• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.43-43
• /
• 2022

유해 조류 대발생은 전국 각지의 인공호소나 하천에서 다발적으로 발생하며, 경관을 해치고 수질을 오염시키는 등 수자원에 부정적인 영향을 미친다. 본 연구에서는 인공호소에서 발생하는 유해 조류 대발생을 예측하기 위해 심층학습 기법을 이용하여 예측 모델을 개발하고자 하였다. 대상 지점은 대청호의 추동 지점으로 선정하였다. 대청호는 금강유역 중류에 위치한 댐으로, 약 150만명에 달하는 급수 인구수를 유지 중이기에 유해 남조 대발생 관리가 매우 중요한 장소이다. 학습용 데이터 구축은 대청호의 2011년 1월부터 2019년 12월까지 측정된 수질, 기상, 수문 자료를 입력 자료를 이용하였다. 수질 예측 모델의 구조는 다중 레이어 퍼셉트론(Multiple Layer Perceptron; MLP)으로, 입력과 한 개 이상의 은닉층, 그리고 출력층으로 구성된 인공신경망이다. 본 연구에서는 인공신경망의 은닉층 개수(1~3개)와 각각의 레이어에 적용되는 은닉 노드 개수(11~30개), 활성함수 5종(Linear, sigmoid, hyperbolic tangent, Rectified Linear Unit, Exponential Linear Unit)을 각각 하이퍼파라미터로 정하고, 모델의 성능을 최대로 발휘할 수 있는 조건을 찾고자 하였다. 하이퍼파라미터 최적화 도구는 Tensorflow에서 배포하는 Keras Tuner를 사용하였다. 모델은 총 3000 학습 epoch 가 진행되는 동안 최적의 가중치를 계산하도록 설계하였고, 이 결과를 매 반복마다 저장장치에 기록하였다. 모델 성능의 타당성은 예측과 실측 데이터 간의 상관관계를 R2, NSE, RMSE를 통해 산출하여 검증하였다. 모델 최적화 결과, 적합한 하이퍼파라미터는 최적화 횟수 총 300회에서 256 번째 반복 결과인 은닉층 개수 3개, 은닉 노드 수 각각 25개, 22개, 14개가 가장 적합하였고, 이에 따른 활성함수는 ELU, ReLU, Hyperbolic tangent, Linear 순서대로 사용되었다. 최적화된 하이퍼파라미터를 이용하여 모델 학습 및 검증을 수행한 결과, R2는 학습 0.68, 검증 0.61이었고 NSE는 학습 0.85, 검증 0.81, RMSE는 학습 0.82, 검증 0.92로 나타났다.

• Journal of the Society of Naval Architects of Korea
• /
• v.37 no.4
• /
• pp.11-18
• /
• 2000

This paper is concerned with the generation of an optimal forward hull form by a nonlinear programming method. A Rankine source panel method based on the inviscid and potential flow approximation is employed to calculate the wave-making resistance and SQP method is also used for the optimization. The hull form is represented by a spline function. The forward hull form of a minimum wave resistance with the given design constraints is generated. In addition, the forward hull form of a minimum total resistance by considering the frictional resistance together with an empirical form factor is produced and compared with the former result.

• The Journal of the Acoustical Society of Korea
• /
• v.41 no.6
• /
• pp.637-646
• /
• 2022

Sub-arrays of arbitrary conformal array have different geometric shape through steering direction, thus the beam patterns of sub-arrays are always non-uniform. In this paper, we apply the beam pattern synthesis method using convex optimization into the conformal array, and shows the improvement of uniformity of beam performance. The simulation is performed with the conformal array of cut-sphere shape. As a result, the standard deviation of 3 dB beamwidth in elevation is greatly reduced but the directivity index is also reduced. To alleviate this trade-off, we propose a convex optimization using a shading function.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.196-196
• /
• 2021

상수관망은 많은 관을 통해 물의 수요가 있는 곳으로 물을 공급해주는 역할을 하는 사회기반 시설물이다. 상수관망 설계의 요점은 두 가지로 구분할 수 있다. 첫 번째 요점은 다양한 종류의 관배치로 인한 상수관망 설계안의 많은 경우의 수이다. 두 번째 요점은 상수관망 내 절점의 최저 요구수압 등의 제약조건이다. 두 가지 요점이 있는 상황에서 상수관망 설계비용의 최소화를 위한 상수관망 최적설계는 많은 계산이 요구된다. 많은 계산이 요구되기 때문에 상수관망 최적설계에 최적화 기법을 적용할 수 있다. 본 연구에서 상수관망 최적설계를 위해 적용된 최적화 기법은 Hybrid Rate(HR)를 개선한 Hybrid Vision Correction Algorithm(HVCA)이다. HVCA는 Vision Correction Algorithm(VCA)을 기반으로 추가적인 전역탐색을 실행하는 Centralized Global Search(CGS)의 적용 및 자가적응형 매개변수인 Hybrid Rate(HR)를 적용하여 사용성과 성능을 개량한 알고리즘이다. HVCA의 기존 HR은 선형적으로 증가하는 형태이다. 선형적으로 증가하는 HR로 인해 HVCA는 최적해 탐색과정에서 지역해에 빠지는 문제가 발생하였다. HVCA의 문제를 해결하기 위해 HR을 비선형적으로 증가하는 형태로 개량하였다. HR이 개량된 HVCA를 수자원공학 문제인 상수관망 최적설계 문제에 적용하여 결과를 비교하였다. 적용결과 HR이 개량된 HVCA가 기존의 HVCA보다 낮은 설계 비용을 나타내었다. 상수관망 최적설계 적용결과를 바탕으로 HR이 개량된 HVCA는 상수관망 최적설계 이외의 수자원공학 문제에도 적용가능할 것이다.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.33 no.2
• /
• pp.22-31
• /
• 2005

