• 한국정보통신학회논문지
• /
• 제26권3호
• /
• pp.374-380
• /
• 2022

본 논문은 기계학습 기반의 LSTM(Long Short Term Memory) 알고리즘을 이용해 바다 양식장의 해수 상태 변화를 예측하는 연구 결과를 보여주고 있다. 바다 양식장의 해수 상태 정보를 수집하기 위해 용존산소량, 염도, 질소이온 농도 및 수온 측정 센서들을 사용해 하드웨어를 구현했으며, LoRa 통신을 이용해 클라우드 기반의 Firebase 데이터베이스로 전달해 저장하도록 구현하였다. 개발한 하드웨어를 이용해 통영과 거제 지역 양식장들 주변 해수 상태 정보들을 수집하였으며, 이들 실제 데이터셋을 사용한 학습 결과에다 LSTM 알고리즘을 적용하여 87%의 정확도를 보여주는 예측 결과를 얻어낼 수 있었다. 용존산소량을 비롯한 4가지 파라미터별 예측 결과를 사용자에게 제공하기 위해 Flask와 REST API를 사용했으며, 이러한 예측 결과는 어민들에게 바다 양식장의 해수 상태 변화를 미리 제공할 수 있어 어류 집단 폐사로 인한 큰 피해를 줄이는 데 도움이 될 것이라 기대한다.

• 해양환경안전학회지
• /
• 제28권2호
• /
• pp.394-405
• /
• 2022

본 논문은 CFD 해석법을 이용한 단일선체 및 이중선체 선박의 손상에 따른 기름 유출량 추정 결과를 소개한다. 수치해석 기법으로 다상유동 해석법을 사용하고 기름 유출 유동에 대한 다양한 수치해석적 조건의 변화에 대한 영향을 조사하였다. 좌초 또는 충돌에 의한 다양한 손상 조건에서의 기름 유출량의 변화를 해석하였으며 실험 결과를 통해 검증하였다. 본 논문의 수치해석 결과는 단일선체와 이중선체의 기하학적 특성을 따라 실험 결과와 잘 일치하였다. 특별히 물과 기름의 복잡한 상호작용이 나타나는 이중선체의 기름 유출 상황도 실험과 동일하게 타당하게 예측하였다. 본 연구를 통해 사고 선박의 기름 유출량 추정에 필요한 CFD기법을 정립할 수 있었다.

• Structural Engineering and Mechanics
• /
• 제83권3호
• /
• pp.293-304
• /
• 2022

In this work, a novel reliability approach for combined high and low cycle fatigue (CCF) estimation is developed by combining active learning strategy with least squares support vector machines (LS-SVM) (named as ALS-SVM) surrogate model to address the multi-resources uncertainties, including working loads, material properties and model itself. Initially, a new active learner function combining LS-SVM approach with Monte Carlo simulation (MCS) is presented to improve computational efficiency with fewer calls to the performance function. To consider the uncertainty of surrogate model at candidate sample points, the learning function employs k-fold cross validation method and introduces the predicted variance to sequentially select sampling. Following that, low cycle fatigue (LCF) loads and high cycle fatigue (HCF) loads are firstly estimated based on the training samples extracted from finite element (FE) simulations, and their simulated responses together with the sample points of model parameters in Coffin-Manson formula are selected as the MC samples to establish ALS-SVM model. In this analysis, the MC samples are substituted to predict the CCF reliability of turbine blades by using the built ALS-SVM model. Through the comparison of the two approaches, it is indicated that the reliability model by linear cumulative damage rule provides a non-conservative result compared with that by the proposed one. In addition, the results demonstrate that ALS-SVM is an effective analysis method holding high computational efficiency with small training samples to gain accurate fatigue reliability.

• Nuclear Engineering and Technology
• /
• 제54권10호
• /
• pp.3745-3765
• /
• 2022

The concrete structures related to nuclear safety are threatened by accidental impact loadings, mainly including the low-velocity drop-weight impact (e.g., spent fuel cask and assembly, etc. with the velocity less than 20 m/s) and high-speed projectile impact (e.g., steel pipe, valve, turbine bucket, etc. with the velocity higher than 20 m/s), while the existing studies are still limited in the impact resistant design of nuclear power plant (NPP), especially the primary RC slab. This paper aims to propose the numerical simulation and theoretical approaches to assist the impact-resistant design of RC slab in NPP. Firstly, the continuous surface cap (CSC) model parameters for concrete with the compressive strength of 20-70 MPa are fully calibrated and verified, and the refined numerical simulation approach is proposed. Secondly, the two-degree freedom (TDOF) model with considering the mutual effect of flexural and shear resistance of RC slab are developed. Furthermore, based on the low-velocity drop hammer tests and high-speed soft/hard projectile impact tests on RC slabs, the adopted numerical simulation and TDOF model approaches are fully validated by the flexural and punching shear damage, deflection, and impact force time-histories of RC slabs. Finally, as for the two low-velocity impact scenarios, the design procedure of RC slab based on TDOF model is validated and recommended. Meanwhile, as for the four actual high-speed impact scenarios, the impact-resistant design specification in Chinese code NB/T 20012-2019 is evaluated, the over conservation of which is found, and the proposed numerical approach is recommended. The present work could beneficially guide the impact-resistant design and safety assessment of NPPs against the accidental impact loadings.

