• Smart Structures and Systems
• /
• 제30권6호
• /
• pp.613-626
• /
• 2022

Guaranteeing the quality and integrity of structural health monitoring (SHM) data is very important for an effective assessment of structural condition. However, sensory system may malfunction due to sensor fault or harsh operational environment, resulting in multiple types of data anomaly existing in the measured data. Efficiently and automatically identifying anomalies from the vast amounts of measured data is significant for assessing the structural conditions and early warning for structural failure in SHM. The major challenges of current automated data anomaly detection methods are the imbalance of dataset categories. In terms of the feature of actual anomalous data, this paper proposes a data anomaly detection method based on data-level and deep learning technique for SHM of civil engineering structures. The proposed method consists of a data balancing phase to prepare a comprehensive training dataset based on data-level technique, and an anomaly detection phase based on a sophisticatedly designed network. The advanced densely connected convolutional network (DenseNet) and Transformer encoder are embedded in the specific network to facilitate extraction of both detail and global features of response data, and to establish the mapping between the highest level of abstractive features and data anomaly class. Numerical studies on a steel frame model are conducted to evaluate the performance and noise immunity of using the proposed network for data anomaly detection. The applicability of the proposed method for data anomaly classification is validated with the measured data of a practical supertall structure. The proposed method presents a remarkable performance on data anomaly detection, which reaches a 95.7% overall accuracy with practical engineering structural monitoring data, which demonstrates the effectiveness of data balancing and the robust classification capability of the proposed network.

• 한국건축친환경설비학회 논문집
• /
• 제12권6호
• /
• pp.579-590
• /
• 2018

Anomaly detection on building energy consumption has been regarded as an effective tool to reduce energy saving on building operation and maintenance. However, it requires energy model and FDD expert for quantitative model approach or large amount of training data for qualitative/history data approach. Both method needs additional time and labors. This study propose a machine learning and data science approach to define faulty conditions on hourly building energy consumption with reducing data amount and input requirement. It suggests an application of Support Vector Data Description (SVDD) method on training normal condition of hourly building energy consumption incorporated with hourly outdoor air temperature and time integer in a week, 168 data points and identifying hourly abnormal condition in the next day. The result shows the developed model has a better performance when the ${\nu}$ (probability of error in the training set) is 0.05 and ${\gamma}$ (radius of hyper plane) 0.2. The model accuracy to identify anomaly operation ranges from 70% (10% increase anomaly) to 95% (20% decrease anomaly) for daily total (24 hours) and from 80% (10% decrease anomaly) to 10%(15% increase anomaly) for occupied hours, respectively.

• 디지털콘텐츠학회 논문지
• /
• 제19권2호
• /
• pp.377-383
• /
• 2018

사물인터넷(IoT) 서비스는 스마트 환경이 발전하면서 다양한 데이터를 생산하고 있다. 이 데이터는 사용자의 상황을 판단하는 중요한 데이터로 사용된다. 그렇기 때문에 센서의 이상 상태를 실시간으로 모니터링하고 이상 데이터를 탐지하는 것이 중요하다. 하지만 데이터 구조와 프로토콜이 다양하기 때문에 표준화된 데이터 구조로 변환하는 과정이 필요하다. 그럼으로써 데이터의 품질을 보장하고 정확한 분석을 통해 서비스의 품질까지 좋아지는 효과를 기대할 수 있다. 본 논문은 수집된 센서의 이상탐지를 위해 빅데이터 기반의 이상탐지 시스템을 제안한다. 제안한 시스템은 이상탐지를 위해 데이터 표준화 전처리와 시계열 기반의 이상탐지가 우수한 SVM(Support Vector Machine) 모델을 적용하였다. 실험에서는 전처리와 전처리되지 않은 데이터를 각각 학습시키고 비교하였다. 그 결과, 전처리된 데이터는 이상 장애를 정확히 탐지하고 예측하였다.

