• Structural Engineering and Mechanics
• /
• 제56권5호
• /
• pp.707-726
• /
• 2015

There is a growing trend of considering uncertainty in optimization process since last few decades. In this regard, Robust Design Optimization (RDO) scheme has gained increasing momentum because of its virtue of improving performance of structure by minimizing the variation of performance and ensuring necessary safety and feasibility of constraint under uncertainty. In the present study, RDO of reinforced concrete folded plate and shell structure has been carried out incorporating uncertainty in the relevant parameters by Monte Carlo Simulation. Folded plate and shell structures are among the new generation popular structures often used in aesthetically appealing constructions. However, RDO study of such important structures is observed to be scarce. The optimization problem is formulated as cost minimization problem subjected to the force and displacements constraints considering dead, live and wind load. Then, the RDO is framed by simultaneously optimizing the expected value and the variation of the performance function using weighted sum approach. The robustness in constraint is ensured by adding suitable penalty term and through a target reliability index. The RDO problem is solved by Sequential Quadratic Programming. Subsequently, the results of the RDO are compared with conventional deterministic design approach. The parametric study implies that robust designs can be achieved by sacrificing only small increment in initial cost, but at the same time, considerable quality and guarantee of the structural behaviour can be ensured by the RDO solutions.

• 한국공작기계학회논문집
• /
• 제14권2호
• /
• pp.27-32
• /
• 2005

This paper presents the optimization of expanding velocity for tube expanding process in the manufacturing of a heat exchanger. In specific, the expanding velocity has a great influence on the performance of a heat exchanger because it is a key variable determining the quantity of tube expending at assembly stage as well as a key Parameter determining overall production rate. The simulation showed that the genetic algorithm used in this paper resulted in the optimal tube expanding velocity by performing the following series of iteration; the generation of arbitrary population for tube expanding parameters, consequently the generation of tube expanding velocities, the evaluation of tube expanding quantity using the pre-trained data of plastic deformation by means of a neural network and finally the generation of next population using a penalty faction and a Roulette wheel method.

• Journal of Electrical Engineering and Technology
• /
• 제12권6호
• /
• pp.2127-2137
• /
• 2017

A virtual power plant (VPP) is a system that virtually integrates power resources based on the VPP participating customer (VPC) unit and operates as a power plant. When VPP operators manage resources to maximize their benefits, load reduction instructions may focus on more responsive VPCs, or those producing high profitability, by using VPC resources with high operation efficiency. VPCs may thus encounter imbalance problems during operation. This imbalance in operation time would bring more participation for some VPCs, causing potential degradation of their resources. Such an operation strategy would be not preferable for VPP operators in managing the relationship with VPCs. This issue impedes both continual VPC participation and economical and reliable VPP operation in the long term. An operation algorithm is therefore proposed that considers the operation time of VPC generators for mandatory reduction of power resource consumption. The algorithm is based on constraints of daily and annual operation times when VPP operators of local generators perform capacity-market power transactions. The algorithm maximizes the operator benefit through VPP operations. The algorithm implements a penalty parameter for imbalances in operation times spent by VPC generators in fulfilling their obligations. An evaluation was conducted on VPP operational effects by applying the algorithm to the Korean power market.

• Steel and Composite Structures
• /
• 제23권1호
• /
• pp.41-51
• /
• 2017

A $C^0$ FE model developed based on an efficient higher order zigzag theory is used for hygrothermal analysis of laminated composite plates. The $C^0$ FE model satisfies the inter-laminar shear stress continuity at the interfaces and zero transverse shear stress conditions at plate top and bottom. In this model the first derivatives of transverse displacement have been treated as independent variables to circumvent the problem of $C^1$ continuity associated with the above plate theory. In the present theory the above mentioned $C^0$ continuity of the present element is compensated in the stiffness matrix formulation by using penalty parameter approach. In order to avoid stress oscillations observed in the displacement based finite element, the stress field derived from temperature/moisture fields (initial strains) must be consistent with total strain field. Special steps are introduced by field consistent approach (e.g., sampling at gauss points) to compensate this problem. A nine noded $C^0$ continuous isoparametric element is used in the proposed FE model. Comparison of present numerical results with other existing solutions shows that the proposed FE model is efficient, accurate and free of locking.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제49권8호
• /
• pp.423-431
• /
• 2000

This paper presents a systematic approach to the input-output data-based fuzzy modeling for the complex and uncertain nonlinear systems, in which the conventional mathematical models may fail to give the satisfying results. To do this, we propose a new method that can yield a successful fuzzy model using a mGA hybrid schemes with a fine-tuning method. We also propose a new coding method fo chromosome for applying the mGA to the structure and parameter identifications of fuzzy model simultaneously. During mGA search, multi-purpose fitness function with a penalty process is proposed and adapted to guarantee the accurate and valid fuzzy modes. This coding scheme can effectively represent the zero-order Takagi-Sugeno fuzzy model. The proposed mGA hybrid schemes can coarsely optimize the structure and the parameters of the fuzzy inference system, and then fine tune the identified fuzzy model by using the gradient descent method. In order to demonstrate the superiority and efficiency of the proposed scheme, we finally show its applications to two nonlinear systems.

