• Journal of Pharmaceutical Investigation
• /
• 제6권2호
• /
• pp.69-82
• /
• 1976

Tablet product design problem was structured as constrained optimization problem and subsequently solved by multiple regression analysis and Lagrangian method of optimization. We used Lagrangian method for the purpose of finding the reason of the previous results. Biodiastase and cellulase were the enzymes, chosen, $Avicel{\circledR}$ and corn starch or calcium carboxy methyl cellulose were the binder and disintegrant, respectively. The effect of the dry binder and disintegrant concentration on tablet hardness, friability, volume, disintegration time was recorded. Optimization of this parameter was studied by using the constrained optimization method. In addition to finding a optimal condition of the enzyme tablets, the application of sensitivity analysis studies to such problems was also illustrated. In order to get a stable preparations of the enzyme tablets, accelerated test of coating tablets was carried out in this study. the results are as follows. 1) The minimum disintegration time, such that the average tablet volume did not exceed 0.0154 cubic inch and the average friability value did not exceed 0.62%, was 6.6 minutes and then $Avicel{\circledR}$ and corn starch were 15.4% and 17.2%, respectively. 2) The multiple-correlation coefficients for the regression models of tablet hardness, friability, disintegration time and volume were with in the 95% confidence range. 3) According to the test results, calcium carboxymethyl cellulose can be used as a disintegrant instead of corn starch.

• 응용통계연구
• /
• 제27권5호
• /
• pp.693-703
• /
• 2014

실험영역을 벗어나는 다중 외삽점들에 관한 실험설계를 기획하는 경우 실험자는 종종 어느 외삽점에 더 많은 노력을 집중하여야 하는지 주어진 모형이 있다하더라도, 고민하는 경우가 있다. 본 연구에서는 이러한 상황에 관한 실험설계 문제를 다루었다. 첫 번째는 주어진 모형이 실험영역을 벗어나더라도 모형이 타당한 경우 다중 외삽점에 관한 실험설계고 다른 하나는 그렇지 않은 경우이다. 첫 번째인 경우는 비교적 기존 문헌에서 알려진 방법들이 적용될 수 있으나 그렇지 않은 경우 즉, 모형의 타당성이 의심되는 경우는 다른 실험설계기준을 제시하여야 한다, 본 연구는 이와 관련 다양한 하이브리드 방법을 제시하여 다중 외삽점에서의 문제가 어떻게 모형 불확실성하에서 전개되어야 하는지 다루어 보았다, 이를 위해 서치알고리즘의 하나인 유전알고리즘을 적용하였다. 왜냐하면 전통적인 교환알고리즘의 복잡성보다는 유전알고리즘의 효율성이 더 뛰어났다고 보기 때문이다.

• 한국강구조학회 논문집
• /
• 제10권3호통권36호
• /
• pp.553-564
• /
• 1998

최근도심지의 극심한 교통체증의 해소를 위한 방안으로 2층도로 강구조 덱 시스템의 채택이 증가하고 있다. 그 주된 이유는 신속한 시공과 콘크리트덱에 비하여 자중의 감소효과가 크고 보다 큰 강도와 효율적인 가설이 가능하기 때문이다. 본 연구의 목적은 강구조 덱 시스템의 최적설계프로그램을 이용한 2층도로용 강구조 덱 시스템의 가장 합리적인 형식 결정에 관한 연구이다. 최적화의 목적함수는 최소 초기비용으로 정식화하였다. 설계 제약조건은 도로교시방서의 ASD설계기준에 맞추어 정식화하였으며 최적설계과정은 두 단계로 이루어져 있다. 첫 번째 단계는 박스 또는 플레이트주형에 대한 강구조 덱 시스템의 최적설계가 수행된다. 그리고 두 번째 단계에서 개단면 또는 폐단면리브를 갖는 강상판에 대하여 최적설계를 한다. 최적설계 프로그램의 구조해석은 주형에 대하여 격자해석 모델링을 이용하였고 강상판에 대해서는 Pelican-Esslinger법을 사용하였다. 최적화 기법은 SQP를 이용하였다. 적용 예의 각 형식별 강구조 덱 시스템의 최적설계 결과의 비교를 통하여 비용의 효율성을 검토하였고 폐단면리브를 가지는 직선 형상의 박스거더 형식이 가장 효율적이고 경제적으로 판단된다.

