Safety Of Tunnel Construction Research Articles

A spatio-temporal forecasting method of fracture distribution using dynamically exposed rock images in tunnel: Methodology and application

Forecasting the geometric characteristics of rock fractures in unexcavated tunnel sections is crucial for construction arrangements. This paper presents a method for predicting fracture distribution near the tunnel face from a spatiotemporal perspective. We Innovatively analogize the tunnel excavation mileage as the time series data. Using fractal geometry and geostatistics, we extract geometric features of peripheral rock fractures and establish a spatiotemporal dataset for dynamic prediction. The SCINet model, a spatial-aware recursive neural network, is employed for deterministic-stochastic dynamic prediction. Numerical simulations are conducted to validate the predictive accuracy of the model for various distributions, including trace length, dip angle, and density. Notably, the Mean Absolute Percentage Error (MAPE) for trace length prediction is remarkably low at 5.49 %, and the Kullback-Leibler (KL) divergence is as low as 4.81 %. The method is further applied to an underground oil storage cavern, China, revealing structural surface information. By incorporating continuously exposed fracture data, prediction accuracy is progressively improved, demonstrating the method's potential as a predictive tool for enhancing tunnel construction safety and efficiency.

Research on safety risk factors of metro shield tunnel construction in China based on social network analysis

PurposeDuring metro construction using the shield method, the construction process's complexity, the construction environment's variability, and other factors can easily lead to tunnel construction accidents. This paper aims to explore the interconnections between risk factors and related accident types, as well as the risk chain formed between risk factors, and to analyze the key risk factors and vulnerabilities in shield tunnel construction through empirical data.Design/methodology/approachBased on the social network analysis theory, the connection of various risk factors in subway shield tunnel construction is explored, and the mechanism of multiple risk factors is studied. Through literature analysis, articles on safety risk factors in metro shield tunnel construction are organized and studied, and the identified safety risk factors can comprehensively reflect the significant risks that need to be concerned in metro shield tunnel construction.FindingsThe results show that a small world characterizes the SNA network of safety risk factors for metro shield tunnel construction: The frequency of association between the five risk factors “unsafe behavior,” “site management,” “safety supervision and inspection,” “safety education system” and “safety protection” is higher than that of other factors. Only a few risks, such as “site management,” “safety supervision and inspection,” and “rapid response capability,” directly lead to accidents. In addition, risk factors such as the “safety education system” and “safety protection” will indirectly cause unsafe behaviors of construction personnel.Research limitations/implicationsDuring construction, the probability of occurrence of risk factors may vary with the construction phase and area and is not considered in this paper. In addition, although this paper identifies, determines and analyzes the risk factors affecting the safety of metro shield tunnel construction, including the importance of each risk factor and the connection between them, more detailed information before and after the accident could not be obtained based on the accident investigation report alone. Therefore, future research can collect the same accident case from more sources to obtain more information.Practical implicationsThe theory of accident causation has been improved at the theoretical level. The identified safety risk factors can comprehensively reflect the significant risks that need to be paid attention to in metro shield tunnel construction. From a practical point of view, the results of the study provide a basis for the rational control of the risk factors in the construction of subway shield tunnels, which can help guide practitioners to do a good job of risk prevention before the construction of metro shield tunnels and reduce the probability of related accidents.Originality/valueThis study expands the application of social network analysis in the field of subway tunnel construction risk, quantitatively analyzes the key risk factors and vulnerabilities in shield method tunnel construction and proposes policy recommendations for future metro tunnel construction safety management.

Enhancing Large-Diameter Tunnel Construction Safety with Robust Optimization and Machine Learning Integrated into BIM

