구조물에 장기적으로 발생하는 노후화를 정량적으로 파악하기 위해 상시진동 데이터를 활용한 일반화된 모니터링 시스템에 관한 연구가 세계적으로 활발히 수행중이다. 본 연구에서는 구조물에서 장기적으로 취득되는 동특성을 앙상블 학습에 활용하여 구조물의 이상을 감지하기 위한 보급형 엣지 컴퓨팅 시스템을 구축하였다. 시스템의 하드웨어는 라즈베리파이와 보급형 가속도계, 기울기센서, GPS RTK 모듈, 로라 모듈로 구성됐다. 실험실 규모의 구조물 모형 진동실험을 통해 동특성을 활용한 앙상블 학습의 구조물 이상 감지를 검증하였으며, 실험을 기반으로 한 실시간 동특성 추출 분산처리 알고리즘을 라즈베리파이에 탑재하였다. 구축된 시스템을 하우징하고 포항시 행정복지센터에 설치하여 데이터를 취득함으로써 개발된 시스템의 현장 적용성을 검증하였다.

Oxide-dispersion-strengthened (ODS) alloy has been developed to increase the mechanical strength of metallic materials; such an improvement can be realized by distributing fine oxide particles within the material matrix. In this study, the ODS layer was formed in the surface region of Zr-based alloy tubes by laser beam treatment. Two kinds of Zr-based alloys with different alloying elements and microstructures were used: KNF-M (recrystallized) and HANA-6 (partial recrystallized). To form the ODS layer, Y2O3-coated tubes were scanned by a laser beam, which induced penetration of Y2O3 particles into the substrates. The thickness of the ODS layer varied from 20 to 55 μm depending on the laser beam conditions. A heat affected zone developed below the ODS layer; its thickness was larger in the KNF-M alloy than in the HANA-6 alloy. The ring tensile strengths of the KNF-M and HANA-6 alloy samples increased more than two times and 20–50%, respectively. This procedure was effective to increase the strength while maintaining the ductility in the case of the HANA-6 alloy samples; however, an abrupt brittle facture was observed in the KNF-M alloy samples. It is considered that the initial microstructure of the materials affects the formation of ODS and the mechanical behavior.

Zirconia nanoparticles were widely used as filler in order to get high refractive index layer. However, dispersion of nanoparticles is difficult due to their agglomeration in solvent. In this study, the dispersibility of the zirconia suspension is promoted by controlling the steric hindrance and electrostatic interactions through the adsorption of PEI according to alkali treatment time. Also, to induce improved dispersibility on suspension, we changed the dispersion conditions variously and fabricated an ink formulation method for the coating layer. Zirconia suspension was characterized by dynamic light scattering (DLS), Zeta potential measurement, Transmission Electron Microscope (TEM) and FT-IR. We were able to confirm that good dispersion of zirconia suspension by alkali treatment and PEI led to high refractive index.

Magnesium alloy is becoming known for the lightest material in the metallic materials. Recently the automotive industry has a variety application to the light weight parts replacement. This study focuses on the mechanical property improving through a tiny amount’s CNT addition into the magnesium alloy as AM60. The CNT material is an arduous combination of the metallic materials. Therefore this study is concentrating on the contact force growth for the CNT material. Consequently, the made CNT is produced by the CVD process using the magnesium catalyst. The CNT material has dispersive with mechanical process into the molten AM60 alloy. The mechanical experiment result that hardness is 18% increasing and tensile strength is 13% increasing, better than the raw AM60 alloy on this investigation.

In the present work, physicochemical treatments were introduced for de-aggregation and stable dispersion of detonation nanodiamonds (DND) in polar solvents. The DNDs in water exhibited a particle size of 138 nm and high dispersion stability without particular treatment. However, the DNDs in ethanol were severely aggregated to several micrometers in size and showed poor dispersion stability with time. To break down aggregates of DNDs and enhance the dispersion stability of them in ethanol, mechanical force and chemical surfactant were introduced as functions of zirconia ball size, kind of surfactant and amount of surfactant added. From the analyses of average particle size and Turbiscan results, it was suggested that the size of DNDs in ethanol can be reduced by only mechanical force; however, the DNDs were re-aggregated due to high surface activity. The long-term dispersion stability can be achieved by applying mechanical force to break down the aggregates of DNDs and by preventing re-aggregation of them using proper surfactant.

Lamb waves are extensively used in plate-like structure inspection because of their guided nature. However, their dispersive properties often limit their use in flaw detection and other applications. Dispersion weakens and defocuses interrogating Lamb waves and makes it difficult to accurately interpret signals reflected from defects or boundaries. Time reversal (TR) processing can be applied to compensate for the effect of dispersive Lamb waves. Thus, the TR operation will enable the amplification of dispersive Lamb wave signals by signal compression in time. In this study, experiments are performed in order to examine the refocusing and recovering the initial input waveform in the long range propagation of dispersive Lamb waves in a plate. Two different time reversal processes (regular TR and reciprocal TR or inverse filtering) are tested and the experimental results are compared

This paper deals with the effects of the surfactant and preplate process (sensitization and activation) on electroless copper plating on carbon nano-fiber (CNF). Ultrasonic irradiation was applied both during dispersion of CNF and during electroless plating containing preplate process. The dispersion of CNF and flatness of the plated copper film were discussed based on the changes in surfactant concentration and preplate process time. It was clearly shown that high concentration of surfactant and long time of preplate process could promote the agglomeration of CNF and uneven copper plating on CNF.

