Fault detection in electromechanical systems plays a significant role in product quality and manufacturing efficiency during the transition to smart manufacturing. Because collecting a sufficient number of datasets under faulty conditions of the system is challenging in practical industrial sites, unsupervised fault detection methods are mainly used. Although fault datasets accumulate during machine operation, it is not straightforward to utilize the information it contains for fault detection after the deep learning model has been trained in an unsupervised manner. However, the information in fault datasets is expected to significantly contribute to fault detection. In this regard, this study aims to validate the effectiveness of the transition from unsupervised to supervised learning as fault datasets gradually accumulate through continuous machine operation. We also focus on experimentally analyzing how changes in the learning paradigm of the deep learning model and the output representation affect fault detection performance. The results demonstrate that, with a small number of fault datasets, a supervised model with continuous outputs as a regression problem showed better fault detection performance than the original model with one-hot encoded outputs (as a classification problem).

선박의 추진전동기는 소량 주문생산되어 고장진단을 위한 신호를 사전에 확보하는 것이 불가능하다. 운용기간 중 계측을 통해 데이터를 확보하는 것은 많은 시간과 비용을 초래하기에 물리모델을 통해 데이터를 확보하는 것이 유일한 방법이다. 물리모델을 통해 얻 은 데이터를 고장진단에 활용하기 위하여 데이터의 정확도를 확보해야 한다. 기존 전동기 물리모델의 경우 전동기에서 발생하는 구조-전 기 연성효과를 온전히 고려하지 않아 진동데이터의 해석 오차가 발생하는 것을 확인할 수 있다. 본 논문에서는 구조-전기 완전연성 물리 모델을 개발하여 물리모델데이터의 정확도를 개선하였다. 실험계측 데이터와 물리모델 데이터의 비교를 통해 전동기 상태별 데이터를 높 은 정확도로 획득할 수 있음을 확인하였다. 본 논문에서 제시한 구조-전기 완전연성 물리모델을 이용하여 정상상태와 결함상태에서 나타 나는 진동수준을 예측할 수 있음을 확인하였으며, 구조-전기 완전연성 반영 필요성을 입증하였다.

Bearing-shaft systems are essential components in various automated manufacturing processes, primarily designed for the efficient rotation of a main shaft by a motor. Accurate fault detection is critical for operating manufacturing processes, yet challenges remain in sensor selection and optimization regarding types, locations, and positioning. Sound signals present a viable solution for fault detection, as microphones can capture mechanical sounds from remote locations and have been traditionally employed for monitoring machine health. However, recordings in real industrial environments always contain non-negligible ambient noise, which hampers effective fault detection. Utilizing a high-performance microphone for noise cancellation can be cost-prohibitive and impractical in actual manufacturing sites, therefore to address these challenges, we proposed a convolution neural network-based methodology for fault detection that analyzes the mechanical sounds generated from the bearing-shaft system in the form of Log-mel spectrograms. To mitigate the impact of environmental noise in recordings made with commercial microphones, we also developed a denoising autoencoder that operates without requiring any expert knowledge of the system. The proposed DAE-CNN model demonstrates high performance in fault detection regardless of whether environmental noise is included(98.1%) or not(100%). It indicates that the proposed methodology effectively preserves significant signal features while overcoming the negative influence of ambient noise present in the collected datasets in both fault detection and fault type classification.

In recent automated manufacturing systems, compressed air-based pneumatic cylinders have been widely used for basic perpetration including picking up and moving a target object. They are relatively categorized as small machines, but many linear or rotary cylinders play an important role in discrete manufacturing systems. Therefore, sudden operation stop or interruption due to a fault occurrence in pneumatic cylinders leads to a decrease in repair costs and production and even threatens the safety of workers. In this regard, this study proposed a fault detection technique by developing a time-variant deep learning model from multivariate sensor data analysis for estimating a current health state as four levels. In addition, it aims to establish a real-time fault detection system that allows workers to immediately identify and manage the cylinder’s status in either an actual shop floor or a remote management situation. To validate and verify the performance of the proposed system, we collected multivariate sensor signals from a rotary cylinder and it was successful in detecting the health state of the pneumatic cylinder with four severity levels. Furthermore, the optimal sensor location and signal type were analyzed through statistical inferences.

