A tunnel gas risk early warning method, system, device and medium

Abstract

The application relates to a tunnel gas risk early warning method, system, device and medium. The method comprises the following steps: acquiring three types of sensing data in a tunnel, and obtaining a unified risk data set by using a data fusion algorithm. A time series analysis algorithm is executed on the data set to analyze the parameter change trend under the dynamic environment of the tunnel, extract a real-time gas accumulation mode, and compare the real-time gas accumulation mode with a preset safety threshold value. If the threshold value is exceeded, a risk alarm is triggered. According to the analysis of the potential influence of the construction behavior on the gas emission according to the real-time gas accumulation mode, the accident risk probability distribution is inferred by using a Bayesian network algorithm combined with historical accident data, and a risk change quantitative index is calculated. The real-time ventilation demand value is determined by combining the index and the real-time environmental variable data of the tunnel, the ventilation parameters are optimized, and the ventilation execution scheme suitable for the current risk state is determined. The method improves the scientificity and accuracy of risk assessment, improves the reliability and efficiency of gas risk early warning, and provides protection for tunnel construction safety.

Description

Technical Field

[0001] This invention belongs to the field of tunnel engineering safety monitoring, and in particular relates to a method, system, equipment and medium for early warning of tunnel gas risks. Background Technology

[0002] During tunnel construction, methane gas, as a flammable, explosive, toxic, and harmful gas, can easily cause explosions, poisoning, and other safety accidents due to its accumulation and outbursts, seriously threatening the lives of construction workers and the progress of the project. Therefore, real-time early warning of tunnel methane risks is one of the core aspects of tunnel engineering safety control.

[0003] Currently, tunnel gas risk early warning mainly adopts the traditional single-parameter monitoring mode, which mostly relies on gas concentration sensors to collect gas concentration data in the tunnel and determines whether to trigger an early warning based on a preset concentration threshold. This method has a single monitoring dimension and ignores the comprehensive impact of various factors such as tunnel geological conditions, ventilation status, and construction behavior on gas risk, resulting in low early warning accuracy and a tendency for false alarms and missed alarms.

[0004] Some improved solutions introduce multi-sensor monitoring, but they only simply overlay and display the data from each sensor without in-depth integration and analysis. This makes it impossible to capture the changing trends of gas parameters in the dynamic environment of the tunnel, hindering early prediction of gas risks and resulting in significant warning lag. Furthermore, most existing early warning methods only achieve a single "monitoring-early warning" process, failing to effectively link warning results with on-site control measures. Even when warning signals are issued, it is difficult to quickly reduce gas risks through control measures.

[0005] Furthermore, existing technologies for assessing gas accident risks largely rely on human experience and lack scientific and quantitative risk inference methods. They cannot accurately reflect the dynamic changes in gas risks and are difficult to adapt to the needs of tunnel gas early warning under different geological conditions and construction conditions, resulting in poor universality and adaptability of early warning methods. Summary of the Invention

[0006] Therefore, it is necessary to provide a tunnel gas risk early warning method, system, equipment, and medium that can improve the scientificity and accuracy of risk assessment, effectively suppress gas accumulation, reduce the probability of gas accidents, and improve the reliability and efficiency of tunnel gas risk early warning, in response to the above-mentioned technical problems.

[0007] Firstly, this application provides a method for early warning of tunnel gas risks, including:

[0008] Data on gas concentration, ventilation status, and geological conditions within the tunnel are acquired by sensor. A data fusion algorithm is then used to integrate the data from each sensor to obtain a unified risk dataset.

[0009] A time series analysis algorithm is performed on a unified risk dataset to analyze the parameter change trend under the dynamic environment of the tunnel and obtain the real-time gas accumulation mode. If the real-time gas accumulation mode exceeds the preset threshold, the risk alarm mechanism is triggered and the current construction behavior data of the tunnel is obtained.

[0010] Based on the analysis of real-time gas accumulation patterns, the potential impact of construction behavior data on gas outbursts is analyzed. A Bayesian network algorithm is used to infer the probability distribution of accident risks based on the potential impacts, and a quantitative index of risk changes is obtained.

[0011] By combining quantitative indicators of risk changes with real-time environmental variable data within the tunnel, the real-time ventilation demand value is determined. Based on the real-time ventilation demand value, ventilation parameters are optimized, and a ventilation implementation plan is determined.

[0012] In one embodiment, the risk alert mechanism is triggered through the following process:

[0013] A time series analysis algorithm was performed on a unified risk dataset to analyze the parameter change trends under the dynamic environment of the tunnel and obtain the real-time model of gas accumulation.

[0014] The real-time model of gas accumulation includes the dynamic change characteristics of gas concentration, the spatiotemporal evolution characteristics of gas accumulation area, and the correlation response characteristics between gas concentration and tunnel dynamic environmental parameters.

[0015] The gas accumulation real-time mode is compared with the preset gas accumulation threshold to obtain the gas accumulation risk assessment result.

[0016] If the gas accumulation risk assessment result is that it exceeds the gas accumulation threshold, the tunnel gas risk alarm mechanism will be triggered.

[0017] While triggering the risk alert mechanism, data on the current construction activities of the tunnel are obtained.

[0018] In one embodiment, based on the analysis of real-time gas accumulation patterns and construction behavior data regarding the potential impact of gas outbursts, a Bayesian network algorithm is used to infer the probability distribution of accident risks based on the potential impacts, yielding quantitative indicators of risk changes, including:

[0019] Drilling depth and drilling speed parameters are extracted from construction behavior data.

[0020] The amount of gas released per unit time is calculated based on the drilling depth and drilling speed parameters.

[0021] Extract the preset gas release threshold corresponding to the current tunnel construction condition in the real-time gas accumulation mode, and compare the gas release amount with the threshold value.

[0022] If the amount of gas released exceeds the preset gas release threshold, a risk warning signal will be triggered.

[0023] By inputting risk warning signals and historical accident data into a Bayesian network algorithm, the probability distribution of accident risks is inferred.

[0024] Based on probability distribution combined with preset quantitative rules and algorithms, a quantitative indicator of risk change is obtained through extrapolation and calculation.

[0025] The quantitative indicators of risk change include the real-time probability value of gas accident occurrence, the quantitative value of gas risk level, the value of gas risk growth rate, the quantitative value of the degree of danger caused by gas accident, and the quantitative value of the scope of risk impact.

