• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.20 no.1
• /
• pp.163-169
• /
• 2020

Recently, research on automation and unmanned operation of machines in the industrial field has been conducted with the advent of AI, Big data, and the IoT, which are the core technologies of the Fourth Industrial Revolution. The machines for these automation processes are controlled based on the data collected from the sensors attached to them, and further, the processes are managed. Conventionally, the abnormalities of sensors are periodically checked and managed. However, due to various environmental factors and situations in the industrial field, there are cases where the inspection due to the failure is not missed or failures are not detected to prevent damage due to sensor failure. In addition, even if a failure occurs, it is not immediately detected, which worsens the process loss. Therefore, in order to prevent damage caused by such a sudden sensor failure, it is necessary to identify the failure of the sensor in an embedded system in real-time and to diagnose the failure and determine the type for a quick response. In this paper, a deep neural network-based fault diagnosis system is designed and implemented using Raspberry Pi to classify typical sensor fault types such as erratic fault, hard-over fault, spike fault, and stuck fault. In order to diagnose sensor failure, the network is constructed using Google's proposed Inverted residual block structure of MobilieNetV2. The proposed scheme reduces memory usage and improves the performance of the conventional CNN technique to classify sensor faults.

• Journal of the Korea Society for Simulation
• /
• v.18 no.4
• /
• pp.67-79
• /
• 2009

Because a product in the car industry has a short life cycle in recent years, the process planning and the manufacturing lines have to be changed frequently. Most of time, repositioning an existing facility and modifying used control information are faster than making completely new process planning. However, control information and control code such as PLC code are difficult to understand. Hence, industries prefer writing a new control code instead of using the existing complex one. It shows the lack of information reusability in the existing process planning. As a result, to reduce this redundancy and lack of reusability, we propose a SOS-Net modeling method. SOS-Net is a standard methodology used to describe control information. It is based on the Device Structure which consists of sensor information derived from device hardware information. Thus, SOS-Net can describe a real control state for automated manufacturing systems. The SOS-Net model is easy to understand and can be converted into PLC Code easily. It also enables to modify control information, thus increases the reusability of the new process planning. Proposed model in this paper plays an intermediary role between the process planning and PLC code generation. It can reduce the process planning and implementation time as well as cost.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.41-42
• /
• 2021

The Green New Deal policy was established to innovate the government's energy consumption structure, establish a third basic energy plan to strengthen the global competitiveness of the energy industry, and realize a carbon neutral society due to the increased need for transition to a low-carbon economy. Waterworks such as drinking water, water purification plant, and pressurization plant analyze control factors and energy consumption status by process to improve energy management efficiency and reduce energy usage through the 4th industrial revolution. Ultimately, we want to realize net-zero water purification plant.

• Journal of Korea Multimedia Society
• /
• v.6 no.7
• /
• pp.1245-1255
• /
• 2003

Request for network becomes complicated gradually and network traffic is increasing, To overcome this situation, the structure of network node should be changed to accept new service quickly and economically by executing program code in node itself. Active Network's research can use net resources more properly because of executing program within node itself. In this paper, we design a programming language, namely ASL, for Active Sensor to describe function and behavior of active sensor. We also design and implement the interpreter for proposed ASL.

• Journal of the Optical Society of Korea
• /
• v.19 no.6
• /
• pp.658-664
• /
• 2015

Lens designers generally refer to flat image fields and attempt to minimize the field curvature. Present-day CMOS image sensors for mobile phone cameras, however, are not flat, but curved. Sometimes it is necessary to generate an intentional field curvature according to the degree and direction of the CMOS image-sensor’s curvature. This paper presents the degree of curvature of a CMOS image sensor measured using an interferometer, and proposes an effective compensation method that minimizes the net field curvature through optimizing the air gap between lens elements, which is demonstrated using simulations and experiments.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2014.07a
• /
• pp.35-36
• /
• 2014

