TY - GEN
T1 - Nonlinear random response of large-scale sparse finite element structural problems
AU - Chokshi, Swati
AU - Mei, Chuh
AU - Nguyen, Duc T.
AU - Rajan, Subramaniam
PY - 2007/8/6
Y1 - 2007/8/6
N2 - Efficient numerical procedures (modal method) for solving large-scale, sparse, nonlinear large-deflection random vibration problems are reviewed. Solution for the small number (say less than 100) of lowest linear eigen-pairs needs to be solved for only once, in order to transform nonlinear large displacements from the conventional "structural" degree-of-freedom (dof) into the "modal" dof. The reduced coupled nonlinear equations of motion in modal dof can be inexpensively solved by the popular Runge-Kurta (RK) or any other time integrating method. The main focus of this work will be placed upon computational issues involved in the nonlinear modal methodologies. Major time consuming portions of the nonlinear modal method are firstly identified. Then, efficient sparse and dense matrix technologies are proposed and incorporated into the developed procedures. Small, medium, and large-scale single panel models are used to validate and evaluate their numerical performance. Comparisons (in terms of numerical accuracy and computational time) between the developed code with existing solution, including popular commercialized finite element software, such as ABAQUS are included, whenever possible. Results obtained to this date indicate that the developed algorithms and software are accurate and highly efficient.
AB - Efficient numerical procedures (modal method) for solving large-scale, sparse, nonlinear large-deflection random vibration problems are reviewed. Solution for the small number (say less than 100) of lowest linear eigen-pairs needs to be solved for only once, in order to transform nonlinear large displacements from the conventional "structural" degree-of-freedom (dof) into the "modal" dof. The reduced coupled nonlinear equations of motion in modal dof can be inexpensively solved by the popular Runge-Kurta (RK) or any other time integrating method. The main focus of this work will be placed upon computational issues involved in the nonlinear modal methodologies. Major time consuming portions of the nonlinear modal method are firstly identified. Then, efficient sparse and dense matrix technologies are proposed and incorporated into the developed procedures. Small, medium, and large-scale single panel models are used to validate and evaluate their numerical performance. Comparisons (in terms of numerical accuracy and computational time) between the developed code with existing solution, including popular commercialized finite element software, such as ABAQUS are included, whenever possible. Results obtained to this date indicate that the developed algorithms and software are accurate and highly efficient.
UR - http://www.scopus.com/inward/record.url?scp=34547502785&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=34547502785&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:34547502785
SN - 1563478927
SN - 9781563478925
T3 - Collection of Technical Papers - AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics and Materials Conference
SP - 7891
EP - 7900
BT - Collection of Technical Papers - 48th AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference
T2 - 48th AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference
Y2 - 23 April 2007 through 26 April 2007
ER -