• Journal of the korean society of oceanography
• /
• 제32권4호
• /
• pp.163-170
• /
• 1997

Some CMP gathers acquired from shallow marine seismic reflection survey in offshore Korea do not show the hyperbolic trend of moveout. It originated from so-called swell effect of source and streamer, which are towed under rough sea surface during the data acquisition. The observed time deviations of NMO-corrected traces can be entirely ascribed to the swell effect. To correct these time deviations, a residual statics is introduced using Genetic Algorithms (GA) into the swell correction. A new class of global optimization methods known as GA has recently been developed in the field of Artificial Intelligence and has a resemblance with the genetic evolution of biological systems. The basic idea in using GA as an optimization method is to represent a population of possible solutions or models in a chromosome-type encoding and manipulate these encoded models through simulated reproduction, crossover and mutation. GA parameters used in this paper are as follows: population size Q=40, probability of multiple-point crossover P$_c$=0.6, linear relationship of mutation probability P$_m$ from 0.002 to 0.004, and gray code representation are adopted. The number of the model participating in tournament selection (nt) is 3, and the number of expected copies desired for the best population member in the scaling of fitness is 1.5. With above parameters, an optimization run was iterated for 101 generations. The combination of above parameters are found to be optimal for the convergence of the algorithm. The resulting reflection events in every NMO-corrected CMP gather show good alignment and enhanced quality stack section.

• Membrane and Water Treatment
• /
• 제9권2호
• /
• pp.115-121
• /
• 2018

Artificial neural network (ANN) simulation is used to predict the dynamic change of permeate flux during wheat starch industry wastewater microfiltration with and without static turbulence promoter. The experimental program spans range of a sedimentation times from 2 to 4 h, for feed flow rates 50 to 150 L/h, at transmembrane pressures covering the range of $1{\times}10^5$ to $3{\times}10^5Pa$. ANN predictions of the wastewater microfiltration are compared with experimental results obtained using two different set of microfiltration experiments, with and without static turbulence promoter. The effects of the training algorithm, neural network architectures on the ANN performance are discussed. For the most of the cases considered, the ANN proved to be an adequate interpolation tool, where an excellent prediction was obtained using automated Bayesian regularization as training algorithm. The optimal ANN architecture was determined as 4-10-1 with hyperbolic tangent sigmoid transfer function transfer function for hidden and output layers. The error distributions of data revealed that experimental results are in very good agreement with computed ones with only 2% data points had absolute relative error greater than 20% for the microfiltration without static turbulence promoter whereas for the microfiltration with static turbulence promoter it was 1%. The contribution of filtration time variable to flux values provided by ANNs was determined in an important level at the range of 52-66% due to increased membrane fouling by the time. In the case of microfiltration with static turbulence promoter, relative importance of transmembrane pressure and feed flow rate increased for about 30%.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2019년도 학술발표회
• /
• pp.363-363
• /
• 2019

활성화 함수(activation function)는 기계학습(machine learning)의 학습과정에 비선형성을 도입하여 심층적인 학습을 용이하게 하고 예측의 정확도를 높이는 중요한 요소 중 하나이다(Roy et al., 2019). 일반적으로 기계학습에서 사용되고 있는 활성화 함수의 종류에는 계단 함수(step function), 시그모이드 함수(sigmoid 함수), 쌍곡 탄젠트 함수(hyperbolic tangent function), ReLU 함수(Rectified Linear Unit function) 등이 있으며, 예측의 정확도 향상을 위하여 다양한 형태의 활성화 함수가 제시되고 있다. 본 연구에서는 기계학습을 통하여 수위예측 시 정확도 향상을 위하여 Hybrid 활성화 함수를 제안하였다. 연구대상지는 조수간만의 영향을 받는 한강을 대상으로 선정하였으며, 2009년 ~ 2018년까지 10년간의 수문자료를 활용하였다. 수위예측 알고리즘은 Python 내 Tensorflow의 RNN (Recurrent Neural Networks) 모델을 이용하였으며, 강수량, 수위, 조위, 댐 방류량, 하천 유량의 수문자료를 학습시켜 3시간 및 6시간 후의 수위를 예측하였다. 예측정확도 향상을 위하여 입력 데이터는 정규화(Normalization)를 시켰으며, 민감도 분석을 통하여 신경망모델의 은닉층 개수, 학습률의 최적 값을 도출하였다. Hybrid 활성화 함수는 쌍곡 탄젠트 함수와 ReLU 함수를 혼합한 형태로 각각의 가중치($w_1,w_2,w_1+w_2=1$)를 변경하여 정확도를 평가하였다. 그 결과 가중치의 비($w_1/w_2$)에 따라서 예측 결과의 RMSE(Roote Mean Square Error)가 최소가 되고 NSE (Nash-Sutcliffe model Efficiency coefficient)가 최대가 되는 지점과 Peak 수위의 예측정확도가 최대가 되는 지점을 확인할 수 있었다. 본 연구는 현재 Data modeling을 통한 수위예측의 정확도 향상을 위해 기초가 되는 연구이나, 향후 다양한 형태의 활성화 함수를 제안하여 정확도를 향상시킨다면 예측 결과를 통하여 침수예보에 대한 의사결정이 가능할 것으로 기대된다.