• Smart Structures and Systems
• /
• 제30권6호
• /
• pp.687-701
• /
• 2022

Inevitable response loss under complex operational conditions significantly affects the integrity and quality of measured data, leading the structural health monitoring (SHM) ineffective. To remedy the impact of data loss, a common way is to transfer the recorded response of available measure point to where the data loss occurred by establishing the response mapping from measured data. However, the current research has yet addressed the structural condition changes afterward and response mapping learning from a small sample. So, this paper proposes a novel data driven structural response reconstruction method based on a sophisticated designed generating adversarial network (UAGAN). Advanced deep learning techniques including U-shaped dense blocks, self-attention and a customized loss function are specialized and embedded in UAGAN to improve the universal and representative features extraction and generalized responses mapping establishment. In numerical validation, UAGAN efficiently and accurately captures the distinguished features of structural response from only 40 training samples of the intact structure. Besides, the established response mapping is universal, which effectively reconstructs responses of the structure suffered up to 10% random stiffness reduction or structural damage. In the experimental validation, UAGAN is trained with ambient response and applied to reconstruct response measured under earthquake. The reconstruction losses of response in the time and frequency domains reached 16% and 17%, that is better than the previous research, demonstrating the leading performance of the sophisticated designed network. In addition, the identified modal parameters from reconstructed and the corresponding true responses are highly consistent indicates that the proposed UAGAN is very potential to be applied to practical civil engineering.

• Nutrition Research and Practice
• /
• 제16권4호
• /
• pp.464-475
• /
• 2022

BACKGROUND/OBJECTIVES: Increased levels of uremic toxins and decreased antioxidant capacity have a significant impact on the progression of chronic kidney disease (CKD). However, it remains unclear whether they interact with each other to mediate the damage of kidney function. The purpose of this study was to investigate whether uremic toxins (i.e., homocysteine and indoxyl sulfate [IS]), as well as glutathione-dependent antioxidant enzyme activities are dependently or independently associated with kidney function during different stages of CKD patients. SUBJECTS/METHODS: One hundred thirty-two patients diagnosed with CKD at stages 1 to 5 participated in this cross-sectional study. RESULTS: Patients who had reached an advanced CKD stage experienced an increase in plasma uremic toxin levels, along with decreased glutathione peroxidase (GSH-Px) activity. Plasma homocysteine, cysteine, and IS concentrations were all positively associated with each other, but negatively correlated to GSH-Px activity levels after adjusting for potential confounders in all CKD patients. Although plasma homocysteine, cysteine, IS, and GSH-Px levels were significantly associated with kidney function, only plasma IS levels still had a significant association with kidney function after these parameters were simultaneously adjusted. In addition, plasma IS could interact with GSH-Px activity to be associated with kidney function. CONCLUSIONS: IS plays a more dominant role than homocysteine and GSH-Px activity in relation to kidney function.

• Wind and Structures
• /
• 제34권2호
• /
• pp.185-197
• /
• 2022

This study aimed to analyze the wind-induced mechanical energy (WME) of a proposed super high-rise and long-span transmission tower-line system (SHLTTS), which, in 2021, is the tallest tower-line system with the longest span. Anew index - the WME, accounting for the wind-induced vibration behavior of the whole system rather than the local part, was first proposed. The occurrence of the maximum WME for a transmission tower, with or without conductors, under synoptic winds, was analyzed, and the corresponding formulae were derived based on stochastic vibration theory. Some calculation data, such as the drag coefficient, dynamic parameters, windshielding areas, mass, calculation point coordinates, mode shape and influence function, derived from wind tunnel testing on reducedscale models and finite element software were used in calculating the maximum WME of the transmission tower under three cases. Then, the influence of conductors, wind speed, gradient wind height and wind yaw angle on WME components and the energy transfer relationship between substructures (transmission tower and conductor) were analyzed. The study showed that the presence of conductors increases the WME of transmission towers and changes the proportion of the mean component (MC), background component (BC) and resonant component (RC) for WME; The RC of WME is more susceptible to the wind speed change. Affected by the gradient wind height, the WME components decrease. With the RC decreasing the fastest and the MC decreasing the slowest; The WME reaches the its maximum value at the wind yaw angle of 30°. Due to the influence of three factors, namely: the long span of the conductors, the gradient wind height and the complex geometrical profile, it is important that the tower-line coupling effect, the potential for fatigue damage and the most unfavorable wind yaw angle should be given particular attention in the wind-resistant design of SHLTTSs

