An Investigation on size effect in fracture of cement based materials ( Non-Linear fracture mechanics ) through artificial neural networks

IV SUMMARY Masters Thesis AN INVESTIGATION ON SIZE EFFECT IN FRACTURE OF CEMENT BASED MATERIALS (NON-LINEAR FRACTURE MECHANICS) THROUGH ARTIFICIAL NEURAL NETWORKS RagıpİNCE Fırat University Graduate School of Natural and Applied Sciences Department of Civil Engineering 1994, Page:96 Computation on civil engineering has concentrated primarily on artificial intelligence (AI) applications in the past few years. These application generally involve expert systems. In this study, a different AI approach involving neural network were investigated and an application were presented. In this study, the experimental size effect theory in fracture mechanics of cementitous materials were modelled through artificial neural networks. At the beginning, an artificial neural network package (NETİCE, Neural nETwork In Civil Engineering) for general purpose were developed. A wide range of literature survey were undertaken, and the results of experimental works involving basic modes of fracture mechanics were gathered. These results were taken as input datas, and NETİCE were interactively used on an PC. The results of NETİCE were compared with the experimental results and somewhat positive and meaningful results were provided. It has been shown that the neural network computation is an alternative solution for decision and design of civil engineering problems. Artificial neural networks, is a quite suitable solution tecnique in the problems tjat has more than one solution methods and the problems that defined through ampirical expressions. Because of most of civil engineering problems involves such properties, the application of neural networks on this particular area shows a great application potential KEYWORDS: Artificial Intelligence, Neural Network, Concrete, Fracture Mechanic, Size Effect Theory.