역사지진 및 계기지진이 보고되어 앞으로 지진이 발생할 가능성이 있는 한반도 중부지역 중, 충청남도 서부지역 에 분포하는 선형구조와 단층지형을 분석하였다. 단층지형을 기반으로 총 151개의 선형구조가 추출하였다. 당진단층과 예산단층이 위치하는 지역에 단층과 주향이 일치하는 선형구조가 밀집하여 분포하는 반면, 홍성단층이 위치한 지역에는 선형구조의 수가 적으며 단층과 유사한 주향을 갖는 선형구조 또한 잘 인지되지 않는다. 이러한 특징은 넓은 충적층과, 오랜 기간의 풍화와 침식, 그리고 경작으로 인한 지표의 변형 등에 의해 단층을 지시하는 지형증거를 인지하기 어렵기 때문으로 판단된다. 단층으로 판명된 5개의 주요지점에서는 단층안부, 제4기 충적층에 나타나는 경사급변점, 선형곡 등 의 단층지형이 선형구조를 따라 인지되었으며 단층지형이 실제 단층을 잘 지시하는 것으로 나타났다. 한편 선형구조 내 에서 감지된 제4기층의 변위는 단층운동에 의해 직접적으로 형성된 것이 아닌 농경지 정리와 같은 인위적 교란이나 하 천 침식의 영향과 같은 외부요인에 의해 형성되었을 가능성이 있는 것으로 파악되었다. 연구지역에서 인지되는 단층지 형의 유형과 한반도 남동부 지역에서 인지되는 단층지형의 유형에 차이가 나타나는데 이는 단층지형의 유형이 단층종 류에 따라 변화되는 한 예를 보여준다.

Recent earthquakes in Korea, like Gyeongju and Pohang, have highlighted the need for accurate seismic hazard assessment. The lack of substantial ground motion data necessitates stochastic simulation methods, traditionally used with a simplistic point-source assumption. However, as earthquake magnitude increases, the influence of finite faults grows, demanding the adoption of finite faults in simulations for accurate ground motion estimates. We analyzed variations in simulated ground motions with and without the finite fault method for earthquakes with magnitude (Mw) ranging from 5.0 to 7.0, comparing pseudo-spectral acceleration. We also studied how slip distribution and hypocenter location affect simulations for a virtual earthquake that mimics the Gyeongju earthquake with Mw 5.4. Our findings reveal that finite fault effects become significant at magnitudes above Mw 5.8, particularly at high frequencies. Notably, near the hypocenter, the virtual earthquake’s ground motion significantly changes using a finite fault model, especially with heterogeneous slip distribution. Therefore, applying finite fault models is crucial for simulating ground motions of large earthquakes (Mw ≥ 5.8 magnitude). Moreover, for accurate simulations of actual earthquakes with complex rupture processes having strong localized slips, incorporating finite faults is essential even for more minor earthquakes.

양산시 동면 금산리 일원의 공사현장 사면 3개 지점에서 미고결 퇴적층을 절단하는 단층이 확인되었으며, 노두 단면에서 관찰되는 단층의 상세 구조분석을 수행하였다. 이곳 금산리 지점은 기존에 제4기 단층운동이 보고된 가산단층 지점으로부터 북쪽으로 약 0 .6 k m 떨어진 곳에 위치한다. 관찰된 총 6조의 단층들은 14o-32oE 주향을 가지고 3조의 단 층들은 77o-87oNW, 나머지 3조의 단층들은 53o-62oSE로 경사진다. 단층에 의해 절단된 미고결 퇴적층은 동편의 금정산 에서 유래된 선상지 역암으로 주로 화강암 또는 화산암 기원의 직경 0.5m 이상의 거력으로 구성된다. 단층면 상에 발 달하는 단층조선은 역이동성 성분이 포함된 우수향 주향이동단층 운동감각을 지시하며, 이러한 변형특성은 한반도 현생 응력환경인 동북동-서남서 압축응력과 부합한다. 사면에서 관찰되는 기반암과 미고결 퇴적층과의 부정합면을 기준으로 산정한 단층의 겉보기 수직변위는 동편이 15 m, 서편이 1 m이다.

