• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.5
• /
• pp.668-674
• /
• 2022

Machine learning techniques using visual data have high usability in fields of industry and service such as scene recognition, fault detection, security and user analysis. Among these, user analysis through the videos from CCTV is one of the practical way of using vision data. Also, many studies about lightweight artificial neural network have been published to increase high usability for mobile and embedded environment so far. In this study, we propose the network combining the object detection and classification for mobile graphic processing unit. This network detects pedestrian and face, classifies age and gender from detected face. Proposed network is constructed based on MobileNet, YOLOv2 and skip connection. Both detection and classification models are trained individually and combined as 2-stage structure. Also, attention mechanism is used to improve detection and classification ability. Nvidia Jetson Nano is used to run and evaluate the proposed system.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.14 no.3
• /
• pp.648-654
• /
• 2010

Request operations and release operations occur repeatedly in resource allocation systems. The process requesting a resource acquires one by any priority-based mechanism, and returns the resource after some periods. In this system, resource failures lead to delay of resource allocation, or to termination of process holding the failed resource. To analyze this process effectively, this paper designs a probabilistic ACSR, a process algebra that extends ACSR with the probabilistic choice operation. The ability to express/analyze both request-release rates and failure-recovery rates is illustrated using probabilistic ACSR.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.11 no.4
• /
• pp.25-31
• /
• 2007

For areas such as the Korean Peninsula, which have moderate seismic activity but no available records of strong ground motion, synthetic seismograms can be used to evaluate ground motion without waiting for a strong earthquake. Such seismograms represent the estimated ground motions expected from a set of possible earthquake scenarios. Local site effects are especially important in assessing the seismic hazard and possible ground motion scenarios for a specific fault. The earthquake source and rupture dynamics can be described as a two-step process of rupture initiation and front propagation controlled by a frictional sliding mechanism. The seismic wavefield propagates through heterogeneous geological media and finally undergoes near-surface modulations such as amplification or deamplification. This is a complex system in which various scales of physical phenomena are integrated. A unified approach incorporates multi-scale problems of dynamic rupture, radiated wave propagation, and site effects into an all-in-one model using a three-dimensional, fourth-order, staggered-grid, finite-difference method. The method explains strong ground motions as products of complex systems that can be modified according to a variety of fine-scale rupture scenarios and friction models. A series of such deterministic earthquake scenarios can shed light on the kind of damage that would result and where it would be located.

• Journal of KSNVE
• /
• v.11 no.1
• /
• pp.125-131
• /
• 2001

There is a clearance between the parts of a machine due to design tolerance, manufacturing error, wear, looseness, or misalignment. If the clearance is large, the vibration and noise of the machine is generally large. Therefore, the analysis on the nitration and noise of a machine can tell the clearance of the machine, which reveals the condition of the machine, i.e., the existence of faults and the safety of the machine. The investigation of this kind of research should be on the basis of experimental results. A link mechanism with a clearance at a joint between the coupler and locker is made for the investigation of the condition monitoring of a machine due to clearance. The vibration and sound are measured from the link driving system during the operation. The signals are clarified using line enhancement technique. The noise removed signals are used to develop the dynamic model of the system for a model based fault diagnosis. Also this study showed that the clarified signals can be used for the calculation of the joint forces between the coupler and rocker and for the correlation between the vibration and sound levels and the clearance sizes.

• Journal of the Korean Society for Precision Engineering
• /
• v.16 no.1 s.94
• /
• pp.211-218
• /
• 1999

The traditional robot manipulators are actuated by continuous range of motion actuators such as motors or hydraulic cylinders. However, there are many applications of mechanisms and robotic manipulators where only a finite number of locations need to be reached, and the robot’s trajectory is not important as long as it is bounded. Binary manipulator uses actuators which have only two stable states. As a result, binary manipulators have a finite number of states. The number of states of a binary manipulator grows exponentially with the number of actuators. This kind of robot manipulator has some advantage compared to a traditional one. Feedback control is not required, task repeatability can be very high, and finite state actuators are generally inexpensive. And this kind of robot manipulator has a fault tolerant mechanism because of kinematic redundancy. In this paper, we solve the inverse kinematic problem of a binary parallel robot manipulator using neural network and test the validity of this structure using some arbitrary points m the workspace of the robot manipulator. As a result, we can show that the neural network can find the nearest feasible points and corresponding binary states of the joints of the robot manipulator

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.58 no.8
• /
• pp.1512-1518
• /
• 2009

Recently, new dispersed power sources such as photovoltaics, wind power, fuel cell etc. are energetically interconnected and operated in the distribution systems, as one of the national projects for alternative energy. This paper deals with the optimal countermeasures for the mal-function of protective devices at primary feeder in distribution systems when new power sources like photovoltaic (PV) systems are interconnected, based on the symmetrical components of short circuit studies. When new power sources are considered to be interconnected to distribution systems, bi-directional power flow and interconnection transformer connection of new power sources may cause the operation problems of protective devices (mainly re-closer), since new power sources can change typical characteristics of distribution systems. Therefore, this paper shows an analysis skill of the mal-functional mechanism of protective relay and proposes the optimal solution for the mal-function problem using the symmetrical components of fault analysis. And, this paper also shows the effectiveness of proposed method by the simulation at the field distribution systems.

