• 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.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.27 no.6
• /
• pp.39-46
• /
• 2023

In this study, the dynamic behavior of friction bearings of PSC (Pre-Stressed Concrete) box train continuous bridge was numerically analyzed at 10 km/h intervals up to 600 km/h according to the increasing speed of the next-generation high-speed train. A frame model was generated targeting the 40-meter single-span and two-span continuous PSC box bridges in the Gyeongbu High-Speed Railway section. The interaction forces including the inertial mass vehicle model with 38 degrees of freedom and the irregularities of the bridge and track were considered. It was calculated the longitudinal displacement, cumulative sliding distance and displacement speed of the bridge bearings at each running speed so that compared with the dynamic behavior trend analysis of the bridge. In addition, long-term friction test standards were applied to evaluate the durability of friction plates.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.36 no.11
• /
• pp.1335-1343
• /
• 2012

In this study, the lightweight modular design of hybrid railway carbody structures made of sandwich composites and aluminum extrusions was investigated by using topology and size optimization techniques. The topology optimum design was used to select the best material for parts of the carbody structure at the initial design stage, and then, the size optimum design was used to find the optimal design parameters of hybrid carbody structures using first-order and sub-problem methods. Through the topology optimization analysis, it was found that aluminum extrusions were suitable for primary members such as the underframe and lower side panel module to improve the stiffness and manufacturability of the carbody structures, and sandwich composites were appropriate for secondary members such as the roof and middle side panel module to minimize its weight. Furthermore, the results obtained by size optimization analysis showed that the weight of hybrid carbody structures composed of aluminum extrusions and sandwich composites could be reduced by a maximum of approximately 17.7% in comparison with carbody structures made of only sandwich composites.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2017.05a
• /
• pp.279-282
• /
• 2017

Kinect depth sensor, a depth camera developed by Microsoft as a natural user interface for game appeared as a very useful tool in computer vision field. In this paper, due to kinect's depth sensor and its high frame rate, we developed a distance measurement system using Kinect camera to test it for unmanned vehicles which need vision systems to perceive the surrounding environment like human do in order to detect objects in their path. Therefore, kinect depth sensor is used to detect objects in its field of view and enhance the distance measurement system from objects to the vision sensor. Detected object is identified in accuracy way to determine if it is a real object or a pixel nose to reduce the processing time by ignoring pixels which are not a part of a real object. Using depth segmentation techniques along with Open CV library for image processing, we can identify present objects within Kinect camera's field of view and measure the distance from them to the sensor. Tests show promising results that this system can be used as well for autonomous vehicles equipped with low-cost range sensor, Kinect camera, for further processing depending on the application type when they reach a certain distance far from detected objects.

• Journal of Korean Society of Forest Science
• /
• v.100 no.2
• /
• pp.154-164
• /
• 2011

This study was conducted to develop the semi-crawler type mini-forwarder that can be operated comfortable small-scale logging operation in the steep terrain and also used at a variety of operations such as the civil work in erosion control and forest-road. Considering the minimum turning radius and the width of forest operation road, the total length, width and loading capacity of the semi-crawler type mini-forwarder is 5,750 mm, 1,900 mm and $2.5m^{3}$, respectively. The maximum engine power is 96ps at 3600 rpm. Selected hydraulic pumps are consists of two main pumps and two sub-main pumps. Main hydraulic pumps are utilized to running motor of the front wheel and rear crawler. Sub-main pumps are utilized to the actuation parts such as steering, crane, out-rigger and dump cylinder. The transmission was adapted as the HST (Hydro-Static Transmission) system. The driving parts are designed and manufactured as the front wheel type and the rear crawler type. The steering type was manufactured as the ackerman type. Driving control parts type was designed and manufactured as driver's seat type of normal cars. It is also attached on auxiliary equipments such as winch, log grapple and out-rigger. The traveling speed of the semi-crawler type mini-forwarder in forest road was 5.3 km/hr to 7.7 km/hr.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.22 no.3
• /
• pp.88-103
• /
• 2023

With the recent advancement of 4th Industrial Revolution technology, transportation systems are generating large amounts of mobility data related to the individual movement trajectories of vehicles and people. There are many constraints on utilizing mobility data containing personal information. Thus, in South Korea, the processing and generation of pseudonymized information and the analysis and utilization of this information have been managed in a dual manner by applying separate agencies and technologies through the revision of the Data 3 Act and the enactment of the Data Basic Act. However, this dual approach fails to securely support the entire data lifecycle and suffers from inefficiencies in terms of processing time and cost. Therefore, to compensate for the problems of the existing Expert Data Combination System and Data Safety Zone, this study proposes an Integrated Data Safety Zone Framework that integrates and unifies the process of generating, processing, analyzing, and utilizing mobility data. The integrated process for data processing was redesigned, and common requirements and core technologies were derived. The result is an architecture for a next-generation Integrated Data Safety Zone system that can manage and utilize the entire life cycle of mobility data at one stop.