A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
An Iterative Convex Optimization Procedure for Structural System Identification
Owing to the complexity of civil structures, structural behavior predicted by finite element models (built according to design drawings) is usually different from behavior of an actual structure in the field. To improve the prediction accuracy, finite element (FE) model updating can be performed based on sensor measurement from the actual structure. Numerous FE model-updating algorithms have been developed in the past few decades. However, most existing algorithms suffer computational challenges when applied to large structures. The challenges are usually because most existing algorithms attempt to solve non-convex optimization problems. As a result, the optimization process encounters convergence difficulty and cannot guarantee global optimum. To address the issue, this paper proposes an iterative convex optimization algorithm for FE model updating. The convex attribute of the optimization problem makes the solution process tractable and efficient. To validate the proposed algorithm, numerical simulation is conducted with a plane truss structure. The proposed approach successfully updates the simulation model and shows certain advantage over a conventional FE model updating algorithm.
An Iterative Convex Optimization Procedure for Structural System Identification
Owing to the complexity of civil structures, structural behavior predicted by finite element models (built according to design drawings) is usually different from behavior of an actual structure in the field. To improve the prediction accuracy, finite element (FE) model updating can be performed based on sensor measurement from the actual structure. Numerous FE model-updating algorithms have been developed in the past few decades. However, most existing algorithms suffer computational challenges when applied to large structures. The challenges are usually because most existing algorithms attempt to solve non-convex optimization problems. As a result, the optimization process encounters convergence difficulty and cannot guarantee global optimum. To address the issue, this paper proposes an iterative convex optimization algorithm for FE model updating. The convex attribute of the optimization problem makes the solution process tractable and efficient. To validate the proposed algorithm, numerical simulation is conducted with a plane truss structure. The proposed approach successfully updates the simulation model and shows certain advantage over a conventional FE model updating algorithm.
An Iterative Convex Optimization Procedure for Structural System Identification
Zhu, Dapeng (author) / Dong, Xinjun (author) / Wang, Yang (author)
ASCE International Workshop on Computing in Civil Engineering ; 2013 ; Los Angeles, California
2013-06-24
Conference paper
Electronic Resource
English
An Iterative Convex Optimization Procedure for Structural System Identification
| British Library Conference Proceedings | 2013
Iterative design and structural optimization
| Engineering Index Backfile | 1966
Sequential Convex Programming for Structural Optimization Problems
| Springer Verlag | 1993
Structural shape optimization using self-adjusted convex approximation
| British Library Online Contents | 2002