• 한국전자거래학회지
• /
• 제20권4호
• /
• pp.41-59
• /
• 2015

시스템이 다양화 되면서 동시에 저장된 로그도 다양하게 분석할 필요가 생겼다. 이러한 로그 데이터 분석에 관한 필요성이 강해지는 환경이 시간 순으로 발생하는 이벤트 단위의 로그로부터 프로세스 모델을 도출하고, 시스템을 개선시키는 활동에 이바지하도록 요구하고 있다. 기존에는 개별 이벤트 단위의 로그를 분석하면서 속성들의 관계를 파악하는 연구가 활발했다. 본 논문에서는 로그 데이터를 활용한 예외적인 형태의 프로세스 인스턴스를 판별하는 방법으로 LAPID(Local Anomaly Process Instance Detection)를 제안한다. LAPID는 액티비티-릴레이션 매트릭스(Activity relation matrix)를 사용해서 계산된 거리 값을 활용하여, API(Anomaly Process Instance)를 탐색한다. 제시한 방법의 유용성을 검증하기 위하여 항만 물류에서 발생하는 컨테이너 이동에 대한 트레이스(Trace)를 포함하는 로그 데이터에서 예외적인 상황의 프로세스 실행이 가지는 특징을 도출하였다. 이를 위하여 본 논문에서는 국내의 실제 항만에서 발생한 이벤트 로그를 이용하여 사례연구를 수행하였다.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2022년도 추계학술대회
• /
• pp.19-21
• /
• 2022

자기 지도 학습을 이용한 이상 탐지는 일반적으로 합성 데이터를 생성해 정상과 이상을 학습하고, 실제 이상 데이터를 테스트 데이터로 사용하여 이상 탐지 성능을 측정한다. 정상 데이터와 유사한 합성 데이터를 생성하기 위해 기존 연구에서는 원본 이미지에서 특정 패치를 자르고 붙이는 식으로 합성 데이터를 생성한다. 이런 방식에서 정상 데이터와 유사한 정도는 패치 개수와 크기에 따라 달라지므로 이상 탐지 성능에 영향을 미칠 수 있다. 본 연구에서는 패치 크기 및 개수를 다르게 하여 합성 데이터를 생성한 뒤 사전 학습된 모델을 사용하여 정상 데이터와의 유사성 측정 및 분석을 진행하였고 모델을 학습시켜 이상 탐지 성능을 측정하여 보았다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권8호
• /
• pp.2787-2800
• /
• 2022

The aging society increases emergency situations of the elderly living alone and a variety of social crimes. In order to prevent them, techniques to detect emergency situations through voice are actively researched. This study proposes CutPaste-based anomaly detection model using multi-scale feature extraction in time series streaming data. In the proposed method, an audio file is converted into a spectrogram. In this way, it is possible to use an algorithm for image data, such as CNN. After that, mutli-scale feature extraction is applied. Three images drawn from Adaptive Pooling layer that has different-sized kernels are merged. In consideration of various types of anomaly, including point anomaly, contextual anomaly, and collective anomaly, the limitations of a conventional anomaly model are improved. Finally, CutPaste-based anomaly detection is conducted. Since the model is trained through self-supervised learning, it is possible to detect a diversity of emergency situations as anomaly without labeling. Therefore, the proposed model overcomes the limitations of a conventional model that classifies only labelled emergency situations. Also, the proposed model is evaluated to have better performance than a conventional anomaly detection model.

• Smart Structures and Systems
• /
• 제29권6호
• /
• pp.757-766
• /
• 2022

To train deep learning algorithms, a sufficient number of data are required. However, in most engineering systems, the acquisition of fault data is difficult or sometimes not feasible, while normal data are secured. The dearth of data is one of the major challenges to developing deep learning models, and fault diagnosis in particular cannot be made in the absence of fault data. With this context, this paper proposes an anomaly detection methodology for rotating machines using only normal data with self-labeling. Since only normal data are used for anomaly detection, a self-labeling method is used to generate a new labeled dataset. The overall procedure includes the following three steps: (1) transformation of normal data to self-labeled data based on a pretext task, (2) training the convolutional neural networks (CNN), and (3) anomaly detection using defined anomaly score based on the softmax output of the trained CNN. The softmax value of the abnormal sample shows different behavior from the normal softmax values. To verify the proposed method, four case studies were conducted, on the Case Western Reserve University (CWRU) bearing dataset, IEEE PHM 2012 data challenge dataset, PHMAP 2021 data challenge dataset, and laboratory bearing testbed; and the results were compared to those of existing machine learning and deep learning methods. The results showed that the proposed algorithm could detect faults in the bearing testbed and compressor with over 99.7% accuracy. In particular, it was possible to detect not only bearing faults but also structural faults such as unbalance and belt looseness with very high accuracy. Compared with the existing GAN, the autoencoder-based anomaly detection algorithm, the proposed method showed high anomaly detection performance.

