• 산업기술연구
• /
• 제22권B호
• /
• pp.35-44
• /
• 2002

To find the optimal solution as rapidly and exactly as possible with Nelder-Mead simplex algorithm, the present values of the reflection, expansion, contraction and/or shrink parameters of this algorithm are needed to be changed at appropriate time during the search process. The reflection parameter is selected in this study in order to be changed because reflection, expansion and contraction process can be simultaneously effected by only this parameter. Two independent indices for determining whether the present value of the reflection parameter of this algorithm should be changed or not during the search process are suggested in this study. Those indices were made of the equations of Nelder-Mead simplex algorithm's convergence criterion and Dennis-Wood's convergence criterion, respectively. It is appeared that the optimal solution can be find with smaller numbers of objective function evaluation than the original Nelder-Mead's one with fixed parameter when the those indices are used during the search process. and the more remarkable reduction effect of the number of an objective function evaluation can be obtained when the latter index is used.

• 산업기술연구
• /
• 제19권
• /
• pp.361-368
• /
• 1999

The Nelder-Mead simplex algorithm has become on of the most widely used methods for nonlinear unconstrained optimization, since 1965. Recently, this algorithm has been reevaluated and many papers on this algorithm are being published. The MATLAB computer software, highly renown in engineering, also provides the Nelder-Mead algorithm and the Denis-Woods modification with FMINS function. The authors made C++ code of these algorithms and compared with FMINS on the convergence behavior and the exactness of solutions. It shows that MATLAB's FMINS is inferior to author's C++ code. So, FMINS should be corrected for every user.

• 산업기술연구
• /
• 제22권A호
• /
• pp.19-28
• /
• 2002

The Nelder-Mead simplex algorithm has been one of the most widely used methods for the nonlinear unconstrained optimization, since 1965. Recently, the new algorithm, (so-called the Fuzzy Simplex Algorithm), with fuzzy logic controllers for the expansion, reflection and contraction process of this algorithm has been proposed. In this paper, this new algorithm is developed. And, the formulation for the optimum design of the power screw's efficiency is made. And then, the developed fuzzy simplex algorithm as well as the original one is applied to this optimum design problem. The Fuzzy simplex algorithm results in a faster convergence in this problem, as reported in other study, too.

• 대한조선학회논문집
• /
• 제35권2호
• /
• pp.29-37
• /
• 1998

물수체의 선형 운동방정식은 형상 설계와 제어기 설계의 지배방정식으로 사용되는 중요한 설계인자이다. 그러나, 경험식, 이론적 계산 및 모형 시험으로부터 획득한 운동방정식 내의 조종 계수들은 오차를 포함하고 있어 해상 시험을 동해 이에 대한 검증을 수행하여야 한다. 본 연구에서는 실측 자료에 대해 병렬 확장 칼만 필터, Nelder & Mead Simplex 탐색법 및 유전적 알고리즘을 적용하여 몰수체의 주요 횡 동요 계수들을 식별/비교하였다. 결론적으로 Nelder & Mead Simplex 탐색법이 사용된 수학 모형과 해상 시험 자료에 대해 가장 만족스런 결과를 주고 있다.

• Steel and Composite Structures
• /
• 제19권4호
• /
• pp.995-1009
• /
• 2015

This study presents design optimization of spot welded structures to attain maximum strength by using the Nelder-Mead (Simplex) method. It is the main idea of the algorithm that the simulation run is executed several times to satisfy predefined convergence criteria and every run uses the starting points of the previous configurations. The material and size of the sheet plates are the pre-assigned parameters which do not change in the optimization cycle. Locations of the spot welds, on the other hand, are chosen to be design variables. In order to calculate the objective function, which is the maximum equivalent stress, ANSYS, general purpose finite element analysis software, is used. To obtain global optimum locations of spot welds a methodology is proposed by modifying the Nelder-Mead (Simplex) method. The procedure is applied to a number of representative problems to demonstrate the validity and effectiveness of the proposed method. It is shown that it is possible to obtain the global optimum values without stacking local minimum ones by using proposed methodology.

• Journal of Microbiology and Biotechnology
• /
• 제22권7호
• /
• pp.939-946
• /
• 2012

Antibiotic production with Streptomyces sindenensis MTCC 8122 was optimized under submerged fermentation conditions by artificial neural network (ANN) coupled with genetic algorithm (GA) and Nelder-Mead downhill simplex (NMDS). Feed forward back-propagation ANN was trained to establish the mathematical relationship among the medium components and length of incubation period for achieving maximum antibiotic yield. The optimization strategy involved growing the culture with varying concentrations of various medium components for different incubation periods. Under non-optimized condition, antibiotic production was found to be $95{\mu}g/ml$, which nearly doubled ($176{\mu}g/ml$) with the ANN-GA optimization. ANN-NMDS optimization was found to be more efficacious, and maximum antibiotic production ($197{\mu}g/ml$) was obtained by cultivating the cells with (g/l) fructose 2.7602, $MgSO_4$ 1.2369, $(NH_4)_2PO_4$ 0.2742, DL-threonine 3.069%, and soyabean meal 1.952%, for 9.8531 days of incubation, which was roughly 12% higher than the yield obtained by ANN coupled with GA under the same conditions.

