• Proceedings of the Korea Information Processing Society Conference
• /
• 2013.05a
• /
• pp.934-936
• /
• 2013

본 시스템은 회귀 알고리즘을 이용한 실시간 모니터링 무선관리 시스템으로 현재 대부분의 업무는 모니터링 시스템과 같은 관제 시스템 및 다수의 PC 통합 환경에 맞춰 대규모, 동적 관리가 필요한 시스템의 요구가 증가하고 있다. 또한 다양한 관제 시스템이 만들어지고 있지만 모니터링 시스템의 경우 관리자가 항상 대기 및 관찰하여 상황에 따른 에러에 대하여즉각적인 대처는 할 수 있지만 예측이나 예방을 하기는 힘든 실정이다. 또한 이러한 예측이 가능할 경우 즉각적인 처리를 위해서 에러가 예상되는 PC의 성능 및 작업을 실시간으로 어떻게 통보하여 처리해야 할지에 대한 방법도 필요한 실정이다. 본 연구에서는 회귀알고리즘을 적용하여 사용자 PC의 성능을 분석 한 뒤 일어날 수 있는 상황이나 에러 등을 PC나 모바일 APP을 통해 실시간으로 처리하기 위하여 이상 이벤트 관리 및 시스템자원 분석 모듈 구현, 다수의 PC를 관리하는 소프트웨어 설계와 구현을 통해 이러한 문제를 해결하여 다수의 PC를 관리하며 실시간적으로 여러 가지 이벤트를 예상하고 시스템의 효율과 최적화를 이루고자 하였다. 또한 모바일 APP을 활용하여 사용자 PC의 실시간 처리 및 관리가 가능하기 때문에 관리자가 좀 더 원활한 업무수행 및 대처에 도움을 줄 수 있다.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.117-127
• /
• 2022

Data anomalies seriously threaten the reliability of the bridge structural health monitoring system and may trigger system misjudgment. To overcome the above problem, an efficient and accurate data anomaly detection method is desiderated. Traditional anomaly detection methods extract various abnormal features as the key indicators to identify data anomalies. Then set thresholds artificially for various features to identify specific anomalies, which is the artificial experience method. However, limited by the poor generalization ability among sensors, this method often leads to high labor costs. Another approach to anomaly detection is a data-driven approach based on machine learning methods. Among these, the bidirectional long-short memory neural network (BiLSTM), as an effective classification method, excels at finding complex relationships in multivariate time series data. However, training unprocessed original signals often leads to low computation efficiency and poor convergence, for lacking appropriate feature selection. Therefore, this article combines the advantages of the two methods by proposing a deep learning method with manual experience statistical features fed into it. Experimental comparative studies illustrate that the BiLSTM model with appropriate feature input has an accuracy rate of over 87-94%. Meanwhile, this paper provides basic principles of data cleaning and discusses the typical features of various anomalies. Furthermore, the optimization strategies of the feature space selection based on artificial experience are also highlighted.

• Proceedings of the Korean Institute of Industrial Safety Conference
• /
• 2003.10a
• /
• pp.335-338
• /
• 2003

최근의 통계에 따르면 국내의 인터넷 인용율은 2000년도에 44.7%, 2001년도에 56.6%, 2002년도에는 55.2%로 아이슬랜드(60.8%), 스웨덴(57.3%)에 이어 세계에서 세번째로 높은 것으로 나타났다. 또한 인구 100명당 PC보급율 또한 미국(62.5대), 덴마크(57.68대)에 이어 한국(55.58대)이 세계3위를 기록해 명실상부한 정보화 선진국임이 입증되고 있다. 따라서 잘 발달된 infrastructure을 이용하는 기술개발은 초고속정보이용을 가능케하고 있다. 근래에 들어 인터넷을 기반으로 기본적인 활용인 문자, 음성 및 영상 전송 등의 데이터 전송뿐만 아니라, 근래에는 인터넷을 통한 remote control 및 monitoring system에 대한 연구가 활발히 진행되고 있다. 기존의 증류공정은 DCS(Dispersion Control System)로써 공정상태(온도, 압력, 유량, 레벨, 순도등)를 Control room까지 선로를 통하여 신호전달이 이루어지고 있다. 따라서, 제한된 지역에서만 I/O(Input/output)이 이루어지고 있다. 이로 인해 안전감시시스템의 효율성이 떨어지고 있는 게 사실이다. 본 연구에서는 PC기반의 프로그램을 바탕으로 인터넷을 이용한 remote control 및 monitoring system을 개발함으로써 close Network에서 Open Network으로 발전시켰다. 여기서 local plant의 안전시스템 알고리즘은 그대로 Server system에 도입하였으며 사용자의 실정에 맞게 손쉽게 설정을 할 수 있도록 확장성을 높였다. 이는 PC기반의 시스템을 구축함으로써 가능하게 된 것이다. 또한, 웹 상에서의 무차별적 접근으로부터 보호하기 위하여 감시 대상의 성질에 따라 프로그램상에서 차별적 접근형식을 택할 수 있으며, control room의 indicator들과 controller를 프로그램상에 그대로 이식시킴으로써 감시자에게 충분한 정보제공 및 원격제어가 가능하도록 하고있다.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.8 no.10
• /
• pp.1435-1441
• /
• 2013

This paper deals a watt-hour meter monitor system applying a Power Line Communication (PLC) MODEM and a Data Concentrator Unit (DCU) which can apply to Internet map based Graphical User Interface (GUI). The PLC MODEM is profitable by saving the cost of installing dedicated wiring, and Internet map GUI also is able to contribute an easy access to monitoring system for non export users. This system makes it possible to monitor the amount of electricity of previous/ current month metering data via Internet mapping service. The implement system can easily apply a PC-management for the electric charges and maintenance with gathering the metering data.