• 한국전산구조공학회논문집
• /
• 제32권4호
• /
• pp.249-256
• /
• 2019

이 논문은 재료의 전기 전도도 분포를 재구성하는 전기임피던스 단층이미지 기법(electrical impedance tomography; EIT)을 제시한다. 이 문제는 구조물 표면의 전극에서 측정된 전위와 계산된 전위의 차를 최소화하여 전기 전도도의 공간적 분포를 재구성하는 최적화 문제로 정의된다. 전류 입력 시 전위를 구하는 정해석 문제의 수학적 모델로서 완전전극모델(complete electrode model; CEM)을 사용하였다. 완전전극모델은 전기 포텐셜에 대한 라플라스 방정식과 전류 입력에 따른 경계조건들로 구성되는 경계값 문제이다. 완전전극모델 해의 정확성을 검증하기 위하여 유한요소법을 이용해 구한 원형 구조물의 전위해와 Technology Computer Aided Design(TCAD) 소프트웨어를 사용해 얻은 결과를 비교하였다. 완전전극모델의 지배방정식과 경계조건을 구속조건으로 하는 최적화 문제를 라그랑주 승수법(lagrange multiplier method)을 이용해 비구속 최적화 문제로 전환하고 라그랑지안의 1차 최적화 조건으로부터 전극에서의 전위 차를 최소화하는 최적의 전기전도도 분포를 도출하였다. 원형 균일영역의 전기 전도도 분포를 재구성하는 역해석 예제를 통해 완전전극모델 기반 EIT 프레임워크의 적용성을 검토하였다.

• 한국소음진동공학회논문집
• /
• 제25권12호
• /
• pp.856-865
• /
• 2015

In a high-speed press, numerous moving links are interconnected and each link executes a constrained motion at high speed. As a consequence, high-level dynamic unbalance force and unbalance moment are transmitted to the main frame of the press, which results in unwanted vibration and significantly degrades manufacturing accuracy. Dynamic unbalance force and unbalance moment inevitably transmits high-level vibrational force to the foundation on which the press is installed. Minimizing the vibrational force transmitted to the foundation is critical for the protection of both the operators and the surrounding structures. The whole task should be carried out in two steps. The first step is to reduce dynamic unbalance based upon kinematic and dynamic analyses. The second step is to design and build an optimal vibration isolation system minimizing the vibrational force transmitted to the foundation. Firstly, the dynamic design method is presented to reduce dynamic unbalance force and moment. For this a 3D CAD software was utilized and a computer program was written to compute dynamic unbalance force and moment. Secondly, the design method for vibration isolation system is presented. The method for designing coil springs and viscous dampers are explained in detail.

• Steel and Composite Structures
• /
• 제3권4호
• /
• pp.261-275
• /
• 2003

Truss members built-up with double angles back-to-back have monosymmetric cross-section and twisting always accompanies flexion upon the onset of buckling about the axis of symmetry. Approximate formulae for calculating the buckling capacity are presented in this paper for routine design purpose. For a member susceptible only to flexural buckling, its optimal cross-section should consist of slender plate elements so as to get larger radius of gyration. But, occurrence of twisting changes the situation owing to the weakness of thin plates in resisting torsion. Criteria for limiting the leg slenderness are discussed herein. Truss web members in compression are usually considered as hinged at both ends for out-of-plane buckling. In case one (or both) end of member is not supported laterally by bracing member, its adjoining members have to provide an elastic support of adequate stiffness in order not to underdesign the member. The stiffness provided by either compression or tension chords in different cases is analyzed, and the effect of initial crookedness of compression chord is taken into account. Formulae are presented to compute the required stiffness of chord member and to determine the effective length factor for inadequately constrained compressive diagonals.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권1호
• /
• pp.1-19
• /
• 2020

