Field data-based probabilistic assessment on degradation of deformational performance for shield tunnel in soft clay

Performance assessment of operated metro shield tunnel is critical for safety, functionality and durability of segmental lining structures. Furthermore, aleatory and epistemic uncertainties existed in the external environments and internal materials inevitably cause random structural performance over its designed service life. In view of the situation, this paper presents a probabilistic assessment on the degradation of deformational performance for shield tunnel in Shanghai soft clay by using well-documented field data. Horizontal convergence of inner diameter is selected as the basic performance indicator. The performance metric is reasonably assumed to be a parabolic transformation from horizontal convergence. Then, stochastic analysis is carried out to demonstrate the evolution of probability density function both over long-term period and short-term period. On the basis of field data, a lognormal random process is proposed and fairly well validated to represent the evolution of long-term performance. It is calculated that the probability of tunnel deformational performance less than 95% (i.e., the serviceability limit of deformational performance) after 10-year operation is equal to 0.70, and the performance less than 75% (i.e., the ultimate limit state of deformational performance) after 50-year operation is about 0.036. In addition, the autoregressive integrated moving average (ARIMA) model is applied to gain insights relating to deformational performance evolution that produces relatively accurate short-term forecasts. The monitoring frequency of twice a month is suggested for the accuracy of prediction based on the field data in Shanghai. The proposed time-dependent deformational performance is helpful to determine the necessity and urgency of preventive and control measures.