[0026] In one embodiment, the amount of gas released per unit time is calculated using the following formula:

[0027]

[0028] in, This indicates the amount of gas released per unit time. This represents the geological correction factor, with a range of values. Based on the current lithology, porosity, and gas occurrence status of the tunnel area, the calibration was performed. This represents the borehole diameter correction factor, with a value range of... To compensate for the deviation between the actual borehole diameter and the nominal borehole diameter. This represents the gas desorption attenuation coefficient, with a value range of... This characterizes the attenuation of gas desorption efficiency with increasing borehole depth. This represents the permeability coefficient of the borehole wall, with a range of values. This reflects the impact of borehole wall integrity on gas escape. Indicates the drilling depth parameter. This represents the drilling speed parameter. This represents the cross-sectional area of ​​the borehole. Indicates the borehole diameter parameter. This indicates the current gas density in the tunnel area.

[0029] In one embodiment, risk change quantification indicators are obtained by performing extrapolation and calculation based on probability distribution combined with preset quantification rules and algorithms, including:

[0030] Extract the probability value of gas accident occurrence corresponding to the current construction conditions of the tunnel from the probability distribution, and use it as the real-time probability value of gas accident occurrence.

[0031] Based on the probability value of the probability distribution, and combined with the preset threshold for classifying gas risk levels, a quantitative value for the gas risk level is obtained.

[0032] By performing time series calculations on the probability distributions at different time points, the slope and growth rate of the probability values ​​are calculated to obtain the gas risk growth rate.

[0033] Based on the probability values ​​of the probability distribution, a weighted calculation is performed using a preset correlation coefficient for the degree of danger to obtain a quantitative value for the degree of danger of a gas accident.

[0034]

[0035] in, This indicates a quantitative value representing the severity of a gas accident. This represents the probability value of a gas accident occurring in a probability distribution. This represents the real-time quantitative value of gas concentration. This represents the quantified risk value of the construction activity. This represents the quantitative value of tunnel environmental risk. , , This represents the preset correlation coefficient for the degree of risk, corresponding to the weighting coefficients of gas concentration, construction behavior, and tunnel environment, respectively.

[0036] Based on the quantitative value of the gas risk level, the probability critical value of the corresponding risk level is determined. The gas diffusion range is extrapolated and calculated by combining the gas diffusion model constructed with tunnel geological parameters and ventilation parameters, and the quantitative value of the risk impact range is obtained.

[0037] The gas diffusion model is constructed using numerical analytical algorithms, fluid dynamics simulation algorithms, or data fitting algorithms.

[0038] By integrating the real-time probability value of gas accidents, the quantitative value of gas risk level, the value of gas risk growth rate, the quantitative value of the degree of danger caused by gas accidents, and the quantitative value of the scope of risk impact, a quantitative indicator of risk change is obtained.

[0039] In one embodiment, the real-time ventilation demand value is determined by combining risk change quantification indicators with real-time environmental variable data within the tunnel. Based on this real-time ventilation demand value, ventilation parameters are optimized to determine the ventilation implementation plan, including:

[0040] The risk prediction report is generated based on quantitative indicators of risk changes; the risk prediction report includes the current risk level data within the tunnel.

[0041] The risk assessment report is transmitted to the tunnel ventilation control system, which then generates a preliminary ventilation parameter adjustment plan based on the risk level data.

[0042] Based on the preliminary ventilation parameter adjustment plan, real-time environmental variable data inside the tunnel are extracted, and the real-time ventilation demand value is determined in combination with the risk level data.

[0043] The real-time ventilation demand value is compared with the preset ventilation demand threshold, and the ventilation parameters in the preliminary ventilation parameter adjustment plan are determined based on the comparison results.

[0044] If the comparison results indicate that the ventilation parameters need to be optimized, then an optimized ventilation control command will be generated based on the real-time ventilation demand value, environmental variable data, and risk level data.

[0045] Based on the optimized ventilation control commands, real-time operating status data of the tunnel ventilation equipment is obtained. Combined with real-time ventilation demand values, equipment operating status data, and gas risk warning information, the final ventilation execution plan is determined.

[0046] The ventilation implementation plan includes ventilation volume, fan speed control parameters, ventilation status feedback indicators, and ventilation control trigger thresholds that match the current gas risk level.

[0047] Secondly, this application also provides a tunnel gas risk early warning system, the system comprising:

[0048] The data fusion module is used to acquire gas concentration sensor data, ventilation status sensor data, and geological condition sensor data in the tunnel. It integrates the sensor data using a data fusion algorithm to obtain a unified risk dataset.

[0049] The early warning analysis module is used to perform time series analysis algorithms on a unified risk dataset to analyze the parameter change trends in the dynamic environment of the tunnel and obtain the real-time gas accumulation mode. If the real-time gas accumulation mode exceeds the preset threshold, the risk alarm mechanism is triggered and the current construction behavior data of the tunnel is obtained.

[0050] The risk inference module is used to analyze the potential impact of construction behavior data on gas outburst based on the real-time gas accumulation pattern. It uses a Bayesian network algorithm to infer the probability distribution of accident risk based on the potential impact and obtain a quantitative index of risk change.

[0051] The ventilation control module is used to determine the real-time ventilation demand value by combining quantitative indicators of risk changes with real-time environmental variable data in the tunnel, optimize ventilation parameters based on the real-time ventilation demand value, and determine the ventilation implementation plan.

[0052] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method described above.

[0053] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the aforementioned method.