In an underlay cognitive simultaneous wireless information and power transfer (SWIPT) network, communication from secondary user (SU) to secondary destination (SD) is accomplished with decode-and-forward (DF) relays. Multiple energy-constrained relays are assumed to harvest energy from SU via power splitting (PS) protocol and complete SU secure information transmission with beamforming. Hence, physical layer security (PLS) is investigated in cognitive SWIPT network. In order to interfere with eavesdropper and improve relay's energy efficiency, a destination-assisted jamming scheme is proposed. Namely, SD transmits artificial noise (AN) to interfere with eavesdropping, while jamming signal can also provide harvested energy to relays. Beamforming vector and power splitting ratio are jointly optimized with the objective of SU secrecy capacity maximization. We solve this non-convex optimization problem via a general two-stage procedure. Firstly, we obtain the optimal beamforming vector through semi-definite relaxation (SDR) method with a fixed power splitting ratio. Secondly, the best power splitting ratio can be obtained by one-dimensional search. We provide simulation results to verify the proposed solution. Simulation results show that the scheme achieves the maximum SD secrecy rate with appropriate selection of power splitting ratio, and the proposed scheme guarantees security in cognitive SWIPT networks.

• 한국지능시스템학회논문지
• /
• 제11권7호
• /
• pp.680-688
• /
• 2001

In this paper, we propose a reliable method for searching the optimally performing generalized brain-state-in-a-box (GBSB) neural associative memory using an evolution program (EP) given a set of prototype patterns to be stored as stable equilibrium points. First, we exploit some qualitative guidelines necessary to synthesize the GBSB model. Next, we parameterize the solution space utilizing the limited number of parameters to represent the solution space. Then, we recast the synthesis of GBSB neural associative memories as two constrained optimization problems, which are equivalent to finding a solution to the original synthesis problem. Finally, we employ an evolution program (EP), which enables us to find an optimal set of parameters related to the size of domains of attraction (DOA) for prototype patterns. The validity of this approach is illustrated by a design example and computer simulations.

• International Journal of CAD/CAM
• /
• 제6권1호
• /
• pp.173-181
• /
• 2006

In this paper, an integration framework of Geometric Constraint Solving Engine and AutoCAD is presented, and a dynamic geometric constraint system is introduced. According to inherent orientation features of geometric entities and various Object Snap results of AutoCAD, the' proposed system can automatically construct an under-constrained geometric constraint model during interactive drawing. And then the directed constraint graph in a geometric constraint model is realtime modified in order to produce an optimal constraint solving sequence. Due to the open object-oriented characteristics of AutoCAD, a set of user-defined entities including basic geometric elements and graphics constraint relations are defined through derivation. And the custom-made Object Reactor and Command Reactor are also constructed. Several powerful characteristics are achieved based on these user-defined entities and reactors, including synchronously processing geometric constraint information while saving and opening DWG files, visual constraint relations, and full adaptability to Undo/Redo operations. These characteristics of the proposed system can help the designers more easily manage geometric entities and constraint relations between them.

• 유통과학연구
• /
• 제13권6호
• /
• pp.11-15
• /
• 2015

Purpose - Robot path planning, a constrained optimization problem, has been an active research area with many methods developed to tackle it. This study proposes the use of a Rapidly-exploring Random Tree and Particle Swarm Optimizer algorithm for path planning. Research design, data, and methodology - The grid method is built to describe the working space of the mobile robot, then the Rapidly-exploring Random Tree algorithm is applied to obtain the global navigation path and the Particle Swarm Optimizer algorithm is adopted to obtain the best path. Results - Computer experiment results demonstrate that this novel algorithm can rapidly plan an optimal path in a cluttered environment. Successful obstacle avoidance is achieved, the model is robust, and performs reliably. The effectiveness and efficiency of the proposed algorithm is demonstrated through simulation studies. Conclusions - The findings could provide insights to the validity and practicability of the method. This method makes it is easy to build a model and meet real-time demand for mobile robot navigation with a simple algorithm, which results in a certain practical value for distribution environments.