Investigation of Prestressed Concrete Frame Behavior with Neural Networks

Abstract

To date, concrete structure modeling has involved the development of mathematical models of concrete structure behavior derived from human observation of, and reasoning with, experimental data. An alternative, discussed in this paper, is to use a computation and knowledge representation paradigm, called neural networks, developed by researchers in connectionism (a subfield of artificial intelligence) to model concrete structure behavior. The main benefits in using a neural-network approach are that all behavior can be represented within a unified environment of a neural network and that the network is built directly from experimental data using the self-organizing capabilities of the neural network, i.e., the network is presented with the experimental data and learns the relationships between stresses and strains. Such a modeling strategy has important implications for modeling the behavior of modern, complex concrete structures. In this paper, the behavior of prestressed concrete frames is modeled with a backpropagation neural network. The preliminary results of using networks to model concrete structure look very promising.