Aim This study aims to enhance safety in large diameter tunnel construction by integrating robust optimization and machine learning (ML) techniques with Building Information Modeling (BIM). By acquiring and preprocessing various datasets, implementing feature engineering, and using algorithms like SVM, decision trees, ANN, and random forests, the study demonstrates the effectiveness of ML models in risk prediction and mitigation, ultimately advancing safety performance in civil engineering projects. Background Large diameter tunnel construction presents significant safety challenges. Traditional methods often fall short of effectively predicting and mitigating risks. This study addresses these gaps by integrating robust optimization and machine learning (ML) approaches with Building Information Modeling (BIM) technology. By acquiring and preprocessing diverse datasets, implementing feature engineering, and employing ML algorithms, the study aims to enhance risk prediction and safety measures in tunnel construction projects. Objective The objective of this study is to improve safety in large diameter tunnel construction by integrating robust optimization and machine learning (ML) techniques with Building Information Modeling (BIM). This involves acquiring and preprocessing diverse datasets, using feature engineering to extract key parameters, and applying ML algorithms like SVM, decision trees, ANN, and random forests to predict and mitigate risks, ultimately enhancing safety performance in civil engineering projects. Methods The study's methods include acquiring and preprocessing various datasets (geological, structural, environmental, operational, historical, and simulation). Feature engineering techniques are used to extract key safety parameters for tunnels. Machine learning algorithms, such as decision trees, support vector machines (SVM), artificial neural networks, and random forests, are employed to analyze the data and predict construction risks. The SVM algorithm, with a 98.76% accuracy, is the most reliable predictor. Results The study found that the Support Vector Machine (SVM) algorithm was the most accurate predictor of risks in large diameter tunnel construction, achieving a 98.76% accuracy rate. Other models, such as decision trees, artificial neural networks, and random forests, also performed well, validating the effectiveness of ML-based solutions for risk assessment and mitigation. These predictive models enable stakeholders to monitor construction, allocate resources, and implement preventative measures effectively. Conclusion The study concludes that integrating machine learning (ML) approaches with Building Information Modeling (BIM) significantly improves safety in large diameter tunnel construction. The Support Vector Machine (SVM) algorithm, with 98.76% accuracy, is the most reliable predictor of risks. Other models, like decision trees, artificial neural networks, and random forests, also perform well, validating ML-based solutions for risk assessment. Adopting these ML approaches enhances safety performance and resource management in civil engineering projects.

Study on Influence Mechanism of Tunnel Construction on Adjacent Pile Foundation and Resilience Assessment

To guarantee the safety of tunnel construction and the continued use of nearby structures, it is crucial to accurately forecast the size and extent of the plastic region that may occur due to tunnel excavation, as well as examine the impact on resilience. In this paper, the influence mechanism of tunnel construction on adjacent pile foundation and resilience assessment is investigated. Firstly, the stratum deformation and stress induced by tunnel construction are derived based on the thin-walled theory considering the influence of tunnel structure stiffness. Moreover, the resilience assessment based on the characteristics of the stratum plastic region is proposed to describe the degree of disturbance caused by tunnel construction to the adjacent pile foundation. Then, a comparison with a numerical simulation is conducted to verify the correctness of the prediction method of the stratum plastic region proposed in this paper. Finally, parameter sensitivity analysis is carried out, which indicates that pile parameters, soil parameters, and different tunnel outline conditions have a great influence on the prediction results. In order to reasonably control the impact of tunnel construction on the surrounding environment, safety control techniques, including advance grouting reinforcement and grouting uplift, need to be carefully designed.

Excavation Method Comparison and Optimization for a Super Large Cross-Section Tunnel

Characterized by long spans, low aspect ratios, and intricate construction sequences, super-large cross-section tunnels present substantial construction risks. Therefore, the selection of the optimal excavation method and construction sequence is crucial for ensuring the safety of tunnel construction and minimizing project costs. This paper takes a super large transverse-section highway tunnel as a case study, employing field monitoring data combined with ABAQUS software to analyze the stress and deformation of surrounding rock and support structures under different excavation methods. The findings reveal that the deformation of surrounding rock and support structures excavated by the Double-Side Drift Method is smaller than those caused by the three-benching seven-step method and the CRD excavation method. Nevertheless, the significant stresses of surrounding rock and support structures are released by the Double-Side Drift Method, leading to potential stress concentrations. Thus, it is necessary to ensure the rapid completion of early support and quick sealing of the tunnel. Furthermore, the sixth process achieves smaller deformation (including arch displacement and surface settlement) of the tunnel, a shorter construction period, and lower economic costs when compared to other construction processes. Consequently, it can obviously be concluded that both the Double-Side Drift Method and the sixth construction process stand out as the most appropriate choices for excavating super large cross-section tunnels. The insights obtained from this study provide theoretical guidance for the design and construction of similar tunnel projects.

