• Structural Engineering and Mechanics
• /
• 제36권5호
• /
• pp.529-544
• /
• 2010

Nonlinear equations of structures are generally solved numerically by the iterative solution of linear equations. However, this iterative procedure diverges when the tangent stiffness is ill-conditioned which occurs near limit points. In other words, a major challenge with simple iterative methods is failure caused by a singular or near singular Jacobian matrix. In this paper, using the Newton-Raphson algorithm based on Davidenko's equations, the iterations can traverse the limit point without difficulty. It is argued that the propose algorithm may be both more computationally efficient and more robust compared to the other algorithm when tracing path through severe nonlinearities such as those associated with structural collapse. Two frames are analyzed using the proposed algorithm and the results are compared with the previous methods. The ability of the proposed method, particularly for tracing the limit points, is demonstrated by those numerical examples.

• 한국지능시스템학회논문지
• /
• 제14권7호
• /
• pp.907-913
• /
• 2004

일반적인 이동 로봇의 주된 관심은 경로 생성과 생성된 경로 추종에 있다. 그러나 일부 고속의 이동성이 필요로 하는 로봇의 경우 동역학적 제한 조건이 존재하며, 이러한 제한 조건 내에서 원하는 움직임에 대한 제어가 요구된다. 된 논문에서 환경 지도를 가지고 있지 않은 상태, 즉 미지의 환경에서 이동 로봇의 경로 추종에 있어서 빠른 이동시에 발생할 수 있는 이동 로봇의 미끄러짐이나 전복 현상을 막기 위해 이동 로봇의 동역학적 제한 조건을 퍼지 논리를 이용하여 기준 속도를 변화시켜 안전하고 빠는 경로 추종 성능을 얻고자 하였다. 특히, 라인 추종 이동 로봇을 모델링하여 실시간으로 변화하는 목표점에 대한 추종 제어기를 설계하고 퍼지 최적 속도 제한 제어기를 통해 연속적으로 변화하는 라인에 대해서 지능적으로 로봇의 속도를 제한하여 안정적인 추종 성능을 발휘함을 확인하였다.

• Structural Engineering and Mechanics
• /
• 제48권6호
• /
• pp.879-914
• /
• 2013

In part I of the article, formulation and characteristics of the several well-known structural geometrical nonlinear solution techniques were studied. In the present paper, the efficiencies and capabilities of residual load minimization, normal plane, updated normal plane, cylindrical arc length, work control, residual displacement minimization, generalized displacement control and modified normal flow will be evaluated. To achieve this goal, a comprehensive comparison of these solution methods will be performed. Due to limit page of the article, only the findings of 17 numerical problems, including 2-D and 3-D trusses, 2-D and 3-D frames, and shells, will be presented. Performance of the solution strategies will be considered by doing more than 12500 nonlinear analyses, and conclusions will be drawn based on the outcomes. Most of the mentioned structures have complex nonlinear behavior, including load limit and snap-back points. In this investigation, criteria like number of diverged and complete analyses, the ability of passing load limit and snap-back points, the total number of steps and analysis iterations, the analysis running time and divergence points will be examined. Numerical properties of each problem, like, maximum allowed iteration, divergence tolerance, maximum and minimum size of the load factor, load increment changes and the target point will be selected in such a way that comparison result to be highly reliable. Following this, capabilities and deficiencies of each solution technique will be surveyed in comparison with the other ones, and superior solution schemes will be introduced.