[0054] The aforementioned method, system, computer equipment, and storage medium for early warning of tunnel gas risks acquire gas concentration sensor data, ventilation status sensor data, and geological condition sensor data within the tunnel. The gas concentration sensor data reflects the distribution and changes in gas content within the tunnel; the ventilation status sensor data characterizes ventilation information such as fan operating parameters and wind speed and volume within the tunnel; and the geological condition sensor data reflects geological parameters affecting gas occurrence and emission, such as the lithology and porosity of the surrounding rock. Subsequently, a data fusion algorithm is used to perform noise reduction, complementation, and integration processing on the three types of sensor data, eliminating errors and redundancy in single-sensor data to obtain a unified risk dataset covering multiple dimensions of gas, ventilation, and geological information. A time-series analysis algorithm is applied to this unified risk dataset to trace the changes in each monitoring parameter at different time points, analyze the changing trends of each parameter under the dynamic construction environment of the tunnel, and extract a real-time gas accumulation pattern. This real-time gas accumulation pattern is compared with a preset safety threshold. If the threshold is exceeded, a risk alarm mechanism is triggered, and simultaneously, current tunnel construction behavior data, including relevant parameters of drilling, tunneling, and other construction procedures, are acquired. Based on the obtained real-time gas accumulation pattern, the potential impact of current construction activity data on gas emission volume and rate is analyzed to clarify the correlation between construction activities and abnormal gas accumulation. Using a Bayesian network algorithm, the aforementioned potential impacts are used as input variables, combined with prior probabilities calibrated from historical accident data, to infer the probability distribution of tunnel gas accident risk. Based on this probability distribution, quantitative indicators of risk change, such as the real-time probability value of gas accident occurrence and the quantified value of gas risk level, are further calculated. Finally, combining these quantitative indicators of risk change with real-time environmental variable data within the tunnel, the real-time ventilation demand value is determined by comprehensively considering the gas risk level and on-site environmental conditions. Based on this real-time ventilation demand value, the existing parameters of the tunnel ventilation system are adjusted and optimized, ultimately determining a ventilation implementation plan suitable for the current risk state. This method overcomes the limitations of single-parameter monitoring by fusing multi-source sensor data, improving the completeness and accuracy of risk data. It captures dynamic patterns of gas accumulation through time-series analysis algorithms, enabling early prediction of gas risks and effectively reducing warning lag. Furthermore, it utilizes Bayesian network algorithms to quantify accident risks, replacing traditional manual experience-based judgment and enhancing the scientific rigor and accuracy of risk assessment. By linking quantitative risk change indicators with ventilation control, it can dynamically optimize ventilation plans in real time, effectively suppressing gas accumulation and reducing the probability of gas accidents. Overall, this method is adaptable to the dynamic construction environment of tunnels, improving the reliability and efficiency of tunnel gas risk early warning and providing strong protection for tunnel construction safety. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0056] Figure 1 A flowchart of a tunnel gas risk early warning method provided in an embodiment of the present invention;

[0057] Figure 2 This is a structural block diagram of a tunnel gas risk early warning system provided in an embodiment of the present invention. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0059] In one embodiment, such as Figure 1 As shown, this application provides a method for early warning of tunnel gas risks, which may include the following steps:

[0060] Step S101: Obtain gas concentration sensor data, ventilation status sensor data, and geological condition sensor data inside the tunnel. Use a data fusion algorithm to integrate the sensor data to obtain a unified risk dataset.

[0061] The data includes three main categories: gas concentration sensing data, collected by gas concentration sensors deployed within the tunnel, covering real-time monitoring values ​​of gas concentration at different cross-sections and heights; ventilation status sensing data, collected by ventilation system sensors, including parameters such as fan speed, real-time wind speed, air volume, and ventilation pressure within the tunnel; and geological condition sensing data, collected by geological monitoring sensors, including parameters such as tunnel surrounding rock lithology, porosity, rock strata fracturing degree, and basic gas occurrence parameters. A data fusion algorithm is used to perform noise reduction, complementation, and redundancy elimination processing on these three types of sensing data, correcting monitoring errors from individual sensors, and converting multi-dimensional, heterogeneous sensing data into a standardized data format, ultimately forming a unified risk dataset covering the three core dimensions of gas, ventilation, and geology.

[0062] Step S102: Perform time series analysis algorithm on unified risk dataset to analyze parameter change trends in tunnel dynamic environment and obtain real-time gas accumulation mode. If the real-time gas accumulation mode exceeds the preset threshold, trigger risk alarm mechanism and obtain current tunnel construction behavior data.

[0063] Time series analysis algorithms extract trends from the time-series data of various monitoring parameters in a unified risk dataset, capturing the changing patterns of gas concentration, ventilation status, and geological conditions at different time points. This allows for the extraction of a real-time gas accumulation model that reflects the current gas accumulation status of the tunnel. This model includes core information such as gas accumulation rate, accumulation range, and changing trends. A preset threshold is a critical safety value for gas accumulation, calibrated based on tunnel construction safety standards and historical accident data. The real-time gas accumulation model is compared with this preset threshold. If it exceeds the preset threshold, it indicates a potential gas risk in the tunnel, immediately triggering a risk alarm mechanism. Simultaneously, current construction activity data is acquired, including construction parameters related to gas outburst, such as construction procedure type, drilling depth, tunneling speed, and worker positions.

[0064] Step S103: Analyze the potential impact of construction behavior data on gas outburst based on the real-time gas accumulation model, and use a Bayesian network algorithm to infer the probability distribution of accident risk based on the potential impact to obtain a quantitative index of risk change.

[0065] First, by combining the real-time gas accumulation model, the influence of various parameters in the current construction data on gas emission volume and rate is analyzed to clarify the correlation between construction activities and gas accumulation anomalies, and to identify the potential influencing variables of construction activities on gas emission. Then, using these potential influencing variables as input, and combining them with the prior probability of accidents obtained from historical gas accident data, a Bayesian network algorithm is used to infer the posterior probability, resulting in the probability distribution of tunnel gas accident risk. This probability distribution reflects the probability of gas accidents occurring at different risk levels. Based on this probability distribution, further quantitative indicators of risk change, such as the real-time probability value of gas accident occurrence, the quantitative value of gas risk level, and the risk growth rate, are calculated.

[0066] Step S104: Combine the risk change quantification index with the real-time environmental variable data in the tunnel to determine the real-time ventilation demand value, optimize the ventilation parameters based on the real-time ventilation demand value, and determine the ventilation implementation plan.

[0067] The real-time environmental variable data includes supplementary environmental parameters such as real-time temperature, humidity, air pressure, and methane concentration distribution within the tunnel. Combined with the current methane risk level reflected by quantitative risk change indicators, and considering the needs for methane dilution and removal, a real-time ventilation demand value is determined through quantitative calculation. This demand value includes core parameters such as required ventilation volume and velocity. Based on this real-time ventilation demand value, the existing ventilation parameters of the tunnel ventilation system are adjusted and optimized, correcting parameters such as fan speed and vent opening. Finally, a ventilation implementation plan adapted to the current methane risk state is determined. This plan includes ventilation parameter control standards, fan operating modes, and ventilation status feedback requirements, used to suppress methane accumulation and reduce the risk of methane accidents through ventilation control.

