The opening of segment joints is considered critical for the stability analysis of a shield tunnel lining. In order to prevent tunnel disasters, an integrated framework was developed in this study based on the laboratory experiment, real-time monitoring, and statistic theory. As a case study, this presented method was employed in a typical underwater shield tunnel to prevent leakage disasters. The water pressure experiment was used to analyze the importance of joint openings for the tunnel waterproofing. Then, an automatic structural health monitoring system was installed in the study site to have real-time monitoring of segment joint opening and the variation of external load applied to the structure. Based on the monitoring data, a multiple linear regression model was developed to explain the response of joint opening to water pressure and temperature at the important positions of arch crown, spandrel, and hance of the tunnel. The multiple linear regression results were verified to be in agreement with those of the numerical simulation; they denote that (1) joint opening decreased with the rise in temperature, but increased with the rise in water pressure at arch crown and hance, and (2) the segment joint opening decreased with the rise in water pressure at spandrel. As a potential application, the developed model was applied to predict future behaviors of structure, which is vital to prevent disasters and provides a reference to underwater constructions.

Structural health monitoring system plays a vital role in smart management of civil engineering. A lot of efforts have been motivated to improve data quality through mean, median values, or simple interpolation methods, which are low-precision and not fully reflected field conditions due to the neglect of strong spatio-temporal correlations borne by monitoring datasets and the thoughtless for various forms of abnormal conditions. Along this line, this article proposed an integrated framework for data augmentation in structural health monitoring system using machine learning algorithms. As a case study, the monitoring data obtained from structural health monitoring system in the Nanjing Yangtze River Tunnel are selected to make experience. First, the original data are reconstructed based on an improved non-negative matrix factorization model to detect abnormal conditions occurred in different cases. Subsequently, multiple supervised learning methods are introduced to process the abnormal conditions detected by non-negative matrix factorization. The effectiveness of multiple supervised learning methods at different missing ratios is discussed to improve its university. The experimental results indicate that non-negative matrix factorization can recognize different abnormal situations simultaneously. The supervised learning algorithms expressed good effects to impute datasets under different missing rates. Therefore, the presented framework is applied to this case for data augmentation, which is crucial for further analysis and provides an important reference for similar projects.

Real-time monitoring of underwater construction is important for structural stability. To prevent tunnel disasters, an integrated analysis framework for tunnel stability evaluation was proposed using real-time monitoring and statistical analysis. A typical underwater shield tunnel, located in the Yangtze River, China, was selected as a case study. A structural health monitoring system (SHMS) was developed to implement real-time monitoring of the water pressure, temperature, and strain during the service period. An analytical model was developed to investigate the mechanical behaviors of the structure based on the monitoring data obtained from the SHMS. To calibrate the model parameters, numerical simulations were carried out using the finite element method with parameters based on the geological conditions for the study site. Consequently, the results indicate that (1) the numerical results agreed well with the analytical results, which showed that the response of the segment strain varied with position and the maximum strain occurred in the arch crown; (2) the segment strain increased with decreasing water pressure, and the maximum value was $4.118\times {10}^{-7}$ when the water pressure changed by 1 kPa; and (3) the strain variation and temperature were proportional, and the maximum value was $13.492\times {10}^{-6}$ when the temperature changed by 1°C. As a potential application, the proposed method was used to predict future behaviors of the structure, which is crucial for preventing disasters, and provides a reference for underwater construction.