Keywords: artificial neural networks, computational intelligence, structural mechanics, finite element method, surrogate modelling, uncertainty.

Computational Intelligence is applied to several fields in structural engineering. Biologically- inspired computational approaches, such as artificial neural networks, evolutionary algorithms and fuzzy logic are used, e.g., for the modelling of material and structural behaviour, structural optimization, reliability assessment, sensitivity analysis, system and parameter identification, structural health monitoring, computer aided steering of structural processes, and structural control. This contribution contains a review on the application of artificial neural network approaches in structural mechanics. First, an overview on different architectures of artificial neural networks used in structural mechanics is presented. It includes feedforward and recurrent network architectures. Also strategies are discussed to consider uncertainties of structural parameters within artificial neural network approaches. Artificial neural networks can be utilized on the material and on the structural level. A review of neural network applications for surrogate modelling of computationally intensive structural models and structural processes is presented. Also, the combination of neural network based material models with the finite element method is shown and special network structures for three-dimensional material models are presented.

purchase the full-text of this paper (price £20)

go to the next paper
return to the table of contents
return to Computational Technology Reviews
purchase this volume (price £75 +P&P)