대형 구조물에 대한 최적설계를 고려할 때 구조해석에 많은 시간과 노력이 소비된다. 한대의 개인용 컴퓨터에 의한 대형 구조물의 구조해석은 대용량의 기억장치와 많은 계산시간이 요구되므로 반복적 해석이 필요한 대형 구조물의 설계에 효율적으로 이용되기 어렵다. 따라서 본 논문에서는 이러한 문제의 대안으로 인터넷이 연결된 다수의 개인용 컴퓨터들로 고성능 병렬연산시스템을 구성하여 구조해석을 분산 처리하여 계산시간을 절감하였다. 아울러 반응표면의 근사를 위해 요구되는 구조해석을 상용 구조해석 어플리케이션으로 해결할 수 있다면 상용성이 확보되어 일반 구조물에 대하여도 반응표면법을 이용한 최적설계를 수행할 수 있을 것이다.

다분야통합해석에 기반한 설계문제는 일반적으로 전체 설계과정에서 매우 큰 계산시간을 요구하며, 이러한 계산시간을 단축하기 위해 병렬처리시스템을 도입하는 것이 필수적이다. 그러나 다분야통합해석에 기존의 병렬처리기법을 적용하기 위해서는 해석에 필요한 모든 CAE 소프트웨어들이 병렬처리시스템의 모든 서버에 설치되어 있어야 하며, 이는 매우 큰 CAE 소프트웨어의 비용을 필요로 한다. 본 논문에서는 이러한 문제점을 해결하기 위해 가중치 기반 멀티큐 부하분산 알고리즘을 제안하였다. 제안된 알고리즘은 서버들의 성능과 설치된 CAE 소프트웨어들의 종류가 각기 다른 이종 병렬처리시스템을 고려하였으며 성능검증을 위해 선입선출(First Come First Servre) 알고리즘을 적용한 경우와 비교한 전산실험을 수행하였다.

Ultra-fine copper powders with particle size about 150 nm were synthesized from copper hydroxide slurry by wet method using hydrazine as reduction agent and several sur factants at below . The particle size distribution and dispersion of synthesized powders as function of temperature, feeding rate of reduction and especially, sur factants were character ized by XRD, BET, PSA and SEM by this process.

해충의 발생예찰자료의 분석에 있어서 지리자료처리체계의 활용 가능성을 검토하였다. 전국의 152개 병해충 발생예찰소의 해충 발생자료를 이용하여 해충지리적 분포정보를 집적하는 "데이터베이스" 체계를 SPANS(Spatial Analysis Systme)로 구축하였다. 구축된 "지리분포 데이터베이스"로부터 최근 10년간 이화명나방 발생량의 분포변동을 추적한 결과 일반계 품종이 확대 재배되기 시작한 80년대 중반터 고밀도 분포지가 급속히 확산됨을 볼 수 있었다. 또한 분포의 확산과 수축의 중심은 전북의 이리지역임이 확실하게 나타났다. 별멸구의 초기(7월초) 비래량의 분포를 비래후 온도조건과 결부시켜 후기(9월중)의 벼멸구 발생위험지역을 예측하는 시뮬레이션 모형을 구성하였다. 모형에 의해 추정된 벼멸구 발생위험지역을 예측하는 시뮬레이션 모형을 구성하였다. 모형에 의해 추정된 벼멸구 발생지의 분포도를 1990년 9월의 실재 발생분포도와 비교한 결과 경북 청도지역을 제외하고는 분포양상이 매우 유사하였다. 경사도, 고도, 재배작물 등의 분포도를 추정된 벼멸구 발생도와 함께 투시.분류(Overlay/Modeling)하는 기법을 사용한 결과 서남부지역의 벼멸구 국지적 발생위험지를 구분해 낼 수 있었다.

In this paper, decentralized data process based system identification(SI) method for building structure is proposed. For verification of the method, an experimental study was performed and the validity of the method was confirmed using unused displacement response in SI.

The dechlorination processes and efficiencies of PCBs contained in transformer dielectric oils were investigated on three commercial-treatment companies. Alkaline dispersion method was used for dechlorination of PCBs mainly using molten metallic sodium or potasium hydroxide with PEG, and its dechlorination efficiency to whole congeners of PCBs was 98.94% on average. Homologues of highly-chlorinated biphenyls showed better efficiencies than those of less chlorinated ones, especially mono-chlorinated biphenyl in para-position (#2) was about 37 times enriched after chemical treatment. Co-planar PCBs' dechlorination efficiency was averaged as high as 99.78%, but 3,3',4,4'-TeCB was relatively low as 96.5%. It was suggested for better dechlorination efficiency to remove the water in transformer oil prior to chemical reaction and to stir the reagents more rapidly when using alkaline dispersion method.