원자력발전소 지진 확률론적 안전성 평가인 PSA(Probabilistic Safety Assessment)는 오랜 기간에 걸쳐 확고히 구축되어 왔다. 반면 에 다양한 공정 기반의 산업시설물의 경우 화재, 폭발, 확산(유출) 재난에 대해 주로 연구되어 왔으며, 지진에 대해서는 상대적으로 연 구가 미미하였다. 하지만, 플랜트 설계 당시와 달리 해당 부지가 지진 영향권에 들어갈 경우 지진 PSA 수행은 필수적이다. 지진 PSA 를 수행하기 위해서는 확률론적 지진 재해도 해석(Probabilistic Seismic Hazard Analysis), 사건수목 해석(Event Tree Analysis), 고장수 목 해석(Fault Tree Analysis), 취약도 곡선 등을 필요로 한다. 원자력 발전소의 경우 노심 손상 방지라는 최우선 목표에 따라 많은 사고 시나리오 분석을 통해 사건수목이 구축되었지만, 산업시설물의 경우 공정의 다양성과 최우선 손상 방지 핵심설비의 부재로 인해 일 반적인 사건수목 구축이 어렵다. 따라서, 본 연구에서는 산업시설물 지진 PSA를 수행하기 위해 고장수목을 바탕으로 확률론적 시각 도구인 베이지안 네트워크(Bayesian Network, BN)로 변환하여 리스크를 평가하는 방법을 제안한다. 제안된 방법을 이용하여 임의로 생성된 가스플랜트 Plot Plan에 대해 최종 BN을 구축하고, 다양한 사건 경우에 대한 효용성있는 의사결정과정을 보임으로써 그 우수 성을 확인하였다.

In the era of the 4th Industrial Revolution, Logistic 4.0 using data-based technologies such as IoT, Bigdata, and AI is a keystone to logistics intelligence. In particular, the AI technology such as prognostics and health management for the maintenance of logistics facilities is being in the spotlight. In order to ensure the reliability of the facilities, Time-Based Maintenance (TBM) can be performed in every certain period of time, but this causes excessive maintenance costs and has limitations in preventing sudden failures and accidents. On the other hand, the predictive maintenance using AI fault diagnosis model can do not only overcome the limitation of TBM by automatically detecting abnormalities in logistics facilities, but also offer more advantages by predicting future failures and allowing proactive measures to ensure stable and reliable system management. In order to train and predict with AI machine learning model, data needs to be collected, processed, and analyzed. In this study, we have develop a system that utilizes an AI detection model that can detect abnormalities of logistics rotational equipment and diagnose their fault types. In the discussion, we will explain the entire experimental processes : experimental design, data collection procedure, signal processing methods, feature analysis methods, and the model development.

It is essential to determine a proper earthquake time history as a seismic load in a seismic design for a critical structure. In the code, a seismic load should satisfy a design response spectrum and include the characteristic of a target fault. The characteristic of a fault can be represented by a definition of a type of possible earthquake time history shape that occurred in a target fault. In this paper, the pseudo-basis function is proposed to be used to construct a specific type of earthquake, including the characteristic of a target fault. The pseudo-basis function is derived from analyzing the earthquake time history of specific fault harmonic wavelet transform. To show the feasibility of this method, the proposed method was applied to the faults causing the Gyeong-Ju ML5.8 and Pohang ML5.3 earthquakes.

Failure diagnoses on large diesel engine are commonly detected when a deviation or fluctuation in its temperature, pressure, vibration or noise set parameter limits arises. These parameters can be easily monitored and can provide information of the engine’s present state depending on external environment and operating conditions. On the other hand, long term monitoring and condition management can be interfaced into the engine’s existing operating system. The approach is seen to keep track of monitored machines’ status using resonance and vibration amplitude. In particular, these signals will be able to identify complex vibration characteristic pertaining to such as engine torque output and support mounts. In this paper, a basic research for large diesel engine diagnosis was carried-out. The failure diagnosis collects and monitors the vibration state time history by using various vibration signals with reference to ISO 13373-1. Further, this monitoring system in the field of large diesel engines has not been applied practically and the results of this study are presented herein.