A design study was conducted for a new concept aircraft(Canard Rotor/Wing: CRW) that has the capability of dual mode flight, a rotorcraft and a fixed wing mode. The CRW can show a vertical take off/landing and a high speed/efficiency cruise performance simultaneously. It is not surprising to develop a new sizing code for this class of aircraft because conventional sizing codes developed solely for either the rotary wing or the fixed wing aircraft are not adequate to design a dual mode aircraft operated both by the rotary wing through tip jet effux and the fixed wing lift. Thus, a new design code was developed based on the conventional sizing code by adding some features including rotor performance, duct flow, and engine flow analysis, hence could eventually predict the performance of reaction driven rotor, the flight performance and the flight characteristics. The various design parameters were investigated to find their influences on the flight performance then, a small UAV(Unmanned Aircraft Vehicle) of 1500 lbs class was optimally designed to have minimum weight using the developed sizing code.

• Journal of KIISE:Computing Practices and Letters
• /
• v.12 no.1
• /
• pp.78-89
• /
• 2006

In development of network-enabled consumer electronics, much of the time and effort is spent analyzing and solving network performance problems. In this paper, we define an instance of such problems discovered while developing a commercial home network gateway. We then analyze its cause and propose a solution mechanism. Our home network gateway uses art asymmetric link (ADSL) and suffers from an undesirable phenomenon where downlink traffic interferes with upload speed. We call this phenomenon the network performance interference problem. While this problem can easily be confused with receive livelock caused by packet contention at the input queue, we and that this is not the case. By performing extensive experiments and analysis, we reveal that our problem is caused by packet contention at the output queue and certain intrinsic characteristics of TCP. We devise an ACK-separation queuing mechanism for this problem and implement it in the home network gateway Our experiments show that it effectively solves the problem.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.38 no.12
• /
• pp.1065-1072
• /
• 2014

The shape of an impeller splitter for a dual-inlet centrifugal blower was optimized to enhance the blower performance. Two design variable, the normalized chord and pitch of a splitter, were used to evaluate the blower performance and internal flow fields based on the three-dimensional flow analysis. The blower performance obtained using this numerical simulation had a maximum error of 4 percent compared to that in an experiment at the design flow condition. The shape optimization of the splitter successfully increased the blower efficiency and pressure by 3.65 and 1.14 percent compared to the reference values. The blower performance was increased by reducing the flow separation near the blade suction surface by optimizing the shape of the splitter, which produced a pressure increase at the outlet of the volute casing.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.20 no.4
• /
• pp.390-395
• /
• 2009

Design and optimization of a broadband meander-line twist reflector was performed for X-band application. Based on the equivalent transmission line model, the polarization twist performance was evaluated. Genetic analysis, particles swarm, and hybrid swarm optimizations were employed to obtain the optimized geometrical parameters. The optimized design exhibits low cross-polarization level below - 25 dB between 8.45 and 11.38 GHz. The polarization twist loss was below 0.2 dB. Comparison between computed and simulated results was also discussed.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.34 no.5
• /
• pp.531-537
• /
• 2010

This paper optimized the design parameters of the offset strip fin in a heat exchanger. To decrease the pressure drop and increase heat transfer, the performance factors such as j/f, $j/f^{1/3}$, and JF, which could be used to estimate the pressure drop and heat transfer simultaneously, were employed as the criteria for optimization. In the present study, STDQAO, PQRSM, and MGA were used for solving the constrained nonlinear optimization problem. The JF factor and heat transfer performance of the optimized offset-strip fin were greater than those of the reference offset-strip fin by 36% and 280%, respectively.

• Proceedings of the Korean Society of Propulsion Engineers Conference
• /
• 2011.04a
• /
• pp.397-400
• /
• 2011

MDO Optimization for a low pressure axial compressor rotor has been carried out to improve aerodynamic performance and structural stability. Global optimized solution was obtained from an artificial neural network model with genetic algorithm. Optimized rotor model has a high blade loading near hub and near zero incidence flow angle near tip region to reduce the incidence loss and flow separation at trailing edge region. Also the rotor shape is converged to a trapezoid shape to reduce the maximum stress occurred at the root of the blade. Numerical simulation results show that rotor has 87.6% rotor efficiency and safety factor over than 3.