• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.12 no.10
• /
• pp.799-807
• /
• 2002

The vibration signal can give an indication of the condition of rotating machinery, highlighting potential faults such as unbalance, misalignment and bearing defects. The features in the vibration signal provide an important source of information for the faults diagnosis of rotating machinery. When additional training data become available after the initial training is completed, the conventional neural networks (NNs) must be retrained by applying total data including additional training data. This paper proposes the fault diagnostics algorithm using the ART-Kohonen network which does not destroy the initial training and can adapt additional training data that is suitable for the classification of machine condition. The results of the experiments confirm that the proposed algorithm performs better than other NNs as the self-organizing feature maps (SOFM) , learning vector quantization (LYQ) and radial basis function (RBF) NNs with respect to classification quality. The classification success rate for the ART-Kohonen network was 94 o/o and for the SOFM, LYQ and RBF network were 93 %, 93 % and 89 % respectively.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.6 no.2
• /
• pp.294-299
• /
• 2002

This paper presents a faults diagnosis technique of induction motors based on a neural network. Only stator current is measured, transformed by using FFT and normalized for the training. Healthy, bearing fault, stator fault and rotor end-ring fault motors are prepared to obtain the learning data and diagnose the several faults. For more effective diagnosis, the load rate is changed by 100%, 60%, 30% of full load and the obtained are applied to the teaming process. The experimental results show the proposed method is very detectable and applicable to the real diagnosis system.

• Proceedings of the KSR Conference
• /
• 2011.05a
• /
• pp.1502-1508
• /
• 2011

Ground anchor expands the hollow wall of settled part and has the structure which resists the designed tensile load by the bearing pressure generated by the wedge of the anchor body pressing in the expanded part. Such ground anchor has been recognized for stability and economicality since 1960s in technologically advanced nations such as Japan and Europe, and in 1970s, the Japan Society of Soil Engineering has established and announced the anchor concept map. The ground anchor introduced in Korea, however, has the structural problem where the tensile strength is comes only from the ground frictional force due to expansion of the wedge body. In an interval where the ground strength is locally reduced due to fault, discontinuation or such, this is pointed out as a critical weakness where the anchor body of around 1.0m must resist the tensile load. Also, in the installation of concrete block, the concentrated stress of concrete block constructed on the uneven rock surface causes damage, and many such issues in the anchor head have been reported. Thus, in this study, by using the expanded bit for precise expansion of settled part, the ground anchor system was completed so that the bearing pressure of ground anchor can be expressed as much as possible, and the bearing plate was inserted into the ground to resolve the existing issues of concrete block. Through numerical analysis and pullout test executed for verification of site applicability, the pullout-behavior characteristics of anchor was analyzed.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.35 no.3
• /
• pp.185-192
• /
• 2015

Since real-time monitoring systems, such as early fault detection, have been very important, an infrared thermography technique was proposed as a new diagnosis method. This study focused on damage detection and temperature characteristic analysis of ball bearings using the non-destructive, infrared thermography method. In this paper, for the reliability assessment, infrared experimental data were compared with finite element analysis (FEA) results from ANSYS. In this investigation, the temperature characteristics of ball bearing were analyzed under various loading conditions. Finally, it was confirmed that the infrared thermography technique was useful for the real-time detection of damage to bearings.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.19 no.3
• /
• pp.151-158
• /
• 2024

Although on-device artificial intelligence (AI) has gained attention to diagnosing machine faults in real time, most previous studies did not consider the model retraining and redeployment processes that must be performed in real-world industrial environments. Our study addresses this challenge by proposing an on-device AI-based real-time machine fault diagnosis system that utilizes continual learning. Our proposed system includes a lightweight convolutional neural network (CNN) model, a continual learning algorithm, and a real-time monitoring service. First, we developed a lightweight 1D CNN model to reduce the cost of model deployment and enable real-time inference on the target edge device with limited computing resources. We then compared the performance of five continual learning algorithms with three public bearing fault datasets and selected the most effective algorithm for our system. Finally, we implemented a real-time monitoring service using an open-source data visualization framework. In the performance comparison results between continual learning algorithms, we found that the replay-based algorithms outperformed the regularization-based algorithms, and the experience replay (ER) algorithm had the best diagnostic accuracy. We further tuned the number and length of data samples used for a memory buffer of the ER algorithm to maximize its performance. We confirmed that the performance of the ER algorithm becomes higher when a longer data length is used. Consequently, the proposed system showed an accuracy of 98.7%, while only 16.5% of the previous data was stored in memory buffer. Our lightweight CNN model was also able to diagnose a fault type of one data sample within 3.76 ms on the Raspberry Pi 4B device.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.8 no.5 s.39
• /
• pp.65-77
• /
• 2004

