• 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.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.2
• /
• pp.1005-1018
• /
• 2017

This paper proposes an App Visualization (AppV) based on IoT Self-diagnosis Micro Control Unit (ISMCU) for accident prevention. It collects a current status of a vehicle through a sensor, visualizes it on a smart phone and prevents vehicles from accident. The AppV consists of 5 components. First, a Sensor Layer (SL) judges noxious gas from a current vehicle and a driver's driving habit by collecting data from various sensors such as an Accelerator Position Sensor, an O2 sensor, an Oil Pressure Sensor, etc. and computing the concentration of the CO collected by a semiconductor gas sensor. Second, a Wireless Sensor Communication Layer (WSCL) supports Zigbee, Wi-Fi, and Bluetooth protocol so that it may transfer the sensor data collected in the SL to ISMCU and the data in the ISMCU to a Mobile. Third, an ISMCU integrates the transferred sensor information and transfers the integrated result to a Mobile. Fourth, a Mobile App Block Programming Tool (MABPT) is an independent App generation tool that changes to visual data just the vehicle information which drivers want from a smart phone. Fifth, an Embedded Module (EM) records the data collected through a Smart Phone real time in a Cloud Server. Therefore, because the AppV checks a vehicle' fault and bad driving habits that are not known from sensors and performs self-diagnosis through a mobile, it can reduce time and cost spending on accidents caused by a vehicle's fault and noxious gas emitted to the outside.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.5
• /
• pp.55-67
• /
• 2022

Sensor data can provide fault diagnosis for equipment. However, the cause analysis for fault results of equipment is not often provided. In this study, we propose an explainable convolutional neural network framework for the sensor-based time series classification model. We used sensor-based time series dataset, acquired from vehicles equipped with sensors, and the Wafer dataset, acquired from manufacturing process. Moreover, we used Cycle Signal dataset, acquired from real world mechanical equipment, and for Data augmentation methods, scaling and jittering were used to train our deep learning models. In addition, our proposed classification models are convolutional neural network based models, FCN, 1D-CNN, and ResNet, to compare evaluations for each model. Our experimental results show that the ResNet provides promising results in the context of time series classification with accuracy and F1 Score reaching 95%, improved by 3% compared to the previous study. Furthermore, we propose XAI methods, Class Activation Map and Layer Visualization, to interpret the experiment result. XAI methods can visualize the time series interval that shows important factors for sensor data classification.

• Geophysics and Geophysical Exploration
• /
• v.7 no.2
• /
• pp.105-108
• /
• 2004

Immersive virtual reality has been used in areas of oil and gas exploration for visualization and analysis of various spatial data, such as wireline logs, 3-dimensional seismic data volume, formational boundaries, fault, and some other reservoir characteristics. Although virtual reality is a valuable tool in this area, in most cases, it requires a large budget. This paper describes the construction of a single screen, passive stereo, virtual reality, display system based on commodity, or otherwise, low-cost components. The core elements of the system are a PC with a two-channel 3-D graphics, two projects, and a polarized stereo. There are many options available for the major elements of such a system, and the basic system can be modified or adapted to many different styles of use.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2001.03a
• /
• pp.501-508
• /
• 2001

A graph or topographic map can often convey larger amounts of information in a shorter time than ordinary text-based methods. To visualize information precisely it is necessary to collect all the geological information at design stage, but actually it is almost impossible to bore or explore the entire area to gather the required data. So, tunnel engineers have to rely on the judgement of expert from the limited number of the results of exploration and experiment. In this study, several programs are developed to handle the results of geological investigation with various data processing techniques. The results of the typical case study are also presented. For the electric survey, eleven points are chosen at the valley to measure the resistivity using Schlumberger array. The measured data are interpolated in 3-dimensional space by kriging and the distribution of resistivity are visualized to find weak or fractured zone. The correlation length appears to be around 5 to 20 meter in depth. Regression analyses were performed to find a correlation length. No nugget effect is assumed, and the topographic map, geologic formation, fault zone, joint geometry and the distribution of resistivity are successfully visualized by using the proposed technique.