• Wind and Structures
• /
• 제19권4호
• /
• pp.443-465
• /
• 2014

Hyperbolic thin-shell cooling towers have complicated vibration modes, and are very sensitive to the effects of group towers and wind-induced vibrations. Traditional aero-elastic models of cooling towers are usually designed based on the method of stiffness simulation by continuous medium thin shell materials. However, the method has some shortages in actual engineering applications, so the so-called "equivalent beam-net design method" of aero-elastic models of cooling towers is proposed in the paper and an aero-elastic model with a proportion of 1: 200 based on the method above with integrated pressure measurements and vibration measurements has been designed and carried out in TJ-3 wind tunnel of Tongji university. According to the wind tunnel test, this paper discusses the impacts of self-excited force effect on the surface wind pressure of a large-scale cooling tower and the results show that the impact of self-excited force on the distribution characteristics of average surface wind pressure is very small, but the impact on the form of distribution and numerical value of fluctuating wind pressure is relatively large. Combing with the Complete Quadratic Combination method (hereafter referred to as CQC method), the paper further studies the numerical sizes and distribution characteristics of background components, resonant components, cross-term components and total fluctuating wind-induced vibration responses of some typical nodes which indicate that the resonance response is dominant in the fluctuating wind-induced vibration response and cross-term components are not negligible for wind-induced vibration responses of super-large cooling towers.

• Geomechanics and Engineering
• /
• 제13권3호
• /
• pp.459-474
• /
• 2017

To investigate the mechanical properties of Expanded Polystyrene (EPS) Beads Stabilized Lightweight Soil (EBSLS), Laboratory studies were conducted. Totally 20 sets of specimens according to the complete test design were prepared and tested with unconfined compressive test and consolidated drained triaxial test. Results showed that dry density of EBSLS ($0.67-1.62g/cm^3$) decreases dramatically with the increase of EPS beads volumetric content, while increase slightly with the increase of cement content. Unconfined compressive strength (10-2580 kPa) increases dramatically in parabolic relationship with the increase of cement content, while decreases with the increase of EPS beads volumetric content in hyperbolic relationship. Cohesion (31.1-257.5 kPa) increases with the increase of cement content because it is mainly caused by the bonding function of hydration products of cement. The more EPS beads volumetric content is, the less dramatically the increase is, which is a result of the cohesion between hydration products of cement and EPS beads is less than that between hydration products of cement and sand particles. Friction angle ($14.92-47.42^{\circ}$) decreases with the increase of EPS beads volumetric content, which is caused by the smoother surfaces of EPS beads than sand grains. The stress strain curves of EBSLS tend to be more softening with the increase of EPS beads content or the decrease of cement content. The shear contraction of EBSLS increases with the increase of $c_e$ or the decrease of $c_c$. The results provided quantitative relationships between physico-mechanical properties of EBSLS and material proportion, and design process for engineering application of EBSLS.

• 한국지반공학회논문집
• /
• 제34권12호
• /
• pp.43-58
• /
• 2018

본 연구에서는 매입말뚝 주면부의 하중전이 거동을 확인하기 위하여 14본의 강관매입말뚝에 대한 정재하시험을 수행하였다. 정재하시험을 통하여 하중-침하 곡선 및 축하중 분포를 계측하였으며, 그 결과 말뚝-지반 사이에서 발생되는 주면마찰력을 정량화 하였다. 이를 토대로 강관매입말뚝의 일반적인 하중전이곡선 형태를 도출하였다. 본 연구 결과, 말뚝의 주면하중전이 거동은 연성과 취성파괴 두 가지 형태로 나타났다. 그 결과, 강관매입말뚝의 t-z 곡선을 쌍곡선(일수현상이 없는 경우)과 톱니 형태(일수현상이 있는 경우 또는 표준관입저항치, $N{\geq}50$인 풍화암 이상)로 각각 제안하였다. 그때의 최대주면마찰력은 쌍곡선의 경우 지반의 N치 또는 암반의 일축압축강도, 톱니 형태의 경우 주입된 시멘트풀에 의해 형성된 쏘일-시멘트층의 일축압축강도의 영향을 받는 것으로 나타났다.

