Structural optimization applying neural networks

Abstract

Structures are designed to fulfill requirements established in numerous design codes or standards. After the design and construction phase is completed, only a few - sophisticated or important structures, such as bridges - are monitored. However, these monitoring systems are mostly non permanent and typically measure only one or two parameters, which leads to incomplete or questionable results. The thesis is focused on the development of an "intelligent monitoring system". This is achieved by applying neural network algorithms to analyze several complex parameters. This approach achieves reliable results, because the neural network is capable to detect an unusual condition through a self learning process.

Structures are designed to fulfill requirements established in numerous design codes or standards. After the design and construction phase is completed, only a few - sophisticated or important structures, such as bridges - are monitored. However, these monitoring systems are mostly non permanent and typically measure only one or two parameters, which leads to incomplete or questionable results. The thesis is focused on the development of an "intelligent monitoring system". This is achieved by applying neural network algorithms to analyze several complex parameters. This approach achieves reliable results, because the neural network is capable to detect an unusual condition through a self learning process.