonlinear model of New England (NE) 39-bus system with the co-simulated tie-line power flow communication delays. The simulation results present very promising performance and good robustness of ADRC in the presence of communication delays, suggesting that the stability of the overall cyber–physical system is enhanced significantly.     

基于数据特征融合的管网信息物理异常诊断方法

随着管网物理空间和信息网络的深度融合,系统面临着物理和信息空间异常带来的运行风险.本文根据管网系统数据量大、耦合性强的特点,提出一种基于数据特征融合的信息物理异常诊断方法.首先通过站场信息数据构建信息增维矩阵并且通过矩阵预分析实现信息传输中断异常的判断.然后基于不同站场信息构建的信息增维协方差矩阵,通过矩阵特征值分布的变化情况对物理异常以及信息传输错误异常进行区分.在此基础上,为了对管网物理异常分类实现系统运行状态的有效分析,将管网信息增维协方差矩阵最大特征向量映射的二维图像作为输入,采用卷积神经网络进行研究,进而实现对物理异常的准确判断.最后通过某实际管网数据进行仿真分析,验证所提方法的有效性.     

A topology and risk-aware access control framework for cyber-physical space

Cyber-physical space is a spatial environment that integrates the cyber world and the physical world, aiming to provide an intelligent environment for users to conduct their day-to-day activities. The interplay between the cyber space and physical space proposes specific security requirements that are not captured by traditional access control frameworks. On one hand, the security of the physical space and the cyber space should be both concerned in the cyber-physical space. On the other hand, the bad results caused by failure in providing secure policy enforcementmay directly affect the controlled physical world. In this paper, we propose an effective access control framework for the cyber-physical space. Firstly, a topology-aware access control (TAAC) model is proposed. It can express the cyber access control, the physical access control, and the interaction access control simultaneously. Secondly, a risk assessment approach is proposed for the policy enforcement phase. It is used to evaluate the user behavior and ensures that the suspicious behaviors executed by authorized users can be handled correctly. Thirdly, we propose a role activation algorithm to ensure that the objects are accessed only by legal and honest users. Finally, we evaluate our approach by using an illustrative example and the performance analysis. The results demonstrate the feasibility of our approach.     

基于能量控制与资源调度的信息物理系统建模

数据中心作为信息物理系统的一种,消耗着巨大的能量。通过对信息物理系统的能量特点进行分析,根据信息物理系统中有大量的具有计算能力的信息设备,将信息物理系统的构件分为两类：计算部件和非计算部件,并以此进行能量系统建模。通过分析信息物理系统资源调度的特点,针对其资源调度的三要素：资源实体能力、资源实时状态和资源上所执行的任务类型进行资源调度建模。最后,以信息物理系统的数据中心为例,针对数据中心的计算部件,给出了一个对能量控制与资源调度进行结合的信息物理系统模型。     

位置约束的访问控制模型及验证方法

随着物联网和信息物理融合系统等新一代信息技术的发展,位置约束的访问控制系统的安全性需求不仅体现在虚拟的信息空间,还体现在现实的物理空间.如何在这种新需求下制定位置约束的访问控制模型与验证方法成为保证访问控制系统安全的关键所在.首先提出位置约束访问控制模型,包括LCRBAC模型和EM模型,实现对信息空间和物理空间的静态结构以及两空间中实体动态行为的刻画;其次利用偶图和偶图反应系统建模位置约束访问控制模型,生成访问控制策略标注转移边的标号变迁系统;然后根据标号变迁系统验证结果,提出针对死锁状态、违反状态和不可达状态的策略修改方案;最后通过银行访问控制系统实例分析说明所提方法能够对信息空间和物理空间以及两空间交互行为的访问控制策略进行建模和验证.     

Deep Learning Based Attack Detection for Cyber-Physical System Cybersecurity: A Survey

With the booming of cyber attacks and cyber criminals against cyber-physical systems(CPSs),detecting these attacks remains challenging.It might be the worst of times,but it might be the best of times because of opportunities brought by machine learning(ML),in particular deep learning(DL).In general,DL delivers superior performance to ML because of its layered setting and its effective algorithm for extract useful information from training data.DL models are adopted quickly to cyber attacks against CPS systems.In this survey,a holistic view of recently proposed DL solutions is provided to cyber attack detection in the CPS context.A six-step DL driven methodology is provided to summarize and analyze the surveyed literature for applying DL methods to detect cyber attacks against CPS systems.The methodology includes CPS scenario analysis,cyber attack identification,ML problem formulation,DL model customization,data acquisition for training,and performance evaluation.The reviewed works indicate great potential to detect cyber attacks against CPS through DL modules.Moreover,excellent performance is achieved partly because of several highquality datasets that are readily available for public use.Furthermore,challenges,opportunities,and research trends are pointed out for future research.     

