• 정보보호학회논문지
• /
• 제29권5호
• /
• pp.1027-1037
• /
• 2019

기존의 사이버 공격 탐지 솔루션은 일반적으로 시그니처 기반 내지 악성행위 분석을 통한 방식의 탐지를 수행하므로, 알려지지 않은 방식에 의한 공격은 탐지하기 어렵다는 한계가 있다. 시스템에서는 상시로 발생하는 다양한 정보들이 시스템의 상태를 반영하고 있으므로, 이들 정보를 수집하여 정상상태를 학습하고 이상상태를 탐지하는 방식으로 알려지지 않은 공격을 탐지할 수 있다. 본 논문은 정상상태 학습 및 탐지에 활용하기 위하여 문자열을 그 순서와 의미를 보존하며 정량적 수치로 변환하는 머신러닝 임베딩(Embedding) 기법과 이상상태의 탐지를 위하여 다수의 정상데이터에서 소수의 비정상 데이터를 탐지하는 머신러닝 이상치 탐지(Novelty Detection) 기법을 이용하여 사이버 공격에 의한 시스템 이상상태를 탐지하는 방안을 제안한다.

• Nuclear Engineering and Technology
• /
• 제55권4호
• /
• pp.1528-1539
• /
• 2023

Since the leakage of sodium in an SFR (sodium-cooled fast reactor) causes an explosion upon reaction with air and water, sodium leakages represent an important safety issue. In this study, a novel technique for improving the reliability of sodium leakage detection applying DDVR (dynamic data validation and reconciliation) is proposed and verified to resolve this technical issue. DDVR is an approach that aims to improve the accuracy of a target system in a dynamic state by minimizing random errors, such as from the uncertainty of instruments and the surrounding environment, and by eliminating gross errors, such as instrument failure, miscalibration, or aging, using the spatial redundancy of measurements in a physical model and the reliability information of the instruments. DDVR also makes it possible to estimate the state of unmeasured points. To validate this approach for supporting sodium leakage detection, this study applies experimental data from a sodium leakage detection experiment performed by the Korea Atomic Energy Research Institute. The validation results show that the reliability of sodium leakage detection is improved by cooperation between DDVR and hardware measurements. Based on these findings, technology integrating software and hardware approaches is suggested to improve the reliability of sodium leakage detection by presenting the expected true state of the system.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2002년도 춘계학술대회 논문집
• /
• pp.1038-1042
• /
• 2002

Monitoring of the drill wear and hole quality change is conducted during the drilling process. Cutting force measured by tool dynamometer is a evident feature estimating abnormal state of drilling. One major difficulty in using tool dynamometer is that the work piece must be mounted on the dynamometer, and thus the machining process is disturbed and discontinuous. Acoustic transducer do not disturb the normal machining process, and provide a relatively easy way to monitor a machining process for industrial application. For this advantage, AE signal is used to estimate the abnormal state. In this study vision system is used to detect flank wear tendency and hole quality, there are many formal factors in hole quality decision circularity, cylindricity, straightness, and so on, but these are difficult to measure in on-line monitoring. The movement of hole center and increasement of hole diameter is presented to determine hole quality As the results of this experiment, AE RMS signal and measurements by vision system are shown the similar tendency as abnormal state of drilling. And detection of the abnormal states using BPNs was achieved 96.4% reliability.

• 센서학회지
• /
• 제21권1호
• /
• pp.39-45
• /
• 2012

Recently the convergence of healthcare technology is used for daily life healthcare monitoring. Cardiac arrhythmia is presented by the state of the heart irregularity. Abnormal heart's electrical signal pathway or heart's tissue disorder could be the cause of cardiac arrhythmia. Fatal arrhythmia could put patient's life at risk. Therefore arrhythmia detection is very important. Previous studies on the detection of arrhythmia in various ECG analysis and classification methods had been carried out. In this paper, an ECG signal processing techniques to detect abnormal ECG based on DTW minimum accumulation distance through the template matching for normalized data and variable threshold method for ECG R-peak detection. Signal processing techniques able to determine the occurrence of normal ECG and abnormal ECG. Abnormal ECG detection algorithm using DTW minimum accumulation distance method is performed using MITBIH database for performance evaluation. Experiment result shows the average percentage accuracy of using the propose method for Rpeak detection is 99.63 % and abnormal detection is 99.60 %.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2000년도 추계학술대회 논문집
• /
• pp.888-891
• /
• 2000

Drilling is one of the most important operations in machining industry and usually the most efficient and economical method of cutting a hole in metal. From automobile parts to aircraft components, almost every manufactured product requires that holes are to be drilled for the purpose of assembly, creation of fluid passages, and so on. It is therefore desirable to monitor drill wear and hole quality changes during the hole drilling process. One important aspect in controlling the drilling process is drill wear status monitoring. With the monitoring, we may decide on optimal timing for tool change. The necessity of the detection of tool wear, fracture and the abnormal tool state has been emphasized in the machining process. Accordingly, this paper deals with the cutting characteristics of the hot-rolled high strength steels using common HSS drill. The performance variables include drill wear data obtained from drilling experiments conducted on the workpiece. The results are obtained from monitoring of the cutting force and Acoustic Emission (AE) signals, and from the detection of the abnormal tool state with the computer vision system.