[0068] The aforementioned tunnel gas risk early warning method acquires three types of sensor data: gas concentration, ventilation status, and geological conditions within the tunnel. A data fusion algorithm is used to reduce noise, complement, and integrate these data, eliminating errors and redundancy from individual sensors to obtain a unified risk dataset covering multi-dimensional information. A time-series analysis algorithm is applied to this dataset to analyze parameter change trends under the tunnel's dynamic environment, extracting real-time gas accumulation patterns and comparing them with preset safety thresholds. If the threshold is exceeded, a risk alarm is triggered, and current tunnel construction data is simultaneously acquired. Based on the real-time gas accumulation patterns, the potential impact of construction activities on gas outbursts is analyzed. A Bayesian network algorithm, combined with historical accident data, is used to infer the probability distribution of accident risks and calculate a quantitative index of risk change. This index, combined with real-time tunnel environmental variable data, determines the real-time ventilation demand value, optimizes ventilation parameters, and determines a ventilation implementation plan suitable for the current risk state. This method overcomes the limitations of single-parameter monitoring, improves the completeness and accuracy of risk data, enables early prediction of gas risks, reduces warning lag, improves the scientific nature and accuracy of risk assessment by replacing manual experience with quantitative inference, and links ventilation control optimization schemes to suppress gas accumulation, reduce the probability of accidents, adapt to the dynamic construction needs of tunnels, improve the reliability and efficiency of gas risk early warning, and provide a guarantee for tunnel construction safety.

[0069] In one embodiment, the risk alert mechanism can be triggered through the following process:

[0070] Step S201: Perform time series analysis algorithm on the unified risk dataset to analyze the parameter change trend under the dynamic environment of the tunnel and obtain the real-time gas accumulation mode.

[0071] Preferably, the real-time gas accumulation model includes the dynamic change characteristics of gas concentration, the spatiotemporal evolution characteristics of the gas accumulation area, and the correlation response characteristics between gas concentration and tunnel dynamic environmental parameters.

[0072] Among them, the dynamic change characteristics of gas concentration are used to characterize the rate of rise and fall and the range of fluctuation of gas concentration over time; the spatiotemporal evolution characteristics of gas accumulation area are used to characterize the spatial distribution range of gas accumulation and the diffusion and migration patterns over time; and the correlation response characteristics between gas concentration and tunnel dynamic environmental parameters are used to characterize the influence of changes in parameters such as ventilation status and geological conditions on gas concentration.

[0073] Step S202: Compare the real-time gas accumulation mode with the preset gas accumulation threshold to obtain the gas accumulation risk assessment result.

[0074] Step S203: If the gas accumulation risk assessment result is that the gas accumulation threshold is exceeded, the tunnel gas risk alarm mechanism is triggered.

[0075] Step S204: While triggering the risk alarm mechanism, acquire the current construction behavior data of the tunnel.

[0076] Specifically, a time series analysis algorithm is executed on the obtained unified risk dataset. By extracting trends and performing correlation analysis on the time series data of various monitoring parameters such as gas concentration, ventilation status, and geological conditions in the dataset, the changing patterns of each parameter under the dynamic construction environment of the tunnel are clarified, thereby obtaining a real-time model of gas accumulation. This real-time model of gas accumulation specifically includes the dynamic change characteristics of gas concentration, the spatiotemporal evolution characteristics of gas accumulation area, and the correlation response characteristics between gas concentration and tunnel dynamic environmental parameters. Among them, the dynamic change characteristics of gas concentration reflect the rate of rise and fall and the fluctuation range of gas concentration over time; the spatiotemporal evolution characteristics of gas accumulation area reflect the spatial distribution range of gas accumulation and the diffusion and migration patterns over time; and the correlation response characteristics between gas concentration and tunnel dynamic environmental parameters reflect the influence relationship between changes in parameters such as ventilation status and geological conditions on gas concentration. The real-time gas accumulation model is compared with a preset gas accumulation threshold. The preset gas accumulation threshold is based on tunnel construction safety standards, gas occurrence conditions, and historical accident data, and covers critical values ​​for gas concentration, accumulation rate, and accumulation range. The gas accumulation risk assessment result is obtained through multi-dimensional comparison. If the gas accumulation risk assessment result exceeds the preset gas accumulation threshold, it indicates that there is a gas accumulation safety hazard in the tunnel, and the tunnel gas risk alarm mechanism is immediately triggered. At the same time, the current construction behavior data of the tunnel is acquired, including real-time construction parameters related to gas emission, such as construction procedures, drilling parameters, and tunneling speed.

[0077] This embodiment utilizes time series analysis algorithms to deeply process a unified risk dataset, accurately capturing the multi-dimensional characteristics of gas accumulation in a dynamic tunnel environment. It clarifies the dynamic changes and influencing factors of gas accumulation, solving the problem of traditional gas accumulation analysis focusing only on a single parameter and failing to comprehensively reflect the gas accumulation state. By clarifying the specific composition of the real-time gas accumulation mode, the characterization of the gas accumulation state becomes more targeted and comprehensive, providing a clear and quantitative basis for risk assessment. Preset gas accumulation thresholds, combined with multiple safety standards and historical data calibration, improve the accuracy and reliability of gas accumulation risk assessment, effectively reducing false alarms and missed alarms. Simultaneously acquiring construction behavior data while triggering the risk alarm mechanism achieves linkage between risk warning and construction data collection, improving the accuracy of tunnel gas accumulation monitoring and early warning, and providing effective protection for tunnel construction safety.

[0078] In one embodiment, the potential impact of construction behavior data on gas outburst is analyzed based on the real-time gas accumulation pattern. A Bayesian network algorithm is then used to infer the probability distribution of accident risk based on the potential impact, yielding a quantitative index of risk change. This process may include the following steps:

[0079] Step S301: Extract drilling depth parameters and drilling speed parameters from the construction behavior data.

[0080] Step S302: Calculate the amount of gas released per unit time based on the drilling depth parameters and drilling speed parameters.

[0081] Step S303: Extract the preset gas release threshold corresponding to the current construction condition of the tunnel in the real-time gas accumulation mode, and compare the gas release amount with the threshold value.

[0082] Step S304: If the gas release exceeds the preset gas release threshold, a risk warning signal is triggered.

[0083] Step S305: Input the risk warning signal and historical accident data into the Bayesian network algorithm to infer the probability distribution of accident risk.

