A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
An efficient Latin hypercube sampling for probabilistic nonlinear finite element analysis of reinforced concrete structures
Probabilistic analysis of concrete members is mainly needed for safety evaluations of existing structures. The well-known Monte Carlo simulation is very time-demanding and is therefore not suitable for use in practice, especially for nonlinear finite element analysis of concrete structures. This paper presents a new optimal Latin hypercube sampling method for design of a computer experiment taking into account the sensitivity factors regarding the behaviour of the considered system. The method avoids extensive numerical calculations. The basic idea of this sampling strategy is to use a weighting factor for each dimension of the probability space in order to consider the different influences of the random variables in the optimisation process of Latin hypercube samples. The method is evaluated using examples of reinforced concrete structures, including an example regarding the spatial variability of concrete properties. The results show that with only 10 simulations the statistical values of the load-bearing capacities of concrete structures can be determined with high accuracy, if the new weighted Latin hypercube sampling is used.
An efficient Latin hypercube sampling for probabilistic nonlinear finite element analysis of reinforced concrete structures
Probabilistic analysis of concrete members is mainly needed for safety evaluations of existing structures. The well-known Monte Carlo simulation is very time-demanding and is therefore not suitable for use in practice, especially for nonlinear finite element analysis of concrete structures. This paper presents a new optimal Latin hypercube sampling method for design of a computer experiment taking into account the sensitivity factors regarding the behaviour of the considered system. The method avoids extensive numerical calculations. The basic idea of this sampling strategy is to use a weighting factor for each dimension of the probability space in order to consider the different influences of the random variables in the optimisation process of Latin hypercube samples. The method is evaluated using examples of reinforced concrete structures, including an example regarding the spatial variability of concrete properties. The results show that with only 10 simulations the statistical values of the load-bearing capacities of concrete structures can be determined with high accuracy, if the new weighted Latin hypercube sampling is used.
An efficient Latin hypercube sampling for probabilistic nonlinear finite element analysis of reinforced concrete structures
Tran, Ngoc Linh (author) / Graubner, Carl-Alexander (author) / Rombach, Günter Axel (author)
2019-08-22
Article/Chapter (Book)
Electronic Resource
English
DDC:
690
Latin Hypercube Sampling for Stochastic Finite Element Analysis
| Online Contents | 2002
RELIABILITY ANALYSIS BASED ON LATIN HYPERCUBE SAMPLING OF CURRENT REINFORCED CONCRETE STRUCTURES
| British Library Conference Proceedings | 2000
| British Library Conference Proceedings | 1997