• ETRI Journal
• /
• 제41권5호
• /
• pp.684-695
• /
• 2019

In a cloud environment, performance degradation, or even downtime, of virtual machines (VMs) usually appears gradually along with anomalous states of VMs. To better characterize the state of a VM, all possible performance metrics are collected. For such high-dimensional datasets, this article proposes a feature extraction algorithm based on unsupervised fuzzy linear discriminant analysis with kernel (UFKLDA). By introducing the kernel method, UFKLDA can not only effectively deal with non-Gaussian datasets but also implement nonlinear feature extraction. Two sets of experiments were undertaken. In discriminability experiments, this article introduces quantitative criteria to measure discriminability among all classes of samples. The results show that UFKLDA improves discriminability compared with other popular feature extraction algorithms. In detection accuracy experiments, this article computes accuracy measures of an anomaly detection algorithm (i.e., C-SVM) on the original performance metrics and extracted features. The results show that anomaly detection with features extracted by UFKLDA improves the accuracy of detection in terms of sensitivity and specificity.

• 정보처리학회논문지A
• /
• 제12A권5호
• /
• pp.395-404
• /
• 2005

컴퓨터 시스템의 내/외부에 침입(attacks), 고장(failures)이 발생되더라도 적절한 방법으로 중요한 임무에(mission-critical) 해당한 역할을 수행하는 능력의 척도로 정의되는 생존성(survivability)에 대한 관심이 커지고 있다. 특히, 침입에 의해 시스템 일부가 손상(partially compromised) 되더라도, 최소한의 필수 서비스를 지속적으로 제공할 수 있게 해주는 침입감내시스템(intrusion tolerance system)의 설계시에 생존성 분석은 신뢰성(reliability), 가용도(availability)등과 같은 컴퓨터 시스템의 정량적 신인도(dependability) 분석과 함께 중요한 요소기술 중의 하나이다. 본 논문에서는 침입감내시스템의 방어능력을 평가하기 위해 자율컴퓨팅(autonomic computing)의 핵심 기술인 자가치유(self-healing) 메커니즘의 두 가지 요소(결함모델 및 시스템반응)를 활용하여, 주서버와 보조서버로 구성된 침입감내시스템의 상태천이(state transition)를 표현하였다. 또한, 침입감내시스템의 생존성, 가용도 및 다운타임 비용(downtime cost)을 정량적으로 정의한 후 시뮬레이션 실험 및 취약성(vulnerability) 공격에 대한 사례 연구를 수행하였다. 이를 통해 시스템의 신인도 향상 측면에서 초기상태에서의 침입감내능력 향상이 가장 중요한 요소임을 검증할 수 있었다.

• 한국기계가공학회지
• /
• 제13권5호
• /
• pp.21-27
• /
• 2014

Currently, wind power has become a major research field in the area of sustainable development. As one important component of a wind turbine transmission system, most instances of downtime due to a gearbox failure are caused by bearing failures. Gearboxes for wind turbines must have the highest levels of reliability over a period of approximately 20 years, withstanding high dynamic loads. At the same time, a lightweight design and cost minimization efforts are required. These demands can only be met with a well-thought-out design, high-quality materials, a high production quality and proper maintenance. In order to design a reliable and lightweight gearbox, it is necessary to analyze methods pertaining to the bearing rating lifetimes of the standard and of different companies, also including calculation methods for modification factors. This can determine the influence of the bearing lifetime.

• 한국지진공학회논문집
• /
• 제17권6호
• /
• pp.279-291
• /
• 2013

The seismic damage of non-structural components, such as communication facilities, causes direct economic losses as well as indirect losses which result from social chaos occurring with downtime of communication and financial management network systems. The current Korean seismic code, KBC2009, prescribes the design criteria and requirements of non-structural components based on their elastic response. However, it is difficult for KBC to reflect the dynamic characteristics of structures where non-structural components exist. In this study, both linear and nonlinear time history analyses of structures with various analysis parameters were carried out and floor acceleration spectra obtained from analyses were compared with both ground acceleration spectra used for input records of the analyses and the design floor acceleration spectrum proposed by National Radio Research Agency. Also, this study investigates to find out the influence of structural dynamic characteristics on the floor acceleration spectra. The analysis results show that the acceleration amplification is observed due to the resonance phenomenon and such amplification increases with the increase of building heights and with the decrease of structure's energy dissipation capacities.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권8호
• /
• pp.3865-3883
• /
• 2016

Infrastructure as a Service (IaaS) encapsulates computer hardware into a large amount of virtual and manageable instances mainly in the form of virtual machine (VM), and provides rental service for users. Currently, VM anomaly incidents occasionally occur, which leads to performance issues and even downtime. This paper aims at detecting anomalous VMs based on performance metrics data of VMs. Due to the dynamic nature and increasing scale of IaaS, detecting anomalous VMs from voluminous correlated and non-Gaussian monitored performance data is a challenging task. This paper designs an anomaly detection framework to solve this challenge. First, it collects 53 performance metrics to reflect the running state of each VM. The collected performance metrics are testified not to follow the Gaussian distribution. Then, it employs independent components analysis (ICA) instead of principal component analysis (PCA) to extract independent components from collected non-Gaussian performance metric data. For anomaly detection, it employs multi-class Bayesian classification to determine the current state of each VM. To evaluate the performance of the designed detection framework, four types of anomalies are separately or jointly injected into randomly selected VMs in a campus-wide testbed. The experimental results show that ICA-based detection mechanism outperforms PCA-based and LDA-based detection mechanisms in terms of sensitivity and specificity.