[0084] Preferably, the probability distribution of accident risk is inferred based on the Bayesian posterior probability formula, and the core calculation formula is as follows:

[0085]

[0086] in, This represents the posterior probability of a gas accident occurring after inputting risk warning signals and historical accident data. Indicates a gas accident risk event. This represents the input characteristic variables (such as gas concentration warning value, ventilation abnormality warning value, etc.). This represents the quantified value of the risk warning signal. This represents the quantitative value of historical accident data characteristics (such as the construction conditions, environmental parameters, equipment status, etc. corresponding to historical accidents). This represents the conditional probability of a gas accident risk event occurring, based on the characteristics of the corresponding risk warning signal and historical accident data. This represents the prior probability of a gas accident risk event occurring (obtained based on historical accident data statistics). This represents the marginal probability of risk warning signals and characteristics of historical accident data.

[0087] Step S306: Based on the probability distribution, combined with preset quantification rules and algorithms, perform extrapolation and calculation to obtain the quantification index of risk change.

[0088] Preferably, the quantitative indicators of risk change include the real-time probability value of a gas accident, the quantitative value of the gas risk level, the value of the gas risk growth rate, the quantitative value of the degree of danger caused by a gas accident, and the quantitative value of the scope of risk impact.

[0089] Specifically, drilling depth and drilling speed parameters are extracted from construction data. Based on these parameters, a quantitative calculation is performed to obtain the gas release per unit time under the current tunnel construction conditions. A preset gas release threshold adapted to the current tunnel construction conditions is extracted from the real-time gas accumulation model. The calculated gas release is compared with this preset threshold. If the gas release exceeds the preset threshold, an abnormal gas outburst is determined, triggering a risk warning signal. This risk warning signal and historical accident data are synchronously input into a Bayesian network algorithm, and a probability distribution of accident risk is obtained through probabilistic inference. Based on this probability distribution, combined with preset quantification rules and corresponding algorithms, a risk change quantification index is finally obtained to characterize the dynamic changes in gas risk.

[0090] This embodiment achieves quantitative calculation of gas release by extracting drilling parameters from construction activities, establishing a direct correlation between construction conditions and gas outburst status, and improving the pertinence of risk assessment. It achieves accurate identification of abnormal gas outbursts through preset threshold comparison, enabling timely triggering of early warning signals and shortening the risk response cycle. The use of a Bayesian network algorithm to fuse early warning signals with historical accident data for probabilistic inference improves the scientific rigor and accuracy of accident risk assessment. The resulting quantitative risk change indicators provide a complete quantitative basis for subsequent gas risk prevention and control and ventilation parameter optimization, effectively enhancing the accuracy and practicality of tunnel gas risk early warning.

[0091] In one embodiment, the amount of gas released per unit time is calculated using the following formula:

[0092]

[0093] in, This indicates the amount of gas released per unit time. This represents the geological correction factor, with a range of values. Based on the current lithology, porosity, and gas occurrence status of the tunnel area, the calibration was performed. This represents the borehole diameter correction factor, with a value range of... To compensate for the deviation between the actual borehole diameter and the nominal borehole diameter. This represents the gas desorption attenuation coefficient, with a value range of... This characterizes the attenuation of gas desorption efficiency with increasing borehole depth. This represents the permeability coefficient of the borehole wall, with a range of values. This reflects the impact of borehole wall integrity on gas escape. Indicates the drilling depth parameter. This represents the drilling speed parameter. This represents the cross-sectional area of ​​the borehole. Indicates the borehole diameter parameter. This indicates the current gas density in the tunnel area.

[0094] In one embodiment, the risk change quantification index is obtained by performing extrapolation and calculation based on probability distribution combined with preset quantification rules and algorithms, which may include the following steps:

[0095] Step S401: Extract the probability value of gas accident occurrence corresponding to the current construction conditions of the tunnel from the probability distribution, and use it as the real-time probability value of gas accident occurrence.

[0096] Step S402: Based on the probability value of the probability distribution, and combined with the preset gas risk level classification threshold, a quantitative value of the gas risk level is obtained.

[0097] Step S403: Perform time series calculations on the probability distributions at different time points, calculate the slope and growth rate of the probability values, and obtain the gas risk growth rate value.

[0098] Step S404: Based on the probability values ​​of the probability distribution, a weighted calculation is performed using a preset correlation coefficient for the degree of danger to obtain a quantitative value for the degree of danger of a gas accident.

[0099]

[0100] in, This indicates a quantitative value representing the severity of a gas accident. This represents the probability value of a gas accident occurring in a probability distribution. This represents the real-time quantitative value of gas concentration. This represents the quantified risk value of the construction activity. This represents the quantitative value of tunnel environmental risk. , , This represents the preset correlation coefficient for the degree of risk, corresponding to the weighting coefficients of gas concentration, construction behavior, and tunnel environment, respectively.

[0101] Step S405: Determine the probability critical value of the corresponding risk level based on the quantified value of the gas risk level, and perform gas diffusion range extrapolation calculations by combining the gas diffusion model constructed with tunnel geological parameters and ventilation parameters to obtain the quantified value of the risk impact range.

[0102] Preferably, the gas diffusion model is constructed using numerical analytical algorithms, fluid dynamics simulation algorithms, or data fitting algorithms.

[0103] Step S406: Integrate the real-time probability value of gas accident occurrence, the quantitative value of gas risk level, the value of gas risk growth rate, the quantitative value of the degree of danger caused by gas accident, and the quantitative value of the scope of risk impact to obtain the quantitative index of risk change.

[0104] Specifically, the probability value of a gas accident under the current construction conditions of the tunnel is extracted from the probability distribution of accident risk and used as the real-time probability value of a gas accident. Based on the probability value of this probability distribution, a quantified value of the gas risk level is obtained by combining it with a preset threshold for classifying gas risk levels. Time-series calculations are performed on the probability distribution of accident risk at different time points, and the gas risk growth rate is obtained by calculating the slope and growth rate of the probability value. Based on the probability value of a gas accident in the probability distribution, a weighted calculation is performed using a preset correlation coefficient of risk level to obtain the quantified value of the risk level of a gas accident. The probability threshold value of the corresponding risk level is determined according to the quantified value of the gas risk level, and the gas diffusion range is extrapolated and calculated using a gas diffusion model constructed from tunnel geological parameters and ventilation parameters to obtain the quantified value of the risk impact range. The gas diffusion model is constructed using numerical analysis algorithms, fluid dynamics simulation algorithms, or data fitting algorithms. The real-time probability value of a gas accident, the quantified value of the gas risk level, the gas risk growth rate value, the quantified value of the risk level of a gas accident, and the quantified value of the risk impact range are integrated to finally obtain a quantified index of risk change.