信息能源系统的信-物融合稳定性分析

尽管信息物理系统的稳定性已经得到了广泛的研究,但大部分的学者皆关注于通信网络延时或攻击下的信息物理系统的稳定性问题,无网络通信的信息物理系统的信物融合稳定性分析策略亟待提出.其中,内嵌数字控制系统的并网逆变器系统是一种最简单、最典型的信息能源系统.同时,从效率的角度出发,逆变器的开关/采样频率总是选择尽可能低的频率,其势必产生系统固有延迟时间(控制理论中称为时间延迟).这种延迟时间往往容易引起系统的低频/次同步振荡,弱电网将加剧此现象.为此,提出一种信息能源系统的信-物融合稳定性分析技术.首先,基于柏德近似方法,建立了具有等效延迟时间的信息物理系统阻抗模型.该等效延迟时间由三部分组成,即信息/物理层的采样延迟时间、信息层的计算延迟时间和物理层的脉宽调制延迟时间,其有效地反映了信息-物理相互融合作用的影响.进而设计了稳定禁止区域判据,利用空间映射使开关/采样频率求解过程转化为Hurwitz矩阵辨识问题.在这些空间映射的基础上,最小开关/采样频率通过自适应步长搜索算法获得.最后,仿真和实验结果验证了该方法的有效性.     

基于Web的节能监察管理系统的设计与实现

节能监察工作经过短短几年的发展日臻成熟,各地节能监察中心的业务也由人工操作日渐转化为网络监察。在分析节能监察工作的核心业务前提下,使用当前流行的B／S开发技术,对节能监察工作的信息化建设提出相应的设计模型,并对Web下的应用工具进行了分析对比。该平台实现了企业对能耗数据上报的快捷及节能监察中心对数据的协同管理,就节能监察工作提高效率、降低成本提供了一种模式。     

集中式超大规模储能电站信息物理系统建模与可靠性评估

集中式超大规模储能电站与其控制系统逐渐发展成为信息物理系统(cyber physical system, CPS),信息技术和监控系统能够使超大规模储能电站应对多样化场景和满足不同的需求,但也带来一定程度的安全运行风险,因此对其进行可靠性建模和分析具有非常重要的意义.首先,建立超大规模储能电站的CPS模型,并分析物理侧与信息侧的交互影响;其次,对信息系统中多种信息扰动的可靠性状态进行建模分析,并提出储能电站CPS可靠性评估指标;然后,分别采用非序贯和序贯蒙特卡洛方法对超大规模储能电站的信息层和物理层元件进行抽样,并量化分析多信息扰动因素对超大规模储能电站CPS可靠性的影响;最后,通过算例仿真结果验证所提模型和方法的有效性,结果表明所提模型可为超大规模储能电站规划和运行提供有效技术支撑.     

A load control method for small data centers participating in demand response programs

This paper presents a load control method for small data centers, which are rarely studied although they account for more than 50% of all data centers. The method utilizes the data network and the electrical network to control power usage for participation in demand response (DR) programs, which are regarded as the killer applications of the emerging smart grid (SG). Traditional data center power management often directly manipulates energy usage, which may be ineffective or impractical for small data centers due to their limited resources. Both the SG and the data centers are considered to be the cyber-physical systems (CPSs). This article proposes an approach that performs the data center DR load management through the cyberspaces of the SG and the targeted data center. The proposed method instructs the workload dispatcher to select the best-suited algorithm when a DR event is issued. Additionally, this method also adjusts the temperature set-points of the air conditioners. The simulation result shows that this approach can achieve a 30% power reduction for DR.     

走向社会信息物理生产系统

随着信息物理系统（Cyber-physical system,CPS）融合深度和融合广度的不断增加,信息物理生产系统（Cyber-physical production system,CPPS）呈现出显著的社会化趋势.通过对信息物理生产系统相关技术的研究,分析了信息物理生产系统的社会化演进历程,建立了社会信息物理生产系统（Social cyber-physical production system,SCPPS）模型;根据人与智能体的信息物理交互行为差异,基于对人类社会行为特点的分析,类比研究了智能体社会与人类社会融合的广义互联社会特点;归纳出信息物理系统的七种交互模式及其在社会信息物理生产系统中的应用;总结出社会信息物理生产系统面临标准化、人性化和安全化的挑战问题.     

Decision-making and control co-design for multi-agent systems: a hierarchical design methodology

Decision-making and control are two of the foremost key ingredients in any autonomous intelligent system. Their codesign has been well-recognized even since the early days of control [1]. Recently, motivated by the wide applications of physical networked systems in different areas to cooperatively meet some cyber computation/communication objectives and constraints, there is an urgent need towards an efficient decision-making and control co-design for these cyber-physical systems. In such designs, we are required to determine distributed rules that can steer these physical plants to a steady-state corresponding to system-level decisionmaking problems. Nevertheless, the twisted complexities resulting from different aspects related to optimization, control, and computation become a great challenge to resolve such problems....     

Fuzzy Control to Improve Energy-Economizing in Cyber-Physical Systems

Energy-supply issues constitute one of the most important issues that we face nowadays. Before a viable alternative energy source of supply is discovered and implemented, saving energy is essential. To improve the energy economizing of actuators, this work proposes a method based on fuzzy logic for scheduling and controlling electrical operators in a cyber-physical system. With the feedbacks of sensor data including output of process and the environmental variations, the intelligent controller can gather the current situation and predicted trend. After analyzing the data with the fuzzy rule, the actuators are controlled with the consideration of the desired set point and energy economizing. The results of a simulation reveal that the fuzzy control method for actuators in a cyber-physical system can be used to minimize the power consumption of the system while accomplishing the desired set point.