• 산업경영시스템학회지
• /
• 제43권2호
• /
• pp.33-38
• /
• 2020

Anomaly detection of Machine Learning such as PCA anomaly detection and CNN image classification has been focused on cross-sectional data. In this paper, two approaches has been suggested to apply ML techniques for identifying the failure time of big time series data. PCA anomaly detection to identify time rows as normal or abnormal was suggested by converting subjects identification problem to time domain. CNN image classification was suggested to identify the failure time by re-structuring of time series data, which computed the correlation matrix of one minute data and converted to tiff image format. Also, LASSO, one of feature selection methods, was applied to select the most affecting variables which could identify the failure status. For the empirical study, time series data was collected in seconds from a power generator of 214 components for 25 minutes including 20 minutes before the failure time. The failure time was predicted and detected 9 minutes 17 seconds before the failure time by PCA anomaly detection, but was not detected by the combination of LASSO and PCA because the target variable was binary variable which was assigned on the base of the failure time. CNN image classification with the train data of 10 normal status image and 5 failure status images detected just one minute before.

• 한국항해항만학회지
• /
• 제40권5호
• /
• pp.271-278
• /
• 2016

Localized atmospheric conditions between multi-reference stations can bring the tropospheric delay irregularity that becomes an error terms affecting positioning accuracy in network RTK environment. Imbalanced network error can affect the network solutions and it can corrupt the entire network solution and degrade the correction accuracy. If an anomaly could be detected before the correction message was generated, it is possible to eliminate the anomalous satellite that can cause degradation of the network solution during the tropospheric delay anomaly. An atmospheric grid that consists of four meteorological stations was used to detect an inhomogeneous weather conditions and tropospheric anomaly applied AWSs (automatic weather stations) meteorological data. The threshold of anomaly detection algorithm was determined based on the statistical weather data of AWSs for 5 years in an atmospheric grid. From the analytic results of anomaly detection algorithm it showed that the proposed algorithm can detect an anomalous satellite with an anomaly flag generation caused tropospheric delay anomaly during localized atmospheric conditions between stations. It was shown that the different precipitation condition between stations is the main factor affecting tropospheric anomalies.

• 한국신뢰성학회지:신뢰성응용연구
• /
• 제18권1호
• /
• pp.20-32
• /
• 2018

Purpose: The purpose of this study is to set up an anomaly detection criteria for sensor data coming from a motorcycle. Methods: Five sensor values for accelerator pedal, engine rpm, transmission rpm, gear and speed are obtained every 0.02 second from a motorcycle. Exploratory data analysis is used to find any pattern in the data. Traditional process control methods such as X control chart and time series models are fitted to find any anomaly behavior in the data. Finally unsupervised learning algorithm such as k-means clustering is used to find any anomaly spot in the sensor data. Results: According to exploratory data analysis, the distribution of accelerator pedal sensor values is very much skewed to the left. The motorcycle seemed to have been driven in a city at speed less than 45 kilometers per hour. Traditional process control charts such as X control chart fail due to severe autocorrelation in each sensor data. However, ARIMA model found three abnormal points where they are beyond 2 sigma limits in the control chart. We applied a copula based Markov chain to perform statistical process control for correlated observations. Copula based Markov model found anomaly behavior in the similar places as ARIMA model. In an unsupervised learning algorithm, large sensor values get subdivided into two, three, and four disjoint regions. So extreme sensor values are the ones that need to be tracked down for any sign of anomaly behavior in the sensor values. Conclusion: Exploratory data analysis is useful to find any pattern in the sensor data. Process control chart using ARIMA and Joe's copula based Markov model also give warnings near similar places in the data. Unsupervised learning algorithm shows us that the extreme sensor values are the ones that need to be tracked down for any sign of anomaly behavior.