• Structural Engineering and Mechanics
• /
• 제71권2호
• /
• pp.185-196
• /
• 2019

In this investigation, study of the static and dynamic behaviors of functionally graded beams (FGB) is presented using a hyperbolic shear deformation theory (HySDT). The simply supported FG-beam is resting on the elastic foundation (Winkler-Pasternak types). The properties of the FG-beam vary according to exponential (E-FGB) and power-law (P-FGB) distributions. The governing equations are determined via Hamilton's principle and solved by using Navier's method. To show the accuracy of this model (HySDT), the current results are compared with those available in the literature. Also, various numerical results are discussed to show the influence of the variation of the volume fraction of the materials, the power index, the slenderness ratio and the effect of Winkler spring constant on the fundamental frequency, center deflection, normal and shear stress of FG-beam.

• 수산경영론집
• /
• 제50권2호
• /
• pp.41-56
• /
• 2019

The aim of this study is to analyze the productivity change of pollock enterprise by applying a mutually quadratic hyperbolic model and a bootstrapping model. This study used 20 units of pollock firms data (from 2013 to 2017). As a result of total productivity analysis of twenty pollock enterprises, total factor productivity was estimated to have decreased by 24.9% over the last five years (2013~2017). The main cause of this productivity decline was analyzed by technical change. In terms of annual productivity change, it showed decrease 3.0% in 2013~2014, 7.8% in 2014~2015, 4.5% in 2015~2016 and 4.7% in 2016~2017 respectively. In the analysis of productivity by corporation type, total factor productivity showed a significant decrease in both general corporation and external corporation, and productivity decrease (-29.3%) was larger than general corporation (-23.0%). In the productivity analysis by type of business, total factor productivity decreased significantly in the order of wholesale and commodity brokerage (-26.3%), food manufacturing (-25.1%) and fisheries (-15.3%). This decrease in productivity was caused by the technological change which indicates a downward shift in the production curve that is significant in all sectors.

• Steel and Composite Structures
• /
• 제32권2호
• /
• pp.157-171
• /
• 2019

In this research, the dynamic instability region (DIR) of the sandwich nano-beams are investigated based on nonlocal strain gradient elasticity theory (NSGET) and various higher order shear deformation beam theories (HSDBTs). The sandwich piezoelectric nano-beam is including a homogenous core and face-sheets reinforced with functionally graded (FG) carbon nanotubes (CNTs). In present study, three patterns of CNTs are employed in order to reinforce the top and bottom face-sheets of the beam. In addition, different higher-order shear deformation beam theories such as trigonometric shear deformation beam theory (TSDBT), exponential shear deformation beam theory (ESDBT), hyperbolic shear deformation beam theory (HSDBT), and Aydogdu shear deformation beam theory (ASDBT) are considered to extract the governing equations for different boundary conditions. The beam is subjected to thermal and electrical loads while is resting on Visco-Pasternak foundation. Hamilton principle is used to derive the governing equations of motion based on various shear deformation theories. In order to analysis of the dynamic instability behaviors, the linear governing equations of motion are solved using differential quadrature method (DQM). After verification with validated reference, comprehensive numerical results are presented to investigate the influence of important parameters such as various shear deformation theories, nonlocal parameter, strain gradient parameter, the volume fraction of the CNTs, various distributions of the CNTs, different boundary conditions, dimensionless geometric parameters, Visco-Pasternak foundation parameters, applied voltage and temperature change on the dynamic instability characteristics of sandwich piezoelectric nano-beam.

• Wind and Structures
• /
• 제29권4호
• /
• pp.235-246
• /
• 2019

Wind loading is one of important loadings that should be considered in the design of large hyperbolic natural-draught cooling towers. Both external and internal surfaces of cooling tower are under the action of wind loading for cooling circulating water. In the previous studies, the wind loads on the external surface attracted concernedly attention, while the study on the internal surface was relatively ware. In the present study, the wind pressure on the internal surface of a 220 m high cooling tower is measured through wind tunnel testing, and the effect of ventilation rate of the packing layer on internal pressure is a major concern. The characteristics of internal wind pressure distribution and its effect on wind-induced responses calculated by finite element method are investigated. The results indicate that the wind loading on internal surface of the cooling tower behaves remarkable three-dimensional effect, and the pressure coefficient varies along both of height and circumferential directions. The non-uniformity is particularly strong during the construction stage. Analysis results of the effect of internal pressure on wind-induced responses show that the size and distribution characteristics of internal pressure will have some influence on wind-induced response, however, the outer pressure plays a dominant role in the wind-induced response of cooling tower, and the contribution of internal pressure to the response is small.