• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.26 no.11C
• /
• pp.75-84
• /
• 2001

In this paper, we propose an on-line FDD(Fault Detection and Diagnosis) system based on the three layer feed-forward neural network which is trained by the back-propagation teaming algorithm. We implement the on-line fault detection and diagnosis system by Visual C++ and Visual Basic. The proposed FDD system is applied to an air handling unit in operation. Experimental results show the high performance of our system in the task of fault detection and diagnosis.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2002.11a
• /
• pp.506-509
• /
• 2002

In this paper. we give pressure stimulation into organic ultra thin films and detected the induced displacement current proper ties, and then manufacture a device under the accumulation condition. In processing of a device manufacture. And electroluminescence(EL) from conjugated polymers has recently received great attention because polymer light-emitting diodes(LEDs) clealy have potential for applications such as large-area displays. The operation of polymer LEDs is based on double injection of electrons and holes from the elextrodes. followed by formation of excitons whose radiative decay results in light emission at wavelength characteristic to the material. In this paper, we fabricated the single layer EL device using $Alq_3$ as emitting material. According as turn on voltage could know about 5.5V in voltage-current characteristics and voltage rise, current could see that increase as non-linear, Current and ruminance can see that express similar relativity in voltage, and could know that ruminance is expressing current relativity.

• Journal of the Korean Society for Aviation and Aeronautics
• /
• v.14 no.2
• /
• pp.27-32
• /
• 2006

The performance-design algorithm of a liquid ramjet engine was studied, which showed the comparable results with CFD calculation except the shock-boundary layer interaction region. In addition to the description of the design algorithm, several important design parameters, such as equivalence ratio, combustion efficiency, air capturing factor, and flight Mach number, are evaluated as predicting one of performance factors, Isp, of a typical ramjet engine, so that the flight envelope might not be determined with narrow perspective performance-operation-area in off-design regime.

• Elastomers and Composites
• /
• v.51 no.4
• /
• pp.280-285
• /
• 2016

The purpose of this study is to get optimum design and operation conditions of the heating rubber roller for laminating process. The cause of performance degradation of heating rubber roller is delamination of rubber on metal tube, rubber aging due to high temperature. We measured the material properties of thermal expansion, thermal conductivity, specific heat and density and analyzed thermal distributions of rubber layer using finite element method. As a result of heat/flow analysis, the density distribution of heating coil must shorten the stabilization time by reducing the temperature deviation on the length direction at the temperature rising section after increasing the density of the area contacting with the laminate film at the center part which is an opposite of the current composition while enabling to maintain the temperature of heater to be consistent while maintaining the temperature deviation to be low when heat loss is created. Finally, we determined optimum heating method of heating rubber roller.

• Environmental Engineering Research
• /
• v.11 no.4
• /
• pp.201-207
• /
• 2006

A simple empirical model with good quantitative prediction of inter-particle and intra-particle distance in a cake layer with respect to ionic strength was developed. The model is an inverse length scale with functions of interaction energy and hydrodynamic factor and it explains that the inter-particle and intra-particle distance in a cake is directly related to the effective size of particles. Particle compressibility with respect to ionic strength was also predicted by the model. The model corroborated very well with experimental results of polystyrene microsphere latex particles microfiltation in a dead end operation. From the results of the model, specific cake resistance could be controlled by the same variables affecting the height of particle energy barrier described by the Derjaguin-Landau-Verwey-Overbeek (DLVO) theory.

• ETRI Journal
• /
• v.41 no.6
• /
• pp.863-872
• /
• 2019

Elliptic curve cryptography is a relatively lightweight public-key cryptography method for key generation and digital signature verification. Some lightweight curves (eg, Curve25519 and Curve Ed448) have been adopted by upcoming Transport Layer Security 1.3 (TLS 1.3) to replace the standardized NIST curves. However, the efficient implementation of Curve Ed448 on Internet of Things (IoT) devices remains underexplored. This study is focused on the optimization of the Curve Ed448 implementation on low-end IoT processors (ie, 8-bit AVR and 16-bit MSP processors). In particular, the three-level and two-level subtractive Karatsuba algorithms are adopted for multi-precision multiplication on AVR and MSP processors, respectively, and two-level Karatsuba routines are employed for multi-precision squaring. For modular reduction and finite field inversion, fast reduction and Fermat-based inversion operations are used to mitigate side-channel vulnerabilities. The scalar multiplication operation using the Montgomery ladder algorithm requires only 103 and 73 M clock cycles on AVR and MSP processors.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.13 no.3
• /
• pp.109-116
• /
• 2018

SSDs that are not allowed in-place update within the allocated page cause another allocation of a new page that will replace the previous page at the moment data modification occurs. This intrinsic characteristic of SSDs requires many changes to the existing HDD-based IO theory. In this paper, we conduct a performance comparison of FTL caching strategy in perspective of cache hashing (Global vs. grouped) and caching algorithm (LRU vs. NUR) through a simulation. Experimental results show that in terms of energy consumption for flash operation the grouped management of cache is not suitable and NUR algorithm is superior to LRU algorithm. In particular, we found that the cache hit ratio of LRU algorithm is about 10% point higher than that of NUR algorithm while the energy consumption of LRU algorithm is about 32% high.

• Journal of the Korean Ceramic Society
• /
• v.43 no.10 s.293
• /
• pp.613-618
• /
• 2006

Graphite is suitable materials as a moderator, reflector, and supporter of a nuclear reactor because of high tolerance to the high temperature and neutron irradiations. Because graphite is so weak to the oxidation, its oxidation study is essentially demanded for the operation and design of the nuclear reactor. This work focuses on the effect of the surface oxidation of graphite according to the surface treatment. With thermogravimeter (TG), oxidation characteristics of the isotropic graphite are measured at the three temperature areas, and oxidation ratio and amounts are estimated as changing the surface roughness. Furthermore, the polished graphite surface produced fom the surface treatment is investigated with the Raman spectroscopic study. Oxidation behaviors of the surface are also evaluated as elimination the polished layer by washing with strong sonication.

• 한국HCI학회:학술대회논문집
• /
• 2007.02c
• /
• pp.322-325
• /
• 2007

Graphics processing unit (GPU) is used for a faster artificial neural network. It is used to implement the matrix multiplication of a neural network to enhance the time performance of a text detection system. Preliminary results produced a 20-fold performance enhancement using an ATI RADEON 9700 PRO board. The parallelism of a GPU is fully utilized by accumulating a lot of input feature vectors and weight vectors, then converting the many inner-product operations into one matrix operation. Further research areas include benchmarking the performance with various hardware and GPU-aware learning algorithms. (c) 2004 Pattern Recognition Society. Published by Elsevier Ltd. All rights reserved.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.29 no.12
• /
• pp.759-763
• /
• 2016

We demonstrate the utilization of ion gel gate dielectrics for operating non-volatile transistor memory devices based on polymer semiconductor thin films. The gating process in typical electrolyte-gated polymer transistors occurs upon the penetration and escape of ionic components into the active channel layer, which dopes and dedopes the polymer film, respectively. Therefore, by controlling doping and dedoping processes, electrical current signals through the polymer film can be memorized and erased over a period of time, which constitutes the transistor-type memory devices. It was found that increasing the thickness of polymer films can enhance the memory performance of device including (i) the current signal ratio between its memorized state and erased state and (ii) the retention time of the signal.