[0105] This embodiment, through multi-dimensional decomposition and quantification of the probability distribution of accident risks, comprehensively characterizes the dynamic changes of gas risks from five dimensions: real-time probability, risk level, growth rate, degree of danger, and scope of impact, achieving quantification and refinement of gas risk assessment. By analyzing the characteristics of probability value changes through time-series calculations, the development trend of gas risks can be intuitively reflected, improving the foresight of risk prediction. The use of weighted calculation formulas combined with multiple influencing parameters to determine the quantified value of the degree of danger strengthens the scientific rigor and rationality of risk assessment. The gas diffusion model is used to extrapolate the scope of risk impact, achieving precise definition of the spatial distribution of risks. Finally, the integrated multi-dimensional quantitative indicators of risk changes effectively improve the accuracy and practicality of tunnel gas risk assessment.

[0106] In one embodiment, the real-time ventilation demand value is determined by combining risk change quantification indicators with real-time environmental variable data within the tunnel. Based on this real-time ventilation demand value, ventilation parameters are optimized, and a ventilation implementation plan is determined. This may include the following steps:

[0107] Step S501: A risk prediction report is generated based on the quantitative indicators of risk changes; the risk prediction report contains the current risk level data within the tunnel.

[0108] Step S502: The risk prediction report is transmitted to the tunnel ventilation control system, which then generates a preliminary ventilation parameter adjustment plan based on the risk level data.

[0109] Preferably, a risk prediction report generated based on quantitative risk change indicators is transmitted to the tunnel ventilation control system. The core data of the risk prediction report is the current risk level data within the tunnel, and it also includes auxiliary data related to the quantitative risk change indicators. After receiving the risk prediction report, the tunnel ventilation control system uses the risk level data in the report as the core basis, combined with the system's built-in tunnel ventilation basic parameters (including fan rated power, ventilation duct specifications, tunnel cross-sectional dimensions, etc.), and the basic ventilation requirement standards corresponding to different risk levels, to generate a preliminary ventilation parameter adjustment plan. This plan includes preliminary ventilation air volume and fan speed adjustment parameters, providing an initial framework for subsequent precise optimization of ventilation parameters.

[0110] Step S503: Extract real-time environmental variable data inside the tunnel based on the preliminary ventilation parameter adjustment plan, and determine the real-time ventilation demand value by combining the risk level data.

[0111] Based on the control direction and basic requirements defined in the generated preliminary ventilation parameter adjustment plan, real-time environmental variable data within the tunnel are extracted. This real-time environmental variable data is collected by environmental sensors deployed within the tunnel, specifically including parameters directly related to ventilation effectiveness, such as real-time temperature, humidity, air pressure, methane concentration distribution, and airflow rate. After extracting the above real-time environmental variable data, and combining it with the risk level data in the risk assessment report, the minimum ventilation standard for methane dilution and discharge under the current risk level is determined. Through quantitative calculations, the on-site environmental conditions reflected by the environmental variable data are combined with the ventilation requirements corresponding to the risk level. Taking into account the methane diffusion law and the impact of environmental parameters on ventilation effectiveness, a real-time ventilation requirement value that can adapt to the current environmental conditions and risk status is finally determined.

[0112] Step S504: Compare the real-time ventilation demand value with the preset ventilation demand threshold, and determine whether the ventilation parameters in the preliminary ventilation parameter adjustment plan need to be optimized based on the comparison result.

[0113] Step S505: If the comparison results indicate that the ventilation parameters need to be optimized, then an optimized ventilation control instruction is generated based on the real-time ventilation demand value, environmental variable data, and risk level data.

[0114] Step S506: Based on the optimized ventilation control instructions, obtain the real-time operating status data of the tunnel ventilation equipment, and combine the real-time ventilation demand value, equipment operating status data and gas risk warning information to determine the final ventilation execution plan.

[0115] Preferably, the ventilation execution plan includes ventilation air volume, fan speed control parameters, ventilation status feedback indicators, and ventilation control trigger thresholds that match the current gas risk level.

[0116] Specifically, a risk prediction report is generated based on quantitative indicators of risk changes. This report includes current gas risk level data within the tunnel. The report is then transmitted to the tunnel ventilation control system, which uses the risk level data from the report as the core basis, combined with the system's basic operating parameters, to generate a preliminary ventilation parameter adjustment plan. Based on this plan, real-time environmental variable data within the tunnel (including real-time temperature, humidity, air pressure, and gas concentration distribution) is extracted and, combined with the current risk level data, quantitative calculations determine the real-time ventilation demand. This demand is then compared to a preset ventilation demand threshold. The comparison results determine whether the ventilation parameters in the preliminary adjustment plan are suitable for the current risk state and environmental conditions, thereby determining the appropriate ventilation parameters. The system determines whether the ventilation parameters need optimization. If the comparison results indicate that the ventilation parameters need optimization, then based on the real-time ventilation demand value, real-time environmental variable data in the tunnel, and current risk level data, and taking into account gas dilution, exhaust requirements, and environmental adaptability, an optimized ventilation control command is generated. Based on the optimized ventilation control command, real-time operating status data of the tunnel ventilation equipment (including fan speed, air volume output value, equipment failure rate, etc.) is collected and acquired. Combined with the real-time ventilation demand value, real-time operating status data of the ventilation equipment, and gas risk warning information, multi-parameter collaborative analysis is performed to determine the final ventilation execution plan. The ventilation execution plan specifically includes ventilation air volume and fan speed control parameters matched with the current gas risk level, ventilation status feedback identifiers for linkage with risk warnings, and ventilation control trigger thresholds linked with risk warning levels.

[0117] This embodiment generates a risk prediction report containing risk level data through risk change quantification indicators, providing accurate risk basis for ventilation control and achieving deep linkage between risk prediction and ventilation control. By extracting real-time environmental variable data and combining it with risk levels to determine real-time ventilation demand values, the targetedness and accuracy of ventilation demand judgment are ensured. Through an optimization mechanism that compares real-time ventilation demand values ​​with preset thresholds, the problem of ventilation parameters being out of touch with actual needs is avoided, improving the adaptability of ventilation parameters. By integrating ventilation equipment operating status data and risk warning information to determine the final ventilation execution plan, both risk prevention and control and equipment operation stability are taken into account. Moreover, the multi-dimensional parameter settings of the execution plan achieve synergistic linkage between ventilation control and risk warning. This effectively improves the scientificity and accuracy of tunnel ventilation control, enabling dynamic real-time optimization of ventilation plans based on gas risk, suppressing gas accumulation, reducing the risk of gas accidents, and providing strong protection for tunnel construction safety.

