• 대한기계학회논문집A
• /
• 제27권8호
• /
• pp.1281-1286
• /
• 2003

A line search algorithm to increase a convergence in Newton's method is developed and applied to nonlinear finite element analysis. The algorithm is based on the slack line search theory which is an efficient algorithm to determine initial acceleration coefficient, variable backtracking algorithm proposed by some researchers, and convergence criterion based on residual norm. Also, it is capable of avoiding exceptional diverging conditions. Developed program is tested in metal forming simulation such as forging and ring rolling. Numerical result shows the validity of the algorithm for a highly nonlinear system .

• 대한토목학회논문집
• /
• 제4권4호
• /
• pp.127-135
• /
• 1984

본 연구는 실제 하중을 받는 강재 아치구조의 최적형상을 다루었다. 목적함수로는 아치의 중량을 취했으며, 제약조건으로는 단면력, 체적, 아치리브의 길이, 아치부채 단연적의 조합으로 구성된 웅력제약초건을 고려하였다. 형상최적화문제는 아치부재의 단면척이 설계변수의 항으로 형성되었으며, 첫단계로 구조해석의 정밀도가 최적 설계의 목적함수값에 미치는 영향을 분석하였다. 형상최적화 알고리즘은 Two-space System 으로 형성되었고, Space 1 에서는 Modified Newton-Raphson Method에 의한 단면최적화, Space 2 에서는 Powell Method 에 의한 형상최적화를 시도하여 형상최적화 알고리즘을 도출하였다. 본 연구에서 도출된 최적화 알고리즘을 이용하여 아치의 단면최적화와 최적 아치형상에 관한 연구가 수행되었다. 이 알고리즘에 의해 실제 조건하에서 아치의 최적 Span-Rise 비를 구할 수 있다.

• Journal of Electrical Engineering and Technology
• /
• 제2권3호
• /
• pp.312-319
• /
• 2007

The solution of the power flow is one of the most important problems in electrical power systems. These traditional methods such as Gauss-Seidel method and Newton-Raphson (NR) method have had drawbacks up to now such as initial values, abnormal operating solutions and divergences in heavy loads. In order to overcome theses problems, the power flow solution incorporating genetic algorithm (GA) is introduced in this paper. General operator of genetic algorithm, arithmetic crossover, and non-uniform mutation operator of GA are suggested to solve the power flow problem. While abnormal solution cannot be obtained by a NR method, multiple power flow solution can be obtained by a GA method. With a heavy load, both normal solution and abnormal solution can be obtained by a proposed method. In this paper, a floating number representation instead of the binary number representation is introduced for accuracy. Simulation results have been compared with traditional methods.

• 한국컴퓨터정보학회논문지
• /
• 제15권4호
• /
• pp.57-63
• /
• 2010

Levenberg-Marquardt 알고리즘에서 감쇠 파라미터는 오류역전파 학습과 Gauss-Newton 학습의 스위치 역할을 하며 학습 속도에 영향을 준다. 이런 감쇠 파라미터를 고정시키는 것은 오차 함수의 진동을 유발하고 학습 속도를 감소시킨다. 따라서 본 논문은 오차 함수의 변화 과정을 참조하여 감쇠 파라미터를 가변적으로 적용하는 방법을 제안한다. 제안된 방법은 오차의 변화량이 크면 감쇠 파라미터를 크게, 오차의 변화량이 작으면 감쇠 파라미터를 작게 조정한다. 이것은 모멘텀과 유사한 역할을 하여 학습 속도를 향상시킨다. 제안된 방법의 검증을 위한 실험으로는 iris 분류 문제와 wine 분류 문제를 사용하였다. 제안된 방법은 iris 분류 문제에서는 67% 학습에서, wine 분류 문제에서는 78% 학습에서 학습 속도가 향상되었으며 기존 방법과 비교하여 오차의 진동도 적은 것을 확인할 수 있었다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.296-301
• /
• 2005

This paper describes the kinematics, dynamics and control of a 6-DOF shaking table with a bell crank structure, which converts the direction of reciprocating movements. In this shaking table, the bell crank mechanism is used to reduce the amount of space needed to install the shaking table and create horizontal displacement of the platform. In kinematics, joint design is performed using $Gr{\ddot{u}}bler's$ formula. The inverse kinematics of the shaking table is discussed. The derivation of the Jacobian matrix is presented to evaluate singularity conditions. Considering the maximum stroke of the hydraulic actuator, collision between links and singularity, workspace is computed. In dynamics, computations are based on the Newton-Euler formulation. To derive parallel algorithms, each of the contact forces is decomposed into one acting in the direction of the leg and the other acting in the plane orthogonal to the direction of the leg. Applying the Newton-Euler approach, the solution of inverse dynamics is almost completely parallel. Only one of the steps-the application of the Newton-Euler equations to the platform-must be performed on one single processor. Finally, the efficient control scheme is proposed for the tracking control of the motion platform.