• International Journal of Aeronautical and Space Sciences
• /
• 제2권2호
• /
• pp.95-107
• /
• 2001

This paper is devoted to the parameter optimization of unmanned aerial vehicle's (UAV) flight control laws. Optimization procedure is based on the ideas of mixed $H_2/H_{\infty}$ control of multi-model plants. By using this approach, some partial $H_2$-terms defining the performance of nominal and parametrically perturbed Flight Control System (FCS) responses to deterministic command signals in stochastic atmosphere as well as $H_{\infty}$-terms defining robustness of the FCS can be incorporated in the composite cost function. Special penalty function imposed on the location of closed-loop system's poles keeps the speed of response and oscillatory properties for both nominal and perturbed FCS in reasonable limits. That is the reason why this procedure may provide reasonable trade-off between the performance and robustness of FCS that are very important especially for UAV. Its practical importance is illustrated by case studies of lateral and longitudinal control of small UAV.

• 한국통신학회논문지
• /
• 제29권10C호
• /
• pp.1402-1413
• /
• 2004

비강체 (non-rigid) 영상 등록에서 추정되는 좌표변환은 가역이어야 함으로 그 변환의 Jacobian 행렬식은 항상 양수 값을 가져야 한다. 본 논문에서는 이러한 가역 조건을 만족하는 좌표변환의 조건을 gradient 크기 제한의 조건으로 구한다. 또한 cubic B-spline을 이용한 변환 모델의 경우, 이 gradient 크기 제한 조건을 만족시키는 인수 집합을 이웃한 두 계수들의 차이가 제한된 인수들의 집합으로 구하였다. 이러한 인수들의 집합은 half space들의 교집합으로 이루어진 convex 집합이다. 본 논문에서는 이 convex 집합에 속하는 인수로 구성되는 좌표변환들 중에서 유사지수 (similarity measure) 를 최대로 만드는 변환을 gradient projection 최적화 기법을 통해 발견하였다. 이론적 분석, 폐 CT (Computed Tomography) 영상을 이용한 시뮬레이션 및 실험을 통하여, 제안된 알고리즘의 성능이 벌칙 함수 penalty function) 를 이용하는 기존의 방법보다 우수함을 증명하였다.

• 한국소음진동공학회논문집
• /
• 제18권11호
• /
• pp.1118-1127
• /
• 2008

The position and size of holes in the partition of a double cavity are known to strongly affect the eigenfrequency of the longitudinal eigenmodes of the double cavity. To maximize or minimize the eigenfrequency of the hole-partitioned double cavity, two acoustical topology optimization problems are formulated and solved. While two sub-cavities are filled with air, a partition between them is assumed to consist of sub-partitions of variable acoustical properties. One design variable is assigned to each sub-partition, whose material properties are interpolated as those of an intermediate material between air and a rigid body. The penalty parameter of the used interpolation function is adjusted to obtain a distinct air and rigid body distribution at the converged stage in each acoustical topology optimization problem. A special attention is paid to the selection of initial values of design variables to obtain solutions as close to global optimum and symmetric as possible. To show numerical characteristics of these optimization problems, the formulated problems are first solved for the one-dimensional partition design domain and then for the two-dimensional partition design domain.

• 한국전자파학회논문지
• /
• 제13권7호
• /
• pp.716-725
• /
• 2002

본 논문에서는 FWM(Four Wave Mixing)에 의한 채널 간 간섭이 존재하는 장거리 WDM 전송 시스템에서의 최적 펌프 전력 조건을 만족하는 전력 대칭 MSSI(Mid-Span Spectral Inversion) 보상법의 적용 가능성과 성능개선 정도를 시뮬레이션과 분석적 방법을 통해 검토해 보았다. 먼저 최적 펌프 전력 조건의 MSSI를 채택한 전송 시스템에 입력 광 펄스의 전력과 주파수 첩 파라미터의 변동에 따른 수신단에서의 EOP(Eye Opening Penalty) 평가를 통해 양호한 수신 성능을 유지할 수 있는 최대 송신 전력의 범위를 알아보았고, WDM 수신단에서 FWM에 의한 누화의 영향을 20 dB SNR 이상으로 할 수 있는 채널당 최대 입력 전력의 크기를 채널 간격에 따라 계산해 보았다. 광 송신단에서 변조된 광 펄스가 정도의 차이 없이 down-chirp되어 있는 WDM 전송 링크에 제안된 MSSI 보상을 적용하면 10 Gbps 전송의 경우 양호한 수신 특성을 유지시킬 수 있는 채널 당 최대 입력 전력을 5.3 dBm으로, 채널 간격을 0.4 nm로 하여 최대 68개의 채널 전송이 가능하다는 것을 확인할 수 있었다. 즉 제안한 최적 펌프 전력 조건 하에서의 전력 대칭 MSSI 보상법이 WDM 시스템의 장거리 초광대역 조건을 만족시킬 적합한 새로운 링크 전송 기술임을 확인할 수 있었다.

• 응용통계연구
• /
• 제33권4호
• /
• pp.365-377
• /
• 2020

선형모형에서 두개 이상의 설명변수들 사이에 존재하는 다중공선성 문제를 변수들 간에 내재되어 있는 공통의 구조인 인자를 구성하고, 인자들을 회귀변수로 사용하여 해결하는 인자회귀모형에 대하여 논의한다. 무한개로 가정 가능한 내재된 인자 중 유의미한 인자적재행렬을 구성하기 위하여 벌점모수의 값이 큰 LASSO 사전분포를 적용하는 베이지안 추정법을 사용한다. 결정된 인자적재행렬과 다른 모수들의 추정값을 각 설명변수의 선형모수로 역변환 하여, 새로운 관측값에 대한 예측 모형으로도 사용한다. 제안한 방법을 제품 서비스 관리 자료에 적용하여 정해진 인자의 개수에 대한 인자가 일반적인 공통인자회귀모형과 동일한 결과를 나타냄을 확인하였고, 일반적인 공통인자회귀모형과 비교를 위해 계산한 평균 제곱 오차값이 더 작다는 것을 알 수 있었다.