[0118] In one embodiment, such as Figure 2 As shown, this application also provides a tunnel gas risk early warning system, which may include:

[0119] The data fusion module 601 is used to acquire gas concentration sensor data, ventilation status sensor data and geological condition sensor data in the tunnel, and to integrate the sensor data using a data fusion algorithm to obtain a unified risk dataset.

[0120] The early warning analysis module 602 is used to perform time series analysis algorithms on a unified risk dataset to analyze the parameter change trends under the dynamic environment of the tunnel and obtain the real-time gas accumulation mode. If the real-time gas accumulation mode exceeds the preset threshold, the risk alarm mechanism is triggered and the current construction behavior data of the tunnel is obtained.

[0121] The risk inference module 603 is used to analyze the potential impact of construction behavior data on gas outburst based on the real-time gas accumulation pattern. It uses a Bayesian network algorithm to infer the probability distribution of accident risk based on the potential impact and obtain a quantitative index of risk change.

[0122] The ventilation control module 604 is used to determine the real-time ventilation demand value by combining the quantitative indicators of risk changes with the real-time environmental variable data in the tunnel, optimize the ventilation parameters based on the real-time ventilation demand value, and determine the ventilation execution plan.

[0123] The aforementioned tunnel gas risk early warning system includes a data fusion module that acquires gas concentration sensor data, ventilation status sensor data, and geological condition sensor data within the tunnel. Gas concentration sensor data reflects the gas content and distribution within the tunnel; ventilation status sensor data characterizes the operating parameters of the ventilation system; and geological condition sensor data reflects the basic characteristics of the tunnel's surrounding rock and gas occurrence. This module employs a data fusion algorithm to reduce noise, complement, and standardize these three types of heterogeneous sensor data, eliminating single-sensor monitoring errors and data redundancy. The final output is a unified risk dataset covering multi-dimensional risk information, providing complete and accurate data support for subsequent early warning analysis. The early warning analysis module receives the unified risk dataset output by the data fusion module, performs time-series analysis on it, traces the temporal changes of each monitoring parameter, analyzes the parameter change trends under the dynamic construction environment of the tunnel, and extracts a real-time gas accumulation pattern that reflects the current gas accumulation state. This module compares the real-time gas accumulation pattern with a preset threshold. If the threshold is exceeded, a tunnel gas risk alarm mechanism is immediately triggered, and simultaneously, current tunnel construction behavior data is acquired, providing construction-related data input for risk inference. The risk inference module takes the real-time gas accumulation pattern and construction behavior data output by the early warning analysis module as input, analyzes the potential impact of construction behavior data on gas outbursts, clarifies the correlation between construction behavior and abnormal gas accumulation, and uses a Bayesian network algorithm to infer the probability distribution of accident risk based on the potential impact and prior probabilities calibrated from historical accident data. Then, through quantitative calculation, it obtains a quantitative index of risk change. The ventilation control module receives the quantitative index of risk change output by the risk inference module, combines it with real-time environmental variable data collected within the tunnel, comprehensively considers the current gas risk level and on-site environmental conditions, and determines the real-time ventilation demand value through quantitative calculation. Based on this real-time ventilation demand value, it optimizes and adjusts the existing parameters of the tunnel ventilation system, and finally determines a ventilation implementation plan suitable for the current risk state, achieving precise control of gas risk.

[0124] This embodiment employs four modules working collaboratively to effectively address the problems of data fragmentation, delayed early warning, inaccurate risk inference, and disconnect between ventilation control and risk status in existing tunnel gas control systems. The data fusion module efficiently integrates multi-source sensor data, improving the completeness and accuracy of risk data; the early warning analysis module captures the dynamic characteristics of gas accumulation through time-series analysis, enabling early risk prediction and reducing false alarms and missed alarms; the risk inference module uses Bayesian network algorithms to quantitatively infer accident risks, improving the scientific rigor and accuracy of risk assessment; and the ventilation control module links risk changes with ventilation control, ensuring that ventilation plans adapt to real-time risk status and effectively suppressing gas accumulation. The clear division of labor and collaborative operation of each module comprehensively improves the accuracy, reliability, and efficiency of tunnel gas risk early warning and control.

[0125] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0126] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of the tunnel gas risk early warning method as described above.

[0127] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0128] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The components described as separate parts may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0129] The above-described embodiments are merely illustrative of several implementation methods of the embodiments of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the embodiments of this application, and these modifications and improvements all fall within the protection scope of the embodiments of this application.

Claims

1. A method for early warning of gas risk in tunnels, characterized in that, The method includes: Data on gas concentration, ventilation status, and geological conditions within the tunnel are acquired by sensor, and a data fusion algorithm is used to integrate the various sensor data to obtain a unified risk dataset. A time series analysis algorithm is executed on the unified risk dataset to analyze the parameter change trend under the dynamic environment of the tunnel and obtain the real-time gas accumulation mode. If the real-time gas accumulation mode exceeds the preset threshold, the risk alarm mechanism is triggered and the current construction behavior data of the tunnel is obtained. Based on the real-time gas accumulation model, the potential impact of the construction behavior data on gas emission is analyzed. A Bayesian network algorithm is used to infer the probability distribution of accident risk based on the potential impact, and a quantitative index of risk change is obtained. By combining the aforementioned risk change quantification indicators with real-time environmental variable data within the tunnel, the real-time ventilation demand value is determined. Based on the real-time ventilation demand value, ventilation parameters are optimized, and a ventilation implementation plan is determined.

2. The method according to claim 1, characterized in that, The risk alert mechanism is triggered through the following process: A time series analysis algorithm is performed on the unified risk dataset to analyze the parameter change trend under the dynamic environment of the tunnel and obtain the real-time gas accumulation mode. The real-time gas accumulation model includes the dynamic change characteristics of gas concentration, the spatiotemporal evolution characteristics of the gas accumulation area, and the correlation response characteristics between gas concentration and tunnel dynamic environmental parameters. The gas accumulation real-time mode is compared with the preset gas accumulation threshold to obtain the gas accumulation risk assessment result. If the gas accumulation risk assessment result exceeds the gas accumulation threshold, the tunnel gas risk alarm mechanism is triggered. While triggering the risk alert mechanism, data on the current construction activities of the tunnel are acquired.