Enhanced rock mass rating prediction from tunnel face imagery: A decision-supportive ensemble deep learning approach

Ensuring safety in tunnel construction necessitates a rapid and objective evaluation of exposed tunnel faces for proactive rock mass risk assessment. The diversity of rock types and discontinuities contributes to structural complexity and local variability, introducing subjectivity and uncertainty in traditional evaluations based on the experience of field engineers. In this study, a convolutional neural network (CNN) is employed for regression to enhance the objectivity and consistency of rock mass classification via rock mass rating (RMR). The model interprets the entire tunnel face image holistically by identifying and learning from complex patterns, rather than detecting specific geological features like fractured or discontinuity zones. The network model, comprising feature extraction and regression blocks, utilizes the EfficientNet family for computational efficiency and accuracy. Data augmentation and ensemble learning address prediction variability due to image quality, optimizing the model for accurate RMR predictions and providing error ranges. The RMR predictions using excavation site images closely follow the field-evaluated evolution. A server communication-based mobile application is developed for real-time RMR evaluation, enhancing its practical field applicability. In geology, tunnel, and geotechnical engineering, where decisions rely on extensive experience, our approach demonstrates that deep learning can enhance decision-making by analyzing large accumulated datasets to predict optimal outcomes.

Rockburst prediction based on multi-featured drilling parameters and extreme tree algorithm for full-section excavated tunnel faces

The suddenness, uncertainty, and randomness of rockbursts directly affect the safety of tunnel construction. The prediction of rockbursts is a fundamental aspect of mitigating or even eliminating rockburst hazards. To address the shortcomings of the current rockburst prediction models, which have a limited number of samples and rely on manual test results as the majority of their input features, this paper proposes rockburst prediction models based on multi-featured drilling parameters of rock drilling jumbo. Firstly, four original drilling parameters, namely hammer pressure (Ph), feed pressure (Pf), rotation pressure (Pr), and feed speed (VP), together with the rockburst grades, were collected from 1093 rockburst cases. Then, a feature expansion investigation was performed based on the four original drilling parameters to establish a drilling parameter feature system and a rockburst prediction database containing 42 features. Furthermore, rockburst prediction models based on multi-featured drilling parameters were developed using the extreme tree (ET) algorithm and Bayesian optimization. The models take drilling parameters as input parameters and rockburst grades as output parameters. The effects of Bayesian optimization and the number of drilling parameter features on the model performance were analyzed using the accuracy, precision, recall and value of the prediction set as the model performance evaluation indices. The results show that the Bayesian optimized model with 42 drilling parameter features as inputs performs best, with an accuracy of 91.89%. Finally, the reliability of the models was validated through field tests.

Evaluation of cooling measures performance and thermal-humidity environment for ultra-high geothermal railway tunnel construction by using WBGT

High-geothermal damage is one of the major challenges in high-speed railway tunnel construction. Current studies have shortcomings in the formation mechanism of tunnel environment and the environment evaluation system. This may lead to an over-optimal environmental acceptability evaluation and jeopardize tunnel construction safety. This study established a mathematical model of the tunnel environment formation mechanism, proposed the wet-bulb globe temperature (WBGT) standard (26.49°C) for the tunnel construction, analysed the cooling efficiency of different environmental control measures. The results show that increasing the mechanical air supply volume can reduce tunnel temperature. Fog cannon can reduce the air temperature near the tunnel working face, the cooling range in front of the tunnel working face is 10.8 m. The cooling shield can control the WBGT below the standard. Under the 90°C rock temperature, the environmental acceptability of the tunnel under pure mechanical ventilation is 0%; the single fog cannon and double fog cannon (both cooperate with double-duct ventilation) are 22.85% and 54.29% respectively; the cooling shield with one fog cannon and double duct ventilation is 95.72%. This study improves the tunnel environment evaluation system during the construction stage, and provides design reference for temperature control measures in tunnel construction. Highlights: Tunnel environment under the impact of multi-physical coupling fields was analysed. Thermal radiation is taken into account in the formation of the tunnel environment. A new benchmark based on WBGT for high-geothermal tunnels construction was proposed. A tunnel environment assessment system was proposed. Cooling shield environmental-control measure was proposed for tunnel construction.

Multiscale evaluation of tunnel construction safety risk: a case study of an offshore tunnel construction in Ningbo

Based on the WBS-RBS method, in this study, the risk factors corresponding to the construction risk events of an offshore tunnel foundation pit in Ningbo were identified, and the fuzzy comprehensive evaluation method was used to evaluate the construction safety risk of the project. The system-level risk value was obtained by using the risk event deformation resulting from changes in two factors, namely, “mechanical property of soil” and “stiffness of the envelope structure”, to calculate the new event-level risk value corresponding to the deformation using a finite element numerical model. The findings indicate that the tunnel project has a risk evaluation score of 62.78 and thus falls within the category of high-risk projects. A change in risk factors will alter the likelihood that risk events will occur, which affects the safety risk status of the entire project. When two factors are coupled, a project’s system-level risk can increase dramatically.