• 한국전산구조공학회논문집
• /
• 제29권3호
• /
• pp.237-244
• /
• 2016

본 연구는 재료비선형 문제를 다루기 위한 비선형 MLS 차분법의 정식화 과정을 제시한다. MLS 차분법은 절점모델을 기반으로 고속 미분근사식을 활용하여 지배 미분방정식을 직접 이산화 하는데, 변수를 변위로 일원화한 Navier 방정식을 사용하여 탄성재료 문제를 다룬 기존의 MLS 차분법은 재료의 구성방정식을 별도로 고려할 수 없다. 본 연구에서는 비선형 재료의 구성방정식을 반영할 수 있는 강정식화를 위해 1차 미분근사를 반복 사용하는 겹미분근사를 고안했다. 응력의 발산으로 표현되는 평형방정식을 그대로 이산화하고 Newton 방법을 적용하여 반복계산을 통해 수렴해를 찾는 비선형 알고리즘을 제시했다. 응력 계산과 내부변수의 갱신은 return mapping 알고리즘을 활용하였고, 알고리즘 접선계수(algorithmic tangent modulus)의 적용을 통해 빠르고 안정적인 반복계산이 가능하도록 하였다. 재생성 시험을 통해 겹미분근사의 정당성을 검증했고, 비선형재료에 대한 인장문제의 해석을 통해 개발된 비선형 MLS 차분 알고리즘의 정확성과 안정성을 확인하였다.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 2004년도 봄 학술발표회 논문집
• /
• pp.1-8
• /
• 2004

For the linearized differential algebraic equation of the nonlinear constrained system, exact initial values of the acceleration are needed to solve itself. It may be very troublesome to perform the inverse operation for obtaining the incremental quantities since the mass matrix contains the zero element in the diagonal. This fact makes the mass matrix impossible to be positive definite. To overcome this singularity phenomenon the mass matrix needs to be modified to allow the feasible application of predictor and corrector in the iterative computation. In this paper the proposed numerical algorithm based on the modified mass matrix combines the conventional implicit algorithm, Newton-Raphson method and Newmark method. The numerical example presents reliabilities for the proposed algorithm via comparisons of the 4th order Runge-kutta method. The proposed algorithm seems to be satisfactory even though the acceleration, Lagrange multiplier, and energy show unstable behaviour. Correspondingly, it provides one important clue to another algorithm for the enhancement of the numerical results.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1999년도 추계학술대회 논문집 전문대학교육위원 P
• /
• pp.43-51
• /
• 1999

The power flow calculations(PFc) are the most important and powerful tools in power systems engineering. The conventional power flow problem is solved generally with numerical methods such as Newton-Raphson(NR). The conventional numerical method generally have some convergency problem, which is sensitive to initial value, and numerical stability problem concerned with jacobian matrix inversion. This paper presents a new PFc algorithm based on the improved genetic algorithm (IGA) which can overcome the disadvantages mentioned above. The parameters of GA, with dynamical hierarchy of the coding system, are improved to make GA a practical algorithm in the problem of real system. Some case studies with test bus system also present to show the performance of proposed algorithm. The results of proposed algorithm are compared with the results of PFc obtained using a conventional NR method.

• Journal of applied mathematics & informatics
• /
• 제22권1_2호
• /
• pp.193-208
• /
• 2006

A modified BFGS algorithm for solving the unconstrained optimization, whose Hessian matrix at the minimum point of the convex function is of rank defects, is presented in this paper. The main idea of the algorithm is first to add a modified term to the convex function for obtain an equivalent model, then simply the model to get the modified BFGS algorithm. The superlinear convergence property of the algorithm is proved in this paper. To compared with the Tensor algorithms presented by R. B. Schnabel (seing [4],[5]), this method is more efficient for solving singular unconstrained optimization in computing amount and complication.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2002년도 합동 추계학술대회 논문집 정보 및 제어부문
• /
• pp.513-516
• /
• 2002

In electrical impedance tomography(EIT), various image reconstruction algorithms have been used in order to compute the internal resistivity distribution of the unknown object with its electric potential data at the boundary. Mathematically the EIT image reconstruction algorithm is a nonlinear ill-posed inverse problem. This paper presents two intelligent optimization algorithm techniques such as genetic algorithm and simulated annealing for the solution of the static EIT inverse problem. We summarize the simulation results for the three algorithm forms: modified Newton-Raphson, genetic algorithm, and simulated annealing.