INVESTIGATION OF FRAMED SHEARWALL BEHAVIOR WITH NEURAL NETWORKS

To date, concrete structure modelling has involved the development of mathematical models of concrete structure behaviour derived from human observation of, and reasoning with, experimental data. An alternative , discussed in this Paper, is to use a computation and knowledge repre sentation paradigm, called neural networks, developed by researchers i n connectionism (a sub-field of artificial intelligence) to model conc rete structure behaviour. The main benefits in using a neural network approach are that all behaviour can be represented within a unified en vironment of a neural network and the network is built directly from o m experimental data using the self-organizing capabilities of the neur al network, i.e. the network is presented with the experimental data a nd learns the relationships between stresses and strains. Such a strat egy has important implications for modelling the behaviour of modern, complex concrete structures. In this Paper, the behaviour of reinforce d concrete framed shearwalls is modelled with a back-propagation neura l network. Preliminary results of the use of networks to model concret e structure look promising.