• Economic and Environmental Geology
• /
• v.54 no.2
• /
• pp.177-186
• /
• 2021

For disaster management and mitigation of earthquakes in Korea Peninsula, active fault investigation has been conducted for the past 5 years. In particular, investigation of sediment-covered active faults integrates geomorphological analysis on airborne LiDAR data, surface geological survey, and geophysical exploration, and unearths subsurface active faults by trench survey. However, the fault traces revealed by trench surveys are only available for investigation during a limited time and restored to the previous condition. Thus, the geological data describing the fault trench sites remain as the qualitative data in terms of research articles and reports. To extend the limitations due to temporal nature of geological studies, we utilized a terrestrial LiDAR to produce 3D point clouds for the fault trench sites and restored them in a digital space. The terrestrial LiDAR scanning was conducted at two trench sites located near the Yangsan Fault and acquired amplitude and reflectance from the surveyed area as well as color information by combining photogrammetry with the LiDAR system. The scanned data were merged to form the 3D point clouds having the average geometric error of 0.003 m, which exhibited the sufficient accuracy to restore the details of the surveyed trench sites. However, we found more post-processing on the scanned data would be necessary because the amplitudes and reflectances of the point clouds varied depending on the scan positions and the colors of the trench surfaces were captured differently depending on the light exposures available at the time. Such point clouds are pretty large in size and visualized through a limited set of softwares, which limits data sharing among researchers. As an alternative, we suggested Potree, an open-source web-based platform, to visualize the point clouds of the trench sites. In this study, as a result, we identified that terrestrial LiDAR data can be practical to increase reproducibility of geological field studies and easily accessible by researchers and students in Earth Sciences.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.23 no.4
• /
• pp.92-100
• /
• 2009

This paper presents a voltage sag assessment program. The program provides various functions for stochastic assessment of voltage sags such as short-circuit analysis, the determination of the area of vulnerability and the calculation of expected sag frequency(ESF). Effective data visualization functions based on computer graphics and animation were also implemented in the developed program. In this paper, the concept of voltage sag assessment and the assessment method considering generator scheduling and time-varying fault rates are presented. The influence of generator scheduling and time-varying fault rates on voltage sag prediction is also described by performing case studies using the developed program.

• Journal of Electrical Engineering and information Science
• /
• v.2 no.6
• /
• pp.64-73
• /
• 1997

Network technology has been developed for very high-speed networking and multimedia data whose characteristics are the continuous and bursty transmission as well as a large amount of data. With this trend users wish to view the information about the application services as well as network devices and system hardware. However, it is rarely available for the users the information of performance or faults of the application services. Most of information is limited to the information related network devices or system hardware. Furthermore, users expect the best services without knowing the service environments in the network and there is no good way of delivering the service related problems and fault information of application services in a high speed network yet. In this paper we present a web-based application management system that we have developed for the past year. It includes a method to build an agent system that uses an existing network management standards, SNMP MIB and SNMP protocols. The user interface of the system is also developed to support visualization effects with web-based Java interface which offers a convenient way not only to access management information but also to control networked applications.

• Journal of the Korean earth science society
• /
• v.36 no.3
• /
• pp.280-299
• /
• 2015

The purpose of this study was to investigate the effects of an instruction using geologic planar figures on high school students' ability of spatial visualization and geologic spatial ability and also to explore its applicability as an instructional strategy through the investigation of students' perception about the instruction using mixed methodology. For this purpose, we developed 10 planar figures of geologic structures (2 horizontal layers, 2 vertical layers, 4 angular layers, 1 fault, and 1 fold), and tested students' spatial visualization ability and geologic spatial ability before and after the implementation in class. In addition, in order to investigate students' perception on the instruction, we conducted quantitative research using questionnaires comprised of the cognitive and the affective domain, and followed by focus group interview that was conducted to obtain deeper understanding of their perception. Findings revealed that the instruction using geologic planar figures was effective to enhance spatial visualization ability and geologic spatial ability. It was also helpful for students to enhance their ability to perceive the spatial configuration of the geologic structures as well as the ability to penetrate visually into the images of the structures. The results of the students' perception on the instruction showed that the students recognized the instruction using geologic planar figures as a strongly positive teaching method both in the cognitive and affective domain. We concluded that geologic planar figures could be used as an effective tool for the lesson of 'mapping and interpreting of geological map', and be highly applicable for the advanced class in high schools.

• Journal of the Korean Society for Aviation and Aeronautics
• /
• v.18 no.4
• /
• pp.9-15
• /
• 2010

This paper presents a duplex flight control system of design concepts and sensor fault detection algorithm using Kalman Filter. The algorithm was verified to use HILS that is composed of two FCCs, motion table, visualization system, cockpit, and flight model computer. The FCC was developed to be able to mount on small aircraft.