Valutazione delle correlazioni tra parametri macchina di TBM-EPB e cedimenti indotti in superficie mediante l'uso di modelli a rete neurale

The aim of the work is to study the application of artificial neural network models to predict surface settlements and to define a hierarchy of TBM drive parameters in correlation with their effect on surface settlement with the scope to improve the advancement of the machine and for risk analysis/management purpose. In particular, a neural network model was created and trained to predict surface settlements giving as inputs the excavation parameters of the machine, the geometry and the geotechnical characteristics of the tunnel. Moreover, it was possible to investigate the correlations between input and output by means of three different methods proposed in literature. This method was applied to the case history of the stretch of Turin Metro (Line 1) from Porta Nuova to Lingotto station.