The recent increase in earthquake activities has highlighted the importance of seismic performance evaluation for civil infrastructures. In particular, the container crane essential to maintaining the national logistics system with port operation requires an exact evaluation of its seismic response. Thus, this study aims to assess the seismic vulnerability of container cranes considering their seismic characteristics. The seismic response of the container crane should account for the structural members’ yielding and buckling, as well as the crane wheel’s uplifting derailment in operation. The crane’s yielding and buckling limit states were defined using the stress of crane members based on the load and displacement curve obtained from nonlinear static analysis. The derailment limit state was based on the height of the rail, and nonlinear dynamic analysis was performed to obtain the seismic fragility curves considering defined limit states and seismic characteristics. The yield and derailment probabilities of the crane in the near-fault ground motion were approximately 1.5 to 4.7 and 2.8 to 6.8 times higher, respectively, than those in the far-fault ground motion.

In this paper, a study was conducted on the analysis of communication circuit faults using oscilloscope waveform analysis. Circuit resistance was calculated based on voltage and operating current values using a simple equation, and it was confirmed that the increase in resistance of the communication circuit could be analyzed by analyzing the voltage level during transmitter operation. By combining information of the controller ID, the location of the fault was identified and it was concluded that the location of the fault can be quickly found by analyzing the oscilloscope waveform and the controller ID information. Additionally, the value of communication line contact resistance can be calculated using a simple equation, and the location of the fault can be found by analyzing the communication voltage level and ID information.

Elevators are the main means of transport in buildings. A malfunction of an elevator in operation may cause in convenience to users. Furthermore, fatal accidents, such as injuries and death, may occur to the passengers also. Therefore, it is important to prevent failure before accidents happen. In related studies, preventive measures are proposed through analyzing failures, and the lifespan of elevator components. However, these methods are limited to existing an elevator model and its surroundings, including operating conditions and installed environments. Vibration occurs when the elevator is operated. Experts have classified types of faults, which are symptoms for malfunctions (failures), via analyzing vibration. This study proposes an artificial intelligent model for classifying faults automatically with deep learning algorithms through elevator vibration data, hereby preventing failures before they occur. In this study, the vibration data of six elevators are collected. The proposed methodology in this paper removes "the measurement error data" with incorrect measurements and extracts operating sections from the input datasets for proceeding deep learning models. As a result of comparing the performance of training five deep learning models, the maximum performance indicates Accuracy 97% and F1 Score 97%, respectively. This paper presents an artificial intelligent model for detecting elevator fault automatically. The users’ safety and convenience may increase by detecting fault prior to the fatal malfunctions. In addition, it is possible to reduce manpower and time by assisting experts who have previously classified faults.

역사지진과 계기지진 기록에 따르면 한반도 남동부는 우리나라에서 지진활성도가 가장 높게 평가되는 곳으로, 최근에 양산단층대와 울산단층대를 따라 제4기 단층이 다수 보고되어 고지진학적 연구가 활발하게 이루어지고 있다. 특 히 울산단층대의 중부지역에 해당하는 경북 경주시 외동읍 말방리 일원은 울산단층대 내에서 가장 많은 활성단층이 보고된 지역이다. 따라서 이 지역에 대한 고지진학적 특성을 이해하기 위하여 먼저 LiDAR 영상 및 항공사진을 이용한 지형 및 선형구조 분석을 실시하여 단층에 의한 기복으로 추정되는 지형인자를 확인하고, 야외답사와 물리탐사를 통해 단층을 추적하여 기 보고된 말방단층 지점에서 약 300 m 북서쪽에 위치한 곳에서 길이 20 m, 너비 5 m, 깊이 5 m의 굴착조사를 실시하였다. 굴착단면을 통해 분석된 제4기 퇴적층의 특징을 바탕으로 단층의 기하학적·운동학적 특성을 해석하여 고지진학적 특성을 규명하고자 하였다. 이번 굴착단면에서 확인된 역단층의 기하를 보이는 단층의 자세는 N26oW/33oNE로 울산단층대를 따라 분포하는 기 보고된 단층들과 유사하다. 약 40 cm의 단일 겉보기 변위가 인지되었 으나 단층조선의 부재로 실변위는 산출할 수 없었다. 선행연구에서 제안된 극저온구조층의 연대결과 값을 토대로 단층 의 최후기 운동시기는 후기 뷔름빙기 이전으로 추정하였다. 기 보고된 연구결과와 본 굴착단면에서 획득한 단층기하를 종합하여 이 지역에 발달하는 단층계를 인편상구조로 해석하였고, 단층특성을 반영한 모델을 제시하였다. 말방리 일원 에서 수 회의 굴착조사를 비롯한 다수의 선행연구가 수행되었음에도 불구하고 구체적인 단층변수에 대한 정보가 미진 하고 각 지점들 간의 상관성이 명확하게 규명되지 않은 것은 역단층의 복잡한 운동학적 특성에 기인한 것으로 판단된 다. 추후 고지진학적 연구가 추가적으로 수행된다면 상기의 문제점들을 해결하여 종합적인 단층의 형태와 운동사가 규 명될 수 있을 것으로 판단된다.