3. The method according to claim 1, characterized in that, The process involves analyzing the potential impact of construction behavior data on gas outbursts based on the real-time gas accumulation model, and using a Bayesian network algorithm to infer the probability distribution of accident risks based on these potential impacts, thereby obtaining quantitative indicators of risk changes, including: Extract drilling depth and drilling speed parameters from the construction behavior data; Based on the drilling depth and drilling speed parameters, the amount of gas released per unit time is calculated. Extract the preset gas release threshold corresponding to the current tunnel construction condition in the real-time gas accumulation mode, and compare the gas release amount with the threshold value. If the gas release exceeds the preset gas release threshold, a risk warning signal is triggered. The risk warning signals and historical accident data are input into a Bayesian network algorithm to infer the probability distribution of accident risks; Based on the probability distribution, combined with preset quantification rules and algorithms, a quantification index of risk change is obtained through extrapolation and calculation. The quantitative indicators of risk change include the real-time probability value of gas accident occurrence, the quantitative value of gas risk level, the value of gas risk growth rate, the quantitative value of the degree of danger caused by gas accident, and the quantitative value of the scope of risk impact.

4. The method according to claim 3, characterized in that, The amount of gas released per unit time is calculated using the following formula: in, This indicates the amount of gas released per unit time. This represents the geological correction factor, with a range of values. Based on the current lithology, porosity, and gas occurrence status of the tunnel area, the calibration was performed. This represents the borehole diameter correction factor, with a value range of... To compensate for the deviation between the actual borehole diameter and the nominal borehole diameter. This represents the gas desorption attenuation coefficient, with a value range of... This characterizes the attenuation of gas desorption efficiency with increasing borehole depth. This represents the permeability coefficient of the borehole wall, with a range of values. This reflects the impact of borehole wall integrity on gas escape. Indicates the drilling depth parameter. This represents the drilling speed parameter. This represents the cross-sectional area of ​​the borehole. Indicates the borehole diameter parameter. This indicates the current gas density in the tunnel area.

5. The method according to claim 3, characterized in that, The process of calculating and deducing risk change quantification indicators based on the probability distribution combined with preset quantification rules and algorithms includes: Extract the probability value of gas accident occurrence corresponding to the current construction conditions of the tunnel from the probability distribution, and use it as the real-time probability value of gas accident occurrence. Based on the probability values ​​of the probability distribution, and combined with the preset gas risk level classification threshold, a quantitative value of the gas risk level is obtained. The probability distribution at different time points is subjected to time series operation to calculate the slope and growth rate of the probability value change, and the gas risk growth rate value is obtained. Based on the probability values ​​of the probability distribution, a weighted calculation is performed using a preset correlation coefficient for the degree of danger to obtain a quantitative value for the degree of danger of a gas accident. in, This indicates a quantitative value representing the severity of a gas accident. This represents the probability value of a gas accident occurring in a probability distribution. This represents the real-time quantitative value of gas concentration. This represents the quantified risk value of the construction activity. This represents the quantitative value of tunnel environmental risk. , , This represents the preset correlation coefficient for the degree of risk, corresponding to the weighting coefficients of gas concentration, construction behavior, and tunnel environment, respectively. Based on the quantified value of the gas risk level, the probability critical value of the corresponding risk level is determined. The gas diffusion range is extrapolated and calculated by combining the gas diffusion model constructed with tunnel geological parameters and ventilation parameters, and the quantified value of the risk impact range is obtained. The gas diffusion model is constructed using numerical analytical algorithms, fluid dynamics simulation algorithms, or data fitting algorithms. The real-time probability value of gas accident occurrence, the quantitative value of gas risk level, the value of gas risk growth rate, the quantitative value of the degree of danger caused by gas accident, and the quantitative value of the scope of risk impact are integrated to obtain the quantitative index of risk change.

6. The method according to claim 1, characterized in that, The process involves combining the risk change quantification index with real-time environmental variable data within the tunnel to determine the real-time ventilation demand value, optimizing ventilation parameters based on the real-time ventilation demand value, and determining the ventilation implementation plan, including: A risk prediction report is generated based on the aforementioned quantitative indicators of risk change; the risk prediction report includes current risk level data within the tunnel. The risk assessment report is transmitted to the tunnel ventilation control system, which then generates a preliminary ventilation parameter adjustment plan based on the risk level data. Based on the preliminary ventilation parameter adjustment plan, real-time environmental variable data inside the tunnel are extracted, and the real-time ventilation demand value is determined in combination with the risk level data. The real-time ventilation demand value is compared with the preset ventilation demand threshold, and the ventilation parameters in the preliminary ventilation parameter adjustment plan are determined based on the comparison results to see if they need to be optimized. If the comparison results determine that the ventilation parameters need to be optimized, then an optimized ventilation control command is generated based on the real-time ventilation demand value, environmental variable data, and risk level data. Based on the optimized ventilation control command, the real-time operating status data of the tunnel ventilation equipment is obtained. Combined with the real-time ventilation demand value, equipment operating status data and gas risk warning information, the final ventilation execution plan is determined. The ventilation execution plan includes ventilation volume, fan speed control parameters, ventilation status feedback indicators, and ventilation control trigger thresholds that match the current gas risk level.

7. A tunnel gas risk early warning system, characterized in that, The system includes: The data fusion module is used to acquire gas concentration sensor data, ventilation status sensor data, and geological condition sensor data in the tunnel, and to integrate the sensor data using a data fusion algorithm to obtain a unified risk dataset. The early warning analysis module is used to perform time series analysis algorithms on the unified risk dataset to analyze the parameter change trend under the dynamic environment of the tunnel and obtain the real-time gas accumulation mode. If the real-time gas accumulation mode exceeds the preset threshold, the risk alarm mechanism is triggered and the current construction behavior data of the tunnel is obtained. The risk inference module is used to analyze the potential impact of the construction behavior data on gas outburst based on the real-time gas accumulation pattern, and to infer the probability distribution of accident risk based on the potential impact using a Bayesian network algorithm to obtain a quantitative index of risk change. The ventilation control module is used to determine the real-time ventilation demand value by combining the risk change quantification index with the real-time environmental variable data in the tunnel, optimize the ventilation parameters based on the real-time ventilation demand value, and determine the ventilation execution plan.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images