In this study, the base isolation systems for equipment in the NPP are presented and the responses of each isolation system are investigated. As for the base isolation systems, a natural rubber bearing (NRB) and a high damping rubber bearing (HDRB) are selected. As input motions, artificial time histories enveloping the US NRC RG 1.60 spectrum and the probability-based scenario earthquake spectra developed for the Korean nuclear power plant site as well as a typical near-fault earthquake record are used. Uniaxial, biaxial, and triaxial excitations are conducted with PGAs of 0.1, 0.2 and 0.25g. The reduction of the seismic forces transmitted to the equipment models are determined for different isolation systems and input motions.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.18 no.5
• /
• pp.718-723
• /
• 2008

Bearing is one of the important mechanical elements used in various industrial equipments. Most of failures occurred during the equipment operation result from bearing defects and breakages. Therefore, monitoring of bearings is essential in preventing equipment breakdowns and reducing unexpected loss. The purpose of this paper is to present an online monitoring method to predict bearing states using vibration signals. Bearing vibrations, which are collected as a form of profile signal, are first analyzed by a discrete wavelet transform. Next, some statistical features are obtained from the resultant wavelet coefficients. In order to select significant ones among them, analysis of variance (ANOVA) is employed in this paper. Statistical features screened in this way are used as input variables to support vector machine (SVM). An hierarchical SVM tree is proposed for dealing with multi-class problems. The result of numerical experiments shows that the proposed SVM tree has a competent performance for classifying bearing fault states.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.3
• /
• pp.845-850
• /
• 2023

In this paper, we propose a method for diagnosing ball bearing vibration using transfer learning. STFT, which can analyze vibration signals in time-frequency, was used as input to CNN to diagnose failures. In order to rapidly learn CNN-based deep artificial neural networks and improve diagnostic performance, we proposed a transfer learning-based deep learning learning technique. For transfer learning, the feature extractor and classifier were selectively learned using a VGG-based image classification model, the data set for learning was publicly available ball bearing vibration data provided by Case Western Reserve University, and performance was evaluated by comparing the proposed method with the existing CNN model. Experimental results not only prove that transfer learning is useful for condition diagnosis in ball bearing vibration data, but also allow other industries to use transfer learning to improve condition diagnosis.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2004.05a
• /
• pp.315-315
• /
• 2004

자기베어링 시스템은 액츄에이터, 센서, 제어기, 전류앰프 등으로 구성되어 있으며 시스템의 신뢰도는 각 구성 요소의 신뢰도와 구성요소의 상호 작용에 의해 결정된다. 자기 베어링 기술이 현재 보다 많은 분야에 적용되기 위해서는 신뢰도의 향상이 필수적이다. 본 논문에서는 자기베어링에서 사용되는 센서 중 일부가 작동하지 않더라도 시스템이 정상적으로 작동하는 센서 고장 허용 자기 베어링 시스템을 기술하였다.(중략)

• Smart Structures and Systems
• /
• v.16 no.4
• /
• pp.701-723
• /
• 2015

Friction isolators are one of the most important types of bearings used to mitigate damages of earthquakes. The adaptive behavior of these isolators allows them to achieve multiple levels of performances and predictable seismic behavior during different earthquake hazard levels. There are three main types of friction isolators. The first generation with one sliding surface is known as Friction Pendulum System (FPS) isolators. The double concave friction pendulum (DCFP) with two sliding surfaces is an advanced form of FPS, and the third one, with fully adaptive behavior, is named as triple concave friction pendulum (TCFP). The current study has been conducted to investigate and compare seismic responses of these three types of isolators. The structure is idealized as a two-dimensional single degree of freedom (SDOF) resting on isolators. The coupled differential equations of motion are derived and solved using state space formulation. Seismic responses of isolated structures using each one of these isolators are investigated under seven near fault earthquake motions. The peak values of bearing displacement and base shear are studied employing the variation of essential parameters such as superstructure period, effective isolation period and effective damping of isolator. The results demonstrate a more efficient seismic behavior of TCFP isolator comparing to the other types of isolators. This efficiency depends on the selected effective isolation period as well as effective isolation damping. The investigation shows that increasing the effective isolation period or decreasing the effective isolation damping improves the seismic behavior of TCFP compared to the other isolators. The maximum difference in seismic responses, the base shear and the bearing displacement, for the TCFP isolator are calculated 26.8 and 13.4 percent less than the DCFP and FPS in effective isolation damping equal to10%, respectively.