The stochastic method is applied to simulate strong ground motions at seismic stations of seven metropolises in South Korea, creating an earthquake scenario based on the causative fault of the 2016 Gyeongju earthquake. Input parameters are established according to what has been revealed so far for the causative fault of the Gyeongju earthquake, while the ratio of differences in response spectra between observed and simulated strong ground motions is assumed to be an adjustment factor. The calculations confirm the applicability and reproducibility of strong ground motion simulations based on the relatively small bias in response spectra between observed and simulated strong ground motions. Based on this result, strong ground motions by a scenario earthquake on the causative fault of the Gyeongju earthquake with moment magnitude 6.5 are simulated, assuming that the ratios of its fault length to width are 2:1, 3:1, and 4:1. The results are similar to those of the empirical Green’s function method. Although actual site response factors of seismic stations should be supplemented later, the simulated strong ground motions can be used as input data for developing ground motion prediction equations and input data for calculating the design response spectra of major facilities in South Korea.

As mechatronic systems have various, complex functions and require high performance, automatic fault detection is necessary for secure operation in manufacturing processes. For conducting automatic and real-time fault detection in modern mechatronic systems, multiple sensor signals are collected by internet of things technologies. Since traditional statistical control charts or machine learning approaches show significant results with unified and solid density models under normal operating states but they have limitations with scattered signal models under normal states, many pattern extraction and matching approaches have been paid attention. Signal discretization-based pattern extraction methods are one of popular signal analyses, which reduce the size of the given datasets as much as possible as well as highlight significant and inherent signal behaviors. Since general pattern extraction methods are usually conducted with a fixed size of time segmentation, they can easily cut off significant behaviors, and consequently the performance of the extracted fault patterns will be reduced. In this regard, adjustable time segmentation is proposed to extract much meaningful fault patterns in multiple sensor signals. By considering inflection points of signals, we determine the optimal cut-points of time segments in each sensor signal. In addition, to clarify the inflection points, we apply Savitzky-golay filter to the original datasets. To validate and verify the performance of the proposed segmentation, the dataset collected from an aircraft engine (provided by NASA prognostics center) is used to fault pattern extraction. As a result, the proposed adjustable time segmentation shows better performance in fault pattern extraction.

A lot of sensor and control signals is generated by an industrial controller and related internet-of-things in discrete manufacturing system. The acquired signals are such records indicating whether several process operations have been correctly conducted or not in the system, therefore they are usually composed of binary numbers. For example, once a certain sensor turns on, the corresponding value is changed from 0 to 1, and it means the process is finished the previous operation and ready to conduct next operation. If an actuator starts to move, the corresponding value is changed from 0 to 1 and it indicates the corresponding operation is been conducting. Because traditional fault detection approaches are generally conducted with analog sensor signals and the signals show stationary during normal operation states, it is not simple to identify whether the manufacturing process works properly via conventional fault detection methods. However, digital control signals collected from a programmable logic controller continuously vary during normal process operation in order to show inherent sequence information which indicates the conducting operation tasks. Therefore, in this research, it is proposed to a recurrent neural network-based fault detection approach for considering sequential patterns in normal states of the manufacturing process. Using the constructed long short-term memory based fault detection, it is possible to predict the next control signals and detect faulty states by compared the predicted and real control signals in real-time. We validated and verified the proposed fault detection methods using digital control signals which are collected from a laser marking process, and the method provide good detection performance only using binary values.