Application of neural network in estimation of air losses in compressed air tunnelling

This paper explores the capabilities of neural networks to predict the air losses from the tunnel in compressed air tunnelling. A neural network is an information processing system, the architecture of which essentially mimics the biological system of the brain. Artificial neural networks are ideally suited to assist engineers to interpret information from in-situ or laboratory measurements in complex problems. A back-propagation neural network has been trained and used to predict the air losses from the tunnel face and the perimeter walls of the Feldmoching tunnel U8 N-8 in Munich, Germany. In this project, compressed air was used to retain the groundwater and stabilize the tunnel face. Shotcrete was used as the temporary lining while the final permanent lining was constructed in free air. The tunnel passed through variable ground conditions ranging from coarse gravel to sand and clay. Grouting, an additional layer of shotcrete and a layer of mortar were occasionally used to control the air losses. The results of the prediction of the air Losses from the tunnel using the neural network have been compared with the field measurements. Comparison of the results show that artificial neural network can be used as an efficient engineering tool in prediction of air losses from tunnels in compressed air tunnelling.