Search

• EICNN and EFBNN are proposed for long-term settlement prediction of shield tunnels. • EICNN, EFBNN and BPNN are compared with a case study in shanghai. • EFBNN had poor spatial and good temporal generalization. • EICNN has excellent temporal and spatial generalization. • The sources of the performance divergence of BPNN, EFBNN and EICNN were analyzed. In recent years, machine learning methods have been widely used to predict the long-term settlement of tunnels. However, data-driven models for long-term settlement prediction often suffer from poor out-of-distribution generalization. Consequently, data-driven models cannot satisfy the requirements of tunnel engineering. This study aims to develop an innovative machine learning methodology with strong out-of-distribution generalization for long-term settlement prediction of tunnel. To implement the data and experience-driven neural networks, we propose two methods: Empirical Formula Based Neural Network (EFBNN) and Empirical Information Constrained Neural Network (EICNN). EFBNN uses an explicit empirical formula to calculate the predicted value, while EICNN calculates it directly. Both methods employ a neural network, but EFBNN estimates the undetermined parameters of the formula, and EICNN constrains the network parameters and loss function with empirical information as prior information. Based on the 20-year settlement monitoring data of a shield tunnel section in Shanghai, the out-of-distribution generalization of EFBNN and EICNN are compared with BPNN. The results show that not all multi-driven methods could improve the out-of-distribution generalization. EFBNN has better temporal out-of-distribution generalization, but worse spatial out-of-distribution generalization, and is sensitive to the choice of empirical formula. EICNN has significant temporal and spatial out-of-distribution generalization. This method can improve the usability of monitoring data, and summarize a model with out-of-distribution generalization. It is a suitable machine learning method for predicting the long-term settlement of tunnels affected by countless and uncertain factors. [ABSTRACT FROM AUTHOR]

• A framework to conduct seismic vulnerability assessment of tunnels is presented. • Fragility and vulnerability curves for tunnels in soft soil are developed. • The importance of local soil condition in the vulnerability of tunnel is stressed. • Shallow tunnels are found to be more vulnerable as compared to deep tunnels. This paper aims at the vulnerability assessment of circular tunnels in soft soil deposits of the metropolitan subway system of Shanghai, accounting for the effects of soil-structure-interaction, local soil conditions and tunnel burial depth. The study focuses on the seismic vulnerability in the transversal direction. A suite of twelve earthquake records is selected and scaled from 0.1 g to 1.0 g to perform non-linear incremental dynamic analysis (IDA), and thus to assess the tunnel response under increasing levels of ground shaking intensity. A visco-elasto-plastic model with Mohr-Coulomb yield criterion is adopted to capture the non-linear behaviour of soil deposits under the seismic excitation. Fragility curves are constructed as a function of free-field peak ground acceleration (PGA) and peak ground velocity (PGV) considering the main sources of uncertainties, with the damage measure (DM) being defined in terms of acting over capacity bending moment ratio of the tunnel lining. The derived fragility curves are compared with existing empirical and analytical ones. Furthermore, vulnerability curves are generated from the derived fragility curves. The significant role of local soil conditions and tunnel burial depth in the vulnerability of tunnels in soft and deep alluvial deposits is highlighted. The proposed fragility and vulnerability functions may be particularly used for the risk analysis of circular tunnels in soft deposits (V s30 < 200 m/s) exposed to seismically-induced transient ground deformations in the transverse direction. [ABSTRACT FROM AUTHOR]