• 한국전자파학회논문지
• /
• 제18권10호
• /
• pp.1203-1210
• /
• 2007

다중 빔 탐색 3D 레이다는 디지털 빔 형성 기술을 이용하여 수신시 적층 빔을 형성하는 최신의 3D 레이다 기술이다. 본 논문에서는 다중 경로 환경 상태에서 다중 빔 3D 레이다의 저고도 표적의 고도 추출을 위한 방법을 제안하고 고찰해 보고자 한다. 다중 빔 레이다에서 저고도 표적에서 발생하는 다중 경로 전파 및 레이다 신호 발생 모델링에 대해 기술하고, 거울 반사 상황에서 효과적으로 표적 고도를 추출하기 위해 nelder-mead simplex multipath reduction(NMSMR) 기법에 대해 기술한다. 제안된 알고리듬의 성능을 다양한 표적 고도와 레이다 주파수에 대해 시뮬레이션으로 확인하였다.

• 한국정보과학회논문지:시스템및이론
• /
• 제32권3호
• /
• pp.133-147
• /
• 2005

LCD BLU(Back-Light Unit)의 설계에는 광선추적 기법이 많이 사용되고 있으며, 이러한 기법을 통하여 도광판 전면으로 출사되는 빛의 휘도, 휘도의 균일도, 광효율 등을 수치적으로 예측할 수 있다. 휘도와 휘도의 균일도 등과 같은 항목은 BLU 설계에 있어서 매우 중요한 요소 중의 하나이며, 이러한 BLU의 광학적 특성은 도광판의 후면에 인쇄된 잉크패턴에 가장 많을 영향을 받고 있다 따라서 높은 휘도와 균일도를 발생시킬 수 있는 최적화된 잉크패턴을 설계하는 것은 BLU의 설계에 있어서 가장 중요한 부분 중의 하나이다. 본 논문에서는 최적의 잉크패턴을 설계하기 위하여 직접탐색(Direct Search) 기법 중에서 Holder Mead의 심플렉스탐색(Simplex Search) 알고리즘을 적용하였다. 직접탐색 기법은 최적의 잉크패턴을 계산하는 것과 같은 비선형적이고, 비연속적이며, 미분불가능한 함수의 최적점을 계산하는 분야에 많이 적용되고 있다. 본 논문에서는 여러 가지 실험을 통하여 심플렉스탐색 기법이 최적의 잉크패턴을 설계하는데 있어서 매우 효율적이며 안정적으로 사용될 수 있음을 보여주며, 이러한 최적화 기법으로 인하여 BLU 설계기능을 개선할 수 있음을 보여준다

• Structural Engineering and Mechanics
• /
• 제80권5호
• /
• pp.539-551
• /
• 2021

With the continuous increase of computational capacity, more and more complex nonlinear elastoplastic constitutive models were developed to study the mechanical behavior of elastoplastic materials. These constitutive models generally contain a large amount of physical and phenomenological parameters, which often require a large amount of computational costs to determine. In this paper, an inverse parameter determination method is proposed to identify the constitutive parameters of elastoplastic materials, with the consideration of both strain rate effect and temperature effect. To carry out an efficient design, a hybrid optimization algorithm that combines the genetic algorithm and the Nelder-Mead simplex algorithm is proposed and developed. The proposed inverse method was employed to determine the parameters for an elasto-viscoplastic constitutive model and Johnson-cook model, which demonstrates the capability of this method in considering strain rate and temperature effect, simultaneously. This hybrid optimization algorithm shows a better accuracy and efficiency than using a single algorithm. Finally, the predictability analysis using partial experimental data is completed to further demonstrate the feasibility of the proposed method.

• Structural Engineering and Mechanics
• /
• 제32권5호
• /
• pp.667-683
• /
• 2009

FE model-based dynamic analysis has been widely used to predict the dynamic characteristics of civil structures. In a physical point of view, an FE model is unavoidably different from the actual structure as being formulated based on extremely idealized engineering drawings and design data. The conventional model updating methods such as direct method and sensitivity-based parameter estimation are not flexible for model updating of complex and large structures. Thus, it is needed to develop a model updating method applicable to complex structures without restriction. The main objective of this paper is to present the model updating method based on the hybrid genetic algorithm (HGA) by combining the genetic algorithm as global optimization method and modified Nelder-Mead's Simplex method as local optimization method. This FE model updating method using HGA does not need the derivation of derivative function related to parameters and without application of complicated inverse analysis methods. In order to allow its application on diversified and complex structures, a commercial FEA tool is adopted to exploit previously developed element library and analysis algorithms. Moreover, an output-level objective function making use of measurement and analytical results is also presented to update simultaneously the stiffness and mass of the analysis model. The numerical examples demonstrated that the proposed method based on HGA is effective for the updating of the FE model of bridge structures.