• Journal of Advanced Research in Ocean Engineering
• /
• 제4권1호
• /
• pp.16-24
• /
• 2018

This paper addresses the heading control of an offshore floating storage and regasification unit (FSRU) using a resolved motion and acceleration control (RMAC) algorithm. A turret moored vessel tends to have the slewing motion. This slewing motion may cause a considerable decrease in working time in loading and unloading operation because the sloshing in the LNG containment tank might happen and/or the collision between FSRU and LNGC may take place. In order to deal with the downtime problem due to this slewing motion, a heading control system for the turret moored FSRU is developed, and a series of model tests with azimuth thrusters on the FSRU is conducted. A Kalman filter is applied to estimate the low-frequency motion of the vessel. The RMAC algorithm is employed as a primary heading control method and modified I-controller is introduced to reduce the steady-state errors of the heading of the FSRU.

• 한국조명전기설비학회:학술대회논문집
• /
• 한국조명전기설비학회 2004년도 학술대회 논문집
• /
• pp.365-370
• /
• 2004

The integration of intelligent devices, devices-level networks, and software into motor control systems can deliver improved diagnostics, fast warnings for increased system reliability, design flexibility, and simplified wiring. Remote access to motor-control information also affords an opportunity for reduced exposure to hazardous voltage and improved personnel safety during startup and trouble-shooting. This paper presents LonWorks fieldbus networked intelligent induction control system architecture. Experimental bed system with two inverter motor driving system for controlling 1.5kW induction motor is configured for LonWorks networked intelligent motor control. In recent years, MCCs have evolved to include component technologies, such as variable-speed drives, solid-state starters, and electronic overload relays. Integration was accomplished through hardwiring to a programmable logic controller (PLC) or distributed control system (DCS). Devicelevel communication networks brought new possibilities for advanced monitoring, control and diagnostics. This LonWorks network offered the opportunity for greatly simplified wiring, eliminating the bundles of control interwiring and corresponding complex interwiring diagrams. An intelligent MCC connected in device level control network proves users with significant new information for preventing or minimizing downtime. This information includes warnings of abnormal operation, identification of trip causes, automated logging of events, and electronic documentation. In order to show the application of the multi-motors control system, the prototype control system is implemented. This paper is the first step to drive multi-motors with serial communication which can satisfy the real time operation using LonWorks network.

• 품질경영학회지
• /
• 제45권1호
• /
• pp.25-38
• /
• 2017

Purpose: Hot strip rolling mill consists of a lot of mechanical and electrical units. In condition monitoring and diagnosis phase, various units could be failed with unknown reasons. In this study, we propose an effective method to detect early the units with abnormal status to minimize system downtime. Methods: The early warning problem with various units is defined. K-means and PAM algorithm with Euclidean and Manhattan distances were performed to detect the abnormal status. In addition, an performance of the proposed algorithm is investigated by field data analysis. Results: PAM with Manhattan distance(PAM_ManD) showed better results than K-means algorithm with Euclidean distance(K-means_ED). In addition, we could know from multivariate field data analysis that the system reliability of hot strip rolling mill can be increased by detecting early abnormal status. Conclusion: In this paper, clustering-based monitoring and fault detection algorithm using Manhattan distance is proposed. Experiments are performed to study the benefit of the PAM with Manhattan distance against the K-means with Euclidean distance.

• 한국정밀공학회지
• /
• 제29권4호
• /
• pp.373-379
• /
• 2012

Disruptions in manufacturing systems caused by system changes and disturbances such as the tool wear, machine breakdown, malfunction of transporter, and so on, reduce the productivity and the increase of downtime and manufacturing cost. In order to cope with these challenges, a new method to build an intelligent manufacturing system with biological principles, namely an ant colony inspired manufacturing system, is presented. In the developed system, the manufacturing system is considered as a swarm of cognitive agents where work-pieces, machines and transporters are controlled by the corresponding cognitive agent. The system reacts to disturbances autonomously based on the algorithm of each autonomous entity or the cooperation with them. To develop the ant colony inspired manufacturing system, the disturbances happened in the machining shop of a transmission case were analyzed to classify them and to find out the corresponding management methods. The system architecture with the autonomous characteristics was generated with the cognitive agent and the ant colony technology. A test bed was implemented to prove the functionality of the developed system.

• Structural Engineering and Mechanics
• /
• 제61권6호
• /
• pp.701-709
• /
• 2017

Seismic dissipation devices can play a crucial role in mitigating earthquake damages, loss of life and post-event repair and downtime costs. This research investigates the use of ring springs with high-force-to-volume (HF2V) dissipaters to create damage-free, recentring connections and structures. HF2V devices are passive rate-dependent extrusion-based devices with high energy absorption characteristics. Ring springs are passive energy dissipation devices with high self-centring capability to reduce the residual displacements. Dynamic behaviour of a system with nonlinear structural stiffness and supplemental hybrid damping via HF2V devices and ring spring dampers is used to investigate the design space and potential. HF2V devices are modelled with design forces equal to 5% and 10% of seismic weight and ring springs are modelled with loading stiffness values of 20% and 40% of initial structural stiffness and respective unloading stiffness of 7% and 14% of structural stiffness (equivalent to 35% of their loading stiffness). Using a suite of 20 design level earthquake ground motions, nonlinear response spectra for 8 different configurations are generated. Results show up to 50% reduction in peak displacements and greater than 80% reduction in residual displacements of augmented structure compared to the baseline structure. These gains come at a cost of a significant rise in the base shear values up to 200% mainly as a result of the force contributed by the supplemental devices.