• Proceedings of the KIEE Conference
• /
• 2000.07a
• /
• pp.470-473
• /
• 2000

Because of the continuous growth of energy consumption, and also the tendency to site power lines and pipelines along the same routes, the close proximity of high voltage structures and metallic pipelines has become more and more frequent. Moreover, normal steady state and fault currents become higher as electric networks increase in size and power. Therefore, there has been and still is a growing concern (safety of people marking contact with pipeline, risk of damage to the pipeline coating, the metal and equipment connected to pipeline. especially cathodic protection system) about possible hazards resulting from the influence of high voltage power system on metallic structures(gas pipeline, oil pipeline and water pipeline etc.). Therefore, we analyze the interference problems when the gas pipeline is buried with power cable in the same submarine tunnel. This paper present the results of the study about interference mechanism, AC corrosion, limitation of safety voltage and analysis of indiction voltage.

• Proceedings of the KIEE Conference
• /
• 1989.07a
• /
• pp.241-245
• /
• 1989

A knowledge based expert system is a computer program that emulates the reasoning process of a human expert in a specific problem domain. This paper presents an expert system to diagnose the various faults in power system. The developed expert system is represented considering two points; the possibility of solution and the fast processing speed. As uncertainties exist in the facts and rules which comprise the knowledge base of the expert system, Certainty Factor, which is based on the confirmation theory is used for the inexact reasoning. Also, as the diagnosis problem requires the inductive reasoning process in nature, the solution is imperfect and not unique in general. So the expert system is designed to generate all the possible hypothesis in order of the possibility and also it can explain the propagation procedure of the faults for each solution using the built in backtracking mechanism. In realization of the expert system, the processing speed is greatly dependent upon the problem representation, reasoning scheme and search strategy. So, in this paper the fault diagnosis problem itself is analysed from the view point of Artificial Intelligence and as a result, the expert system has the following basic features. 1) The certainty factor is adopted in the inference engine for inexact reasoning. 2) Problem apace is represented using the problem reduction technique. 3) Bidirectional reasoning scheme is used. 4) Best first search strategy is adopted for rapid processing. The expert system was developed us ing PROLOG language.

• Earthquakes and Structures
• /
• v.18 no.3
• /
• pp.297-312
• /
• 2020

The KiK-net by NIED is a vertical array measurement system. In the database of KiK-net, singular pulse waves were observed in the seismic record at the borehole of TTRH02 during the mainshock (the magnitude of Japan Meteorological Agency (M_J) 7.3, M_W 6.8) and aftershock (M_j 4.2) of Tottori-ken Seibu earthquake in 2000. Singular pulse waves were also detected in the seismic records at the borehole of IWTH25 during the Iwate-Miyagi Nairiku earthquake in 2008 (M_J 7.2, M_W 6.9). These pulse waves are investigated by using the wave shape analysis methods, e.g., the non-stationary Fourier spectra and the double integrated displacement profiles. Two types of vibration modes are discriminated as the occurrence mechanism of the singular pulse waves. One corresponds to the reversal points in the displacement profile with the amplitude from 10^-4 m to 10^-1 m, which is mainly related to the fault activity and the amplification pass including the mechanical amplification (collision) of the seismograph in the casing pipe. The other is the cyclic pulse waves in the interval of reversal points, which is estimated as the backlash of the seismograph itself with the amplitude from 10^-5 m to 10^-4 m.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.16 no.5
• /
• pp.63-68
• /
• 2016

Audio mixer is an electronic device which performs a mixing of multiple audio signals. Digital mixer having various functions and scalability is spreaded thanks to advanced DSP and IT technology. However, digital mixer is more vulnerable to stability comparing to conventional analog mixer in the digital error or software error sense because its control is executed by SW. To solve this problem, in this paper, we propose a multi-channel digital analog hybrid mixer scheme, digital mixer error detection mechanism and malfunctioning switching technique. Also we develop the audio mixer having digital-analog hybrid structure. By simulation, we can sense the error of digital mixer except power loss in a 120ms, change into analog mixer mode automatically and provide continuous broadcasting function without mixer function loss.