APPLICATION OF ARTIFICIAL NEURAL NETWORK IN PREDICTING THE FLEXURAL STRENGTH OF RIGID PAVEMENT

• Main Authors: Yi-Jiuan Chou, 周義娟
• Other Authors: Yi-Yu Kuo
• Format: Others
• Language: zh-TW
• Published: 2000
• Online Access: http://ndltd.ncl.edu.tw/handle/73756606004931982012

碩士 === 中華大學 === 土木工程學系碩士班 === 88 === ABSTRACT The use of rigid pavement in massive traffic and heavily loaded areas has been proved to have very good effects in engineering performance and maintenance. In this research, 125 sets of concrete beams with a variety of aggregate volume, water-cement ratio, superplasticizer ratio and steel fiber ratio were tested for the data of flexural strength. The objective of this research is to develop an artificial neural network (ANN) learning model for the relationship of steel fiber reinforced concrete (SFRC) and its components. In this work, several different ANN models have been used to investigate the prediction effect of flexural strength. This application shows that artificial neural network can provide an effective way to predict the flexural strength of rigid pavement and it has been proved to be superior to the conventional empirical equations. keywords: artificial neural network, rigid pavement. steel fiber reinforced concrete, flexural strength.