• International Journal of Computer Science & Network Security
• /
• 제21권4호
• /
• pp.131-139
• /
• 2021

In current time, anomaly detection is the primary concern of the administrative authorities. Suspicious activity identification is shifting from a human operator to a machine-assisted monitoring in order to assist the human operator and react to an unexpected incident quickly. These automatic surveillance systems face many challenges due to the intrinsic complex characteristics of video sequences and foreground human motion patterns. In this paper, we propose a novel approach to detect anomalous human activity using a hybrid approach of statistical model and Genetic Programming. The feature-set of local motion patterns is generated by a statistical model from the video data in an unsupervised way. This features set is inserted to an enhanced Genetic Programming based classifier to classify normal and abnormal patterns. The experiments are performed using publicly available benchmark datasets under different real-life scenarios. Results show that the proposed methodology is capable to detect and locate the anomalous activity in the real time. The accuracy of the proposed scheme exceeds those of the existing state of the art in term of anomalous activity detection.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2001년도 추계학술대회논문집A
• /
• pp.821-826
• /
• 2001

Out of all metal-cutting processes, the hole-making process is the most widely used. It is estimated to be more than 30% of the total metal-cutting process. It is therefore desirable to monitor and detect drill wear during the hole-drilling process. One important aspect in controlling the drilling process is monitoring drill wear status. Accordingly, this paper deals with Basic system and Online system. Basic system comprised of spindle rotational speed, feed rates, thrust, torque and flank wear measured tool microscope. Online system comprised of spindle rotational speed, feed rates, AE signal, flank wear area measured computer vision. On-line monitoring system does not need to stop the process to inspect drill wear. Backpropagation neural networks (BPNs) were used for on-line detection of drill wear. This paper deals with an on-line drill wear monitoring system to fit the detection of the abnormal tool state.

• 한국공작기계학회논문집
• /
• 제14권6호
• /
• pp.16-21
• /
• 2005

The cutting characteristics of hardened steel(AISI 52100) by PCBN tools is investigated with respect to cutting force, workpiece surface roughness and tool flank wear by the vision system. Hard Owning is carried out with various cutting conditions; spindle rotational speed, depth of cut and feed rate. Backpropagation neural networks(BPNs) are used for detection of tool wear. The input vectors of neural network comprise of spindle rotational speed, feed rates, vision flank wear, and thrust force signals. The output is the tool wear state which is either usable or failure. The detection of the abnormal states using BPNs achieves $96.8\%$ reliability even when the spindle rotational speed and feedrate are changed.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제18권2호
• /
• pp.348-369
• /
• 2024

With the advancement of Industry 4.0 and Industrial Internet of Things (IIoT), manufacturing increasingly seeks automation and intelligence. Temperature and vibration monitoring are essential for machinery health. Traditional abnormal state detection methodologies often overlook the intricate frequency characteristics inherent in vibration time series and are susceptible to erroneously reconstructing temperature abnormalities due to the highly similar waveforms. To address these limitations, we introduce synergistic, end-to-end, unsupervised Frequency-Time Domain Memory-Enhanced Autoencoders (FTD-MAE) capable of identifying abnormalities in both temperature and vibration datasets. This model is adept at accommodating time series with variable frequency complexities and mitigates the risk of overgeneralization. Initially, the frequency domain encoder processes the spectrogram generated through Short-Time Fourier Transform (STFT), while the time domain encoder interprets the raw time series. This results in two disparate sets of latent representations. Subsequently, these are subjected to a memory mechanism and a limiting function, which numerically constrain each memory term. These processed terms are then amalgamated to create two unified, novel representations that the decoder leverages to produce reconstructed samples. Furthermore, the model employs Spectral Entropy to dynamically assess the frequency complexity of the time series, which, in turn, calibrates the weightage attributed to the loss functions of the individual branches, thereby generating definitive abnormal scores. Through extensive experiments, FTD-MAE achieved an average ACC and F1 of 0.9826 and 0.9808 on the CMHS and CWRU datasets, respectively. Compared to the best representative model, the ACC increased by 0.2114 and the F1 by 0.1876.

• Journal of Advanced Marine Engineering and Technology
• /
• 제30권2호
• /
• pp.247-252
• /
• 2006

Almost ship monitoring systems are event driven alarm system which warn only when the measurement value is over or under set point. These kinds of system cannot warn until signal is growing to abnormal state that the signal is over or under the set point. therefore cannot play a role for preventive maintenance system. This paper proposes fault diagnosis method which is able to diagnose and forecast the fault from present operating condition by analyzing monitored signals with present ship monitoring system without any additional sensors. By analyzing the data with high correlation coefficient(CC), correlation level of interactive data can be defined. Knowledge base of abnormal detection can be built by referring level of CC(Fault Detection CC. FDCC) to detect abnormal data among monitored data from monitoring system and knowledge base of diagnosis built by referring CC among interactive data for related machine each other to diagnose fault part.