본 논문에서는 Kinect 센서의 동작 인식 기능을 활용한 운동요법 앱을 설계하고 구현한다. 이 앱은 사용자의 상체, 하체 관절의 움직임을 인식하여 체조의 올바른 자세를 배울 수 있도록 한다. 이 앱의 특징은 Kinect Sensor로 인식한 관절 요소를 읽어 들이고, 각 관절의 각도를 계산하여 원하는 동작을 표현할 수 있도록 한다. 또한, 사용자의 동작과 기본 동작을 비교하여 오차 범위 내이면 새로운 동작이 진행되도록 한다. 출력되는 각 동작들은 Library를 이용하여 많은 포즈들을 입력하고 출력할 수 있기 때문에 새로운 포즈를 추가하기에 쉽다.

• Nuclear Engineering and Technology
• /
• v.54 no.9
• /
• pp.3283-3292
• /
• 2022

Based on the Deep Q-Network(DQN) algorithm of reinforcement learning, an active fault-tolerance method with incremental action is proposed for the control system with sensor faults of the once-through steam generator(OTSG). In this paper, we first establish the OTSG model as the interaction environment for the agent of reinforcement learning. The reinforcement learning agent chooses an action according to the system state obtained by the pressure sensor, the incremental action can gradually approach the optimal strategy for the current fault, and then the agent updates the network by different rewards obtained in the interaction process. In this way, we can transform the active fault tolerant control process of the OTSG to the reinforcement learning agent's decision-making process. The comparison experiments compared with the traditional reinforcement learning algorithm(RL) with fixed strategies show that the active fault-tolerant controller designed in this paper can accurately and rapidly control under sensor faults so that the pressure of the OTSG can be stabilized near the set-point value, and the OTSG can run normally and stably.

• Nuclear Engineering and Technology
• /
• v.55 no.8
• /
• pp.3114-3120
• /
• 2023

A high-accuracy magnetic field measurement device based on a cyclotron is being developed for accelerator mass spectrometry (AMS). In this study, a magnetic field measurement device consisting of a Hall probe sensor, piezo-motor, and step motor was developed to measure the magnetic field of the AMS cyclotron magnet. The Hall probe sensor was calibrated to achieve positional accuracy by using polar coordinates. The measurement results between the ratchet gear and piezo-motor, which are the instruments used for driving the measurement device, were analyzed. The measurement result of the device with a piezo-motor exhibits a difference of 5 Gauss (0.04%) as compared with the simulation result.

• Nuclear Engineering and Technology
• /
• v.52 no.11
• /
• pp.2535-2542
• /
• 2020

To understand the fundamental parameters of Alvand tokamak, A Rogowski coil with an active integrator was designed and constructed. Considering the characteristics of the Alvand tokamak, the structural and electrical parameters affecting the sensor function, were designed. Calibration was performed directly in the presence of plasma. The sensor has a high resistance against interference of external magnetic fields. Plasma current was measured in various experiments. Based on the plasma current profile and loop voltage signal, the time evolution of plasma discharge was investigated and plasma behavior was analyzed. Alvand tokamak discharge was divided into several regions that represents different physical phenomena in the plasma. During the plasma discharge time, plasma had significant changes and its characteristic was not uniform. To understand the plasma behavior in each of the phases, the Rogowski sensor should have sufficient time resolution. The Rogowski sensor with a frequency up to 15 kHz was appropriate for this purpose.

• The Journal of Korean Institute of Information Technology
• /
• v.17 no.8
• /
• pp.11-19
• /
• 2019

Against advanced image-related crimes, a high level of digital forensic methods is required. However, feature-based methods are difficult to respond to new device features by utilizing human-designed features, and deep learning-based methods should improve accuracy. This paper proposes a deep learning model to identify camera models based on DenseNet, the recent technology in the deep learning model field. To extract camera sensor features, a HPF feature extraction filter was applied. For camera model identification, we modified the number of hierarchical iterations and eliminated the Bottleneck layer and compression processing used to reduce computation. The proposed model was analyzed using the Dresden database and achieved an accuracy of 99.65% for 14 camera models. We achieved higher accuracy than previous studies and overcome their disadvantages with low accuracy for the same manufacturer.