| With the development of China's transportation industry,more and more tunnel projects will pass through the gas geological area,resulting in huge safety hazards in tunnel construction.Gas safety accident is one of the major disasters encountered in the process of tunnel construction,which poses a great threat to the health and safety of construction personnel and property.After excavating the working face of the drilling and blasting construction tunnel,the gas will be released quickly.The distribution and concentration of gas under the tunnel ventilation conditions are the key factors affecting the risk of gas disasters in the tunnel.Analysis of the gas diffusion and migration law,gas distribution characteristics and prediction of gas concentration in the tunnel are the key steps in studying the gas concentration in the tunnel.How to pre-warn the risk of gas tunnel construction according to the gas concentration of the tunnel is an important component to ensure the safe construction of the gas tunnel.This article relies on the Like Tunnel of the New Chengdu-Kunming Railway to conduct research.Based on extensive collection of predecessor research data and results,a detailed investigation and study of engineering construction conditions and on-site inspection and collection of a large amount of gas concentration data.The method of combining theoretical analysis,numerical simulation and nonlinear analysis is used to analyze the characteristics of gas concentration data in tunnels,the law of gas diffusion and transport,and the characteristics of gas space-time distribution.On this basis,the research of gas concentration prediction and gas risk early warning during tunnel construction is carried out.The main research results of the paper are as follows:(1)Based on the gas concentration data collected on the site of the Like Tunnel of the New Chengdu-Kunming Railway,the characteristics analysis of the gas concentration data was carried out.It was determined that the gas concentration in the vault and floor of the tunnel is relatively large,and it is concluded that the general trend of gas concentration change is decreasing,but there will be periodic rises and local peaks.From the perspective of influencing factors of gas concentration data changes,it is determined that the most direct cause and the most important cause of exceeding the gas concentration value are natural factors and human factors respectively.(2)Combined with the gas tunnel ventilation technology and fluid mechanics knowledge,the microscopic analysis of the gas diffusion and transport law in the tunnel is analyzed.The ventilation mode and air demand calculation of the gas tunnel are summarized and analyzed,and the characteristics of the wind flow in the tunnel are studied.It is determined that the process of gas diffusion and migration in the tunnel is mainly divided into four forms.Based on the study of gas migration caused by ventilation and gas turbulence diffusion,the equation of gas diffusion and migration in the tunnel is derived and established,which proves that the gas diffusion and migration in the tunnel has a spatial and temporal correlation.(3)The numerical simulation method is used to study the spatial and temporal distribution characteristics of tunnel gas.Based on the Fluent simulation software,a numerical calculation and analysis model for gas tunnel excavation under ventilation conditions is established.A point source of gas emission was set on the palm surface after the excavation of the upper steps of the tunnel.The spatial distribution and time-varying characteristics of gas concentration in the case of conventional and unconventional gas emission were studied.It macroscopically reflects the law of gas diffusion and migration in the tunnel.Combined with the numerical simulation results and on-site gas concentration data,the automatic gas monitoring and manual detection of the location of the distribution point were optimized.(4)The artificial neural network is used to dynamically predict the gas concentration in the tunnel.The analysis of the dynamic prediction of gas concentration after tunnel excavation proves the necessity of intelligent prediction of gas concentration dynamics from the gas concentration sequence data.Elman neural network is used to dynamically predict the gas concentration at the larger level of the vault and the right side of the two excavation faces of the tunnel.Comparing the predicted data with the real data,it can be found that except for the individual prediction values,there is a large error In addition to the problem,other predicted values are closer to the true value.It is concluded that the intelligent prediction method of tunnel gas concentration dynamics based on Elman neural network has high accuracy and can be successfully applied to the dynamic prediction of gas concentration after excavation of the gas tunnels.(5)Using the risk early warning theory,on-site gas data and gas concentration prediction values,a gas concentration risk early warning model based on monitoring data and a tunnel construction risk early warning model based on gas concentration were established and applied to actual projects.Based on the gas concentration monitoring data and the gas concentration prediction value,the calculation methods and grade division methods of the five gas concentration change indicators are given,and the comprehensive warning level judgment method is determined in combination with the gas concentration level.The principal component analysis method was used to extract and analyze the influencing factors of gas tunnel construction and 12 early warning indicators were determined.According to the early warning indicators and the hierarchical relationship between the indicators,a Bayesian network structure of gas tunnel construction risk warning was established.Combined with fuzzy set theory and expert evaluation language,the tunnel construction risk early warning value is 67.6%.Through Bayesian reverse reasoning,the most probable hazard chain of gas tunnel construction risk is determined. |
PDF Full Text Request