• Journal of the Institute of Convergence Signal Processing
• /
• v.16 no.2
• /
• pp.56-62
• /
• 2015

In this study, we implemented an unconstraint ECG monitoring system on a chair. Nowadays, modern people spend most of their time sitting on the chair. Therefore, it is necessary to have an unconstraint ECG monitoring system that can be used for a long time. The implemented system can perform measurement even with clothes on and it has great advantage on motion artifacts. A pair of unconstraint electrodes are placed on the back of the chair. Amplifier and filters are designed to remove motion artifacts, The baseline noise and power line noise are filtered and very low level of bio-signal is amplified to give a final measurement. Control unit and wireless transmission unit are implemented. Analog signal is converted into digital signal and transmits biological signal to the PC and the smart phone. Therefore continuous ECG monitoring in daily life is made possible. A comparison experiment between Ag / AgCl electrode and unconstraint electrode is conducted to evaluate the performance of the implemented system. As a result, we confirm our unconstraint system can be used for daily life ECG monitoring.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.2
• /
• pp.311-317
• /
• 2014

Buoy has different motion characteristics depends on the sea weather situations. The motion characteristics has an impact on antenna, solar power generation system and etc. installed within a buoy. Therefore, it is important to analyse motion characteristics for management and analyse the buoy conditions. This paper's Buoy motion monitoring system uses gyro sensor to detect motions of a light buoy, and the measured data transfers to the PC on the shore using signal processing algorithm. The aim of this research is to develop monitoring and management mechanism of a buoy by applying motion monitoring system. In this paper, the operation characteristic of movement monitoring system is verified through experiment. Further, in this paper, it can apply such as real-time visibility into the status of the buoy or many ocean facility's motion estimation of the future.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.12 no.2
• /
• pp.290-298
• /
• 2008

This paper presents an efficient data storage and reconstruction method in data acquisition and processing of monitoring system. The proposed method extracts minimum data using an interpolation from raw data which are acquired from a target system. They are transferred and saved in a monitoring PC via TCP/IP communication, and then reconstructed as original signals. Therefore, it is possible to design an efficient monitoring system by the improved data communication speed due to the reduced communication packet, and it reduces the storage space. The algorithm for data acquisition and reconstruction is based on Cubic Hermite interpolation. To verify the validity of the proposed scheme, we presents simulation results compared with other interpolation based approaches. Finally, it is applied to a monitoring system for grid-connected photovoltaic power generation system to prove the high-performance of the proposed method.

• Proceedings of the KIEE Conference
• /
• 2005.05a
• /
• pp.93-95
• /
• 2005

When a control system is designed computer simulations are important because they provide a way to test the performances of the control law and determine some design parameters of the control system. Since most real situations are different from those of the simulations, however, it is required to manage uncertainties of the plant and analyze errors fast and correctly. For this we consider a convenient monitoring system which can display real systems responses exactly and check the performances of the control systems. In this paper, an approach to design a monitoring system is presented by using C++Builder program which is based on PC. Firstly we introduce the structure and functions of the monitoring system. Then, an inverted-pendulum control system is investigated by using the monitoring system.

• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.64 no.2
• /
• pp.50-56
• /
• 2015

This paper presents a smart monitoring technique for photovoltaic power systems by using wire and wireless communication networks in which the RS-232/484 and the Zigbee communication networks are inherently established respectively. In the proposed monitoring systems, environmental data sequences and the output power measured by sensors in photovoltaic systems are transferred to PC systems via two communication networks. We made electronic hardware boards for sensors and communication networks to construct its real-time monitoring system and carry out experiments for demonstrating reliability of the proposed monitoring system.

• Proceedings of the KIEE Conference
• /
• 2004.11c
• /
• pp.200-202
• /
• 2004

This paper deals with the real-time monitoring and control system using PC, PDA(Win CE embedded device) and PCS(based BREW platform). The camera attached to the server captures the moving target, and the captured frame of color image is encoded in JPEG for image compression at the server. The client(PC, PDA, PCS) receives the image data from the remote server and the received image is decoded from decompression. We use the TCP/IP protocol to send the image frames. The client can control the position of the camera by sending the control command to the server. Two DC servo motors for the camera are controlled in any directions, up-down and left-right, by the controller which is communicating with the server via the serial communication to get the control command. In this way, on the client we can monitor the moving images at the server and also control the position of the camera.