Damage evolution of tunnel lining under creep action considering interlayer effect in gently inclined layered surrounding rock

The anisotropy and creep characteristics of layered sandstone mudstone with different structural inclinations have a significant impact on the safety of tunnel construction and operation. we conducted research of the layered sand-mudstone of the Yunding Tunnel on the Dazhou-Chengdu line, based on the mechanical response mechanism of layered surrounding rock and employing a numerical coupling analysis method, established a creep damage mechanical model of layered rock to study the progressive failure characteristics of layered rock and the failure mechanism of tunnel lining. The results indicate that:(1) The response mechanism of layered rock mechanics based on the delamination effect can well reflect the progressive fracture of the layered rock and distinguish the failure mode of layered rock; (2) The established layered rock mechanics contact model can effectively characterize the anisotropic macroscopic mechanical properties of layered sand-mudstone in biaxial test and triaxial creep test; (3) The fracture characteristics of the layered surrounding rock matrix change significantly with the change of the dip angle of the structural plane, the cracks mainly appear in the direction perpendicular to the structural plane. And it can be determined that the critical angle of sliding failure of layered sand mudstone is α1 = 28.77°, α2 = 65.47° (4) The damage evolution of lining can be divided into four states: 1. elastic state 2. injury working state 3. ultimate bearing state 4. damage state. With the increase of the dip angle of the structural plane, the time for the lining to enter the ultimate bearing state is gradually delayed, and the delayed rate shows the characteristics of ’slower-faster-slower’. This research achievement can provide a certain foundation for the analysis of the failure mode and the failure mechanism of tunnel lining.

EVALUATION OF SAFETY MANAGEMENT DURING TUNNELS CONSTRUCTION IN SAUDI ARABIA

Tunnel construction projects are complex engineering projects that pose numerous safety challenges. Accordingly, prioritizing the protection of workers, the public, and the environment is imperative. This study aimed to comprehensively analyze the most significant risks associated with tunnel construction projects. Twenty-nine hazards were identified through site inspections and interviews with engineers and managers engaged in various tunnel projects in Saudi Arabia. These risks were categorized into six groups, namely, technical, geological, environmental, social, human, political, contractual, and cost-related risks. Data were acquired from 98 of 108 respondents completing a questionnaire. The data were statistically analyzed using the SPSS V20 package. The Total Importance Index Impact (TIII) application determined the most crucial risks affecting tunnel construction safety management. The findings demonstrate that risk analysis facilitates project managers in identifying and focusing on principal risks, which supports the implementation and evaluation of safety management plans. By providing practical guidelines, risk analysis enhances the success rate of tunnel construction projects. Implementing measures to mitigate exposure to elevated noise levels, air supply hazards, fires, and oxygen deficiency is anticipated to result in numerous benefits, including improved worker health and performance, reduced time and expenses, enhanced productivity, and an improved reputation among staff and clients.

Evolution Law and Grouting Treatment of Water Inrush in Hydraulic Tunnel Approaching Water-Rich Fault: A Case Study

Tunnel construction adjacent to the fault fracture zone is prone to water inrush disasters, which pose a serious threat to the safety of tunnel construction. To provide theoretical support for the early warning and prevention of water inrush disasters of the tunnel adjacent to the water-rich faults, a numerical analysis based on the three-dimensional discrete element method (DEM) was performed to study the evolution of the displacement and seepage fields of the water-resistant rock mass of a tunnel adjacent to a water-rich fault during the water inrush process by taking the Xianglushan tunnel as the research project. With reference to the obtained results, a grouting reinforcement scheme was developed, and its effectiveness was evaluated. The results indicated that as the tunnel face approached the water-rich fault fracture zone, the effect of water pressure gradually became obvious, and the displacement at the face continuously increased. When the tunnel face was excavated to the position 5 m from the fault, the displacement at the center of the face changed suddenly with a sudden increase in water pressure. The water-resistant rock mass ahead of the center of the face was damaged, and a water inrush disaster occurred in the tunnel. Numerical simulation results demonstrated the feasibility of the grouting reinforcement scheme. The assessment based on the borehole acoustic waves, borehole TV, geological radar detection, and convergence monitoring as well as the excavation results confirmed that the water inrush disasters in the 2# adit of Xianglushan tunnel adjacent to the water-rich fault were effectively prevented and controlled, which can provide a reference for the prevention and treatment of the frequent water inrush disasters in underground projects constructed in the water-rich fault area.

