Identification of Surrounding Rock in TBM Excavation with Deep Neural Network

In this paper, based on the measured data of a water diversion project and combined with the existing research on the artificial neural network technology, a deep neural network model is trained to realize the real-time identification of surrounding rock in tunnel boring machine (TBM) excavation. The overall accuracy is above 85%. The result shows that deep learning technology can play a role in TBM geological prediction, and TBM operation can be guided by this method.