• Korean Journal of Acupuncture
• /
• 제39권4호
• /
• pp.132-141
• /
• 2022

Objectives : Some acupoints are commonly utilized to treat a variety of diseases. The acupoints appear to have a wide range of effects caused by several mechanisms. The purpose of this study is to investigate into the potential role of microRNAs (miRNAs) in the multipotent effects of individual acupoint stimulation. Methods : We examined the miRNA expressions in the dorsal root ganglia (DRG) of neuropathic or inflammatory pain rats following ST36 and GB34 electroacupuncture (EA) stimulation. Neuropathic pain was induced by L5 spinal nerve ligation. Inflammatory pain was induced by knee joint injection of Complete Freund's adjuvant (CFA). EA was given under gaseous anesthesia with the same parameters (1mA, 2Hz, 30 min) in 5 consecutive days. Pain behaviors and miRNA expressions were analyzed. Results : In rats with neuropathic and inflammatory pain, EA treatments significantly enhanced the paw withdrawal threshold and weight-bearing force. After nerve injury, 36 miRNAs were upregulated in the DRG of neuropathic rats, while EA downregulated 10 of them. Furthermore, 14 miRNAs were downregulated following nerve damage, while one was increased by EA. 15 miRNAs were increased in the DRG of inflammatory rats following CFA injection, while 5 were downregulated by EA. Furthermore, 17 miRNAs were downregulated following CFA injection, while 7 were increased by EA. The miRNAs rno-miR-335, rno-miR-381-5p, rno-miR-1306-3p, and rno-miR-1839-3p were regulated by EA in both models. Conclusions : In two pain models, EA applied to ST36 and GB34 regulated miRNA expression differently. There appeared to be both acupoint-specific and non-specific miRNAs, and miRNA regulation of differential protein expression may modulate a variety of EA mechanisms.

• 한국축산식품학회지
• /
• 제43권5호
• /
• pp.840-858
• /
• 2023

This study was designed to compare the quality changes in mackerel fillets stored under different conditions by using hyperspectral imaging (HSI) techniques. Fillets packaged in vacuum were stored for six days under five different conditions: refrigerated at 4℃ (R group); iced at 5±3℃ (I group); kept at an ambient of 17±2℃ (A group); frozen at -18℃ for 24 h and thawed in a refrigerator at 4℃ for 5 h on the sampling day (FTR group); FTR thawed in tap water instead of thawing in a refrigerator (FTW group). The FTR group had the lowest total bacterial count, drip loss, 2-thiobarbituric acid reactive substances, volatile basic nitrogen, and texture profile analysis values among groups during the entire storage period (p<0.05). Scanning electron microscopy revealed that the FTR group had less damage, while the other groups had shrunken muscle tissues. HIS integrated with the partial least squares model yielded reliable and efficient results, with high R²_cv values, for several quality parameters of the mackerel fillets. Overall, the FTR group, involving freezing and thawing in a refrigerator, appears to be the most favorable option for maintaining the quality of mackerel fillets, which could be practically implemented in the industry. HSI is a suitable and effective technique for determining the quality of mackerel fillets stored under different conditions.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2022년도 춘계학술대회
• /
• pp.491-493
• /
• 2022

세계 각지에서 집중호우, 태풍 등으로 인한 홍수 피해가 많이 발생하고 있으며, 이러한 피해를 줄이기 위해 홍수를 미리 예측하는 것은 수해 피해 관리 차원에서 필수적인 요소이다. 본 논문에서는 홍수예측을 위한 핵심 파라미터인 수위, 강수량, 그리고 습도 데이터를 입력 데이터로 활용한 수위 예측 모델을 제안한다. 많은 연구 분야에서 이미 시계열 데이터 예측 성능이 검증된 LSTM 및 GRU 모델을 기반으로 기상청에서 제공하는 종관기상관측 자료와, 방재기상관측 자료를 활용하여 입력 데이터셋을 다르게 구축하고, 성능 비교 실험을 진행하였다. 결과적으로 종관기상관측 자료를 사용했을 때 가장 좋은 결과를 얻었다. 본 논문을 통해 입력 데이터에 따른 성능 비교 실험을 진행하였고, 향후 연구로 홍수 위험도 판별 모델과 연계하여 사전에 대피 결정이 가능한 시스템 개발의 초기 연구로서 활용될 수 있을 것으로 사료된다.