Mechanism and prevention of “Closed Door” collapse in tunnel construction: A case study

The prevention of “closed door” collapse is crucial for tunnel construction safety and progress assurance. To address the challenge of preventing “closed door” collapse, a study was conducted based on a case of tunnel collapse. This tunnel passes through sections of fractured rock with groundwater development. The main research methods are on-site monitoring and mechanical model calculations. The research findings indicate that weak fractured rock, groundwater, initial support quality issues, long trailing distances, and extensive excavation footage all contributed to the “closed door” collapse. On-site monitoring in the fractured rock section showed the tunnel deformation is high sensitivity to construction disturbances. Even without disturbances, tunnel deformation gradually increases. This contributes about 20% of the total deformation, worsening surrounding rock damage and loosening. Additionally, the initial support exhibits cantilever arch mechanical response characteristics due to excavation of the surrounding rock on one side of its lower part. The mechanical model calculation has verified that under a cantilever arch state, the initial support lacks the bearing capacity to withstand loose load. The continuous deformation of the surrounding rock damages the collapsed body, generating loose load. Excavation of the surrounding rock beneath the initial support arch foot changed the structure from a fixed arch to a cantilever arch, ultimately leading to the joint failure of the surrounding rock-support structure system. This is the mechanism behind the collapse. By combining the convergence confinement method, the influence mechanism of factors such as excavation footage and trailing distance on the “closed door” collapse was further elucidated. Finally, according to the stability evaluation of the collapsed cavity, a strategy of “supporting first and then grouting” is adopted for collapse treatment. Drawing from the understanding of the “closed door” collapse mechanism, the following principles for collapse prevention are proposed: “small excavation footage, ensuring the quality of initial support, short trailing distance, quick support, frequent measurement.”

The application of evidence-entropy weight gray incidence theory on the risk assessment of rockburst intensity in the Daxiangling tunnel

The risk assessment of rockburst intensity is significant for tunnel construction safety. First, the depth of the rockburst (X1), the uniaxial compressive strength of the rocks (X2), the brittleness coefficient of the rocks (X3), the stress coefficients of the rocks (X4), and the elastic energy index (X5) are adopted as the evidence body, and their essential certainty and reliability is determined using the entropy-gray correlation theory. Second, the synthetic certainty reliability of other samples is calculated based on the evidence theory. Relatively to the traditional gray extension model, it can improve the predictive accuracy and determine the certainty and reliability of different evidence bodies. The difference of importance between other evidence bodies can be reflected; and an interval scale can be taken into consideration in the evaluation process, so the proposed theory can reasonably predict the grade criterion which is interval form. Conclusion demonstrated that the suggested approach is entirely consistent with the actual investigation. The proposed model not only considers the unreliability or reliability of the problem but also solves some degrees of uncertainty and ambiguity of the datum; it enhances the predictive efficiency and provides a new way and thought for future risk assessment of rockburst intensity.

Determination of Supporting Time of Tunnels in the Xigeda Stratum Based on the Convergence-Confinement Method

The mechanical properties of the surrounding rock of the Xigeda stratum are easily affected by water content. In order to obtain the support characteristics of Xigeda strata, the finite difference method was used to obtain the longitudinal deformation of the surrounding rock at a certain distance from the tunnel excavation face under different water contents. Then, the longitudinal deformation profiles of a Xigeda stratum tunnel were obtained under different water content conditions. The accuracy and applicability of the results were verified through error analysis and comparison with existing research results. Based on the convergence-confinement principle, it is proposed that the best time to apply support is when the displacement increment of the surrounding rock has a sharp increase point. The support construction time under different water content conditions was obtained with the distance from the tunnel excavation face as the control index. The results show that with the increase in water content, the longitudinal deformation profile’s growth trend is steeper near the excavation surface and it is gentler when the distance from the excavation face becomes large. At a water content of 20%, the support should be applied 2.67 m behind the excavation face; at a water content of 25%, the support should be applied 1.46 m behind the excavation face. The result has a certain guiding significance for the safety of tunnel construction in the Xigeda stratum.

Application of Electromagnetic Wave Computerized Tomography in Metro Tunnel Detection in Karst Area

When tunnelling in karst areas, a series of geological hazards caused by karst are often encountered, resulting in huge economic losses as well as human casualties. In order to ensure the safety of tunnel construction, risk assessment and early warning of karst-related geological hazards, it is necessary to use geophysical methods to find out the location, size, shape and filling of karst, and to provide necessary reference information for traffic route planning and tunnel construction planning. Based on theoretical model research, the construction plan of electromagnetic wave CT method was designed for data acquisition in an underground tunnel detection project in a city of Jiangsu province based on existing geological data, and the inverse interpretation of the measured data was carried out in combination with borehole core data to study the feasibility, effectiveness and accuracy of the electromagnetic wave CT method for karst detection in urban underground space.

Limit analysis on roof failure mechanism of tunnels subjected to sequential excavations

Compared to traditional full-face excavation methods, sequential excavation methods (SEM), such as central diaphragm (CD) and sidewall drift (SD) methods, can significantly reduce rock disturbance and minimize the extent of roof failure in tunnel crowns. Therefore, they are particularly suitable for tunnel projects in weak rock formations and large cross-sectional underground chambers. This paper presents a kinematic approach to predict the roof failure mechanism of tunnels excavated using the CD method. The collapse blocks at the top are influenced by excavation steps and initial in-situ stresses, and the direction of their failure velocity field is typically non-vertical, posing challenges for limit analysis applications. To address this, a tilted velocity field for the collapse failure of the tunnel crown in the CD method is constructed based on the initial normal stress distribution along the tunnel’s top contour, obtaining an upper bound solution for tunnel collapse failure. The proposed method is compared with the full-face excavation method, demonstrating the superiority of the CD method over the full-section method. The rationality of the proposed tilted velocity field mechanism is verified through numerical validation using finite difference methods. Building upon the upper bound solution, safety assessment indicators for tunnels excavated using the CD method, namely allowable rock stress release and safety factor, are proposed. Furthermore, we extend this method to analyze the roof failure mechanism in tunnels constructed using the SD method and confirm its effectiveness through comparison with FDM analysis. Theoretical analysis and numerical validation both indicate that sequential excavation methods can significantly reduce the potential collapse range at the top and improve the safety of tunnel construction.

Stress analysis of elliptical tunnels in an orthogonally anisotropic elastic full‐space under non‐uniform in‐situ stress

AbstractNon‐uniformly distributed in‐situ stress around the tunnel is a significant factor inducing various geohazards during tunnel excavation in the rock mass, and the anisotropy of the rock mass would further differentiate the in‐situ stress state around the tunnel. To ensure the safety of tunnel construction and operation, it is necessary to deeply analyze the stress distribution law and mechanical mechanism of the orthogonally anisotropic elastic media the around tunnel. This paper presents a complex variable function method for the elliptical tunnel in an orthogonally anisotropic elastic full‐space under non‐uniform in‐situ stress. To obtain the analytical solution, the affine transform and conformal mapping are applied to consider the orthogonal anisotropy of material and the ellipticity of the tunnel. A numerical series fitting method is introduced to describe the non‐uniformly distributed load, and then the stress solution around tunnel is further obtained. The comparisons with the existing literature, numerical solutions and model tests are provided to verify the correctness of the proposed method. Finally, the convergence study and the numerical analysis on the stress distribution are given graphically.

An integrated system for tunnel construction safety control based on BIM–IoT–PSO

An integrated system for tunnel construction safety control based on BIM–IoT–PSO

Developing a cloud evidence method for dynamic early warning of tunnel construction safety risk in undersea environment

Traditional methods have limitations in achieving precise predictions of risk occurrence at an exact future time and have difficulties transforming between qualitative and quantitative indicators and handling multi-source heterogeneous risk data. This study quantifies and analyzes the multi-source construction safety risks classified into the categories of man, machine, material, method and environment (4M1E), and presents a cloud evidence method that integrates wavelet de-noising algorithm, cloud model, and Dempster-Shafer (D-S) evidence theory. A real-time risk prediction and warning is provided using this method after the fusion of multi-source uncertain information and the transformation between qualitative and quantitative indicators, enabling the timely detection of potential risks for project managers. This method analyzing “uncertainty” with “certainty” is verified by an undersea tunnel construction project. The result shows that this method is effective in early warning risks two days before their actual occurrence, providing reference significance for risk early warning of the tunnel construction project.