A multi-information fusion method for identifying water hazard hazards in tunnel advanced geological forecasting

Abstract

The present invention discloses a multi-information fusion method for distinguishing water disaster hazards in tunnel advanced geological prediction, which relates to the technical field of advanced geological prediction. When implementing the advanced geological prediction work, information of various tunnel advanced geological prediction methods can be collected. The information fusion scheme formula is realized based on the dynamic weighted average method, and the weight scores of various information are quantified. The weighted average method is to assign a weight to each parameter or indicator, and then perform weighted average on the values ​​of each parameter or indicator to obtain a comprehensive value. It solves the problems of scattered reference information for distinguishing water disaster hazards, many human interferences, and inaccurate judgments encountered in the work of tunnel advanced geological prediction using means such as geophysical exploration and drilling. It helps to more accurately distinguish the spatial location characteristics of hidden dangers in water-rich areas such as caves and fractures in front of the face, quantify the probability of development of water disaster hazards in front of the tunnel, and provide a reference for tunnel excavation construction to reduce the safety risk hidden dangers of tunnel excavation construction.

Description

Technical Field

[0001] The present invention relates to the technical field of advanced geological prediction, and in particular to a multi-information fusion method for distinguishing hidden dangers of water disasters in tunnel advanced geological prediction. Background Art

[0002] Advanced geological prediction work can detect the spatial location of water disaster hazards in front of the tunnel face and infer and judge the type of hazards.

[0003] Commonly used methods for tunnel advance geological prediction are advance drilling and geophysical methods (including seismic wave, geological radar, transient electromagnetic, and direct current methods). Each type of geophysical method is based on physical differences in geological media properties (such as elasticity, conductivity, and moisture content). By detecting these differences in the physical properties of the geological media ahead, they infer the development of geological hazards ahead. Therefore, each technique has its own scope of application, sensitivity, and ability to identify water hazards. Advance drilling offers the advantages of directness and reliability, providing a direct and intuitive understanding of the geological conditions ahead of tunnel excavation. However, this only provides a partial view and cannot fully assess the geological conditions ahead or the distribution of water bodies. Among geophysical prospecting methods, the seismic wave method is more advantageous in identifying the integrity of rock masses and has a longer detection range, generally reaching 100m, but its ability to identify water body development is poor; the geological radar method has a good effect on identifying geological hazards such as faults, fracture zones, karst cavities, and groundwater in the surrounding rock ahead, but its detection range is short, generally not exceeding 30m, and it is easily interfered with, causing distortion of the results; the transient electromagnetic method has a detection range generally not exceeding 80m, and is sensitive to low-resistance geological bodies such as water bodies in the surrounding rock ahead, but is more susceptible to interference; the direct current method has a relatively strong anti-interference ability and can identify hidden dangers such as fracture zones and water-rich areas, but its detection range is short, generally not exceeding 18m. In short, the advance drilling method is stable and reliable but obtains less information, while the geophysical prospecting method obtains rich information, but the on-site data collection process is easily interfered with. In addition, the geophysical prospecting results are multi-solution, which restricts the accuracy of its results. This leads to the idea of ​​the present invention, which is to use multiple methods for comprehensive analysis, obtain multi-factor information, and integrate and quantify it to improve the accuracy of water disaster hidden danger identification in tunnel advance geological prediction.

[0004] At present, the advanced geological prediction work of tunnels is often carried out simultaneously using multiple methods, mainly geophysical methods supplemented by advanced drilling methods. However, there is no suitable solution to integrate and quantify the various information obtained by these methods. The development of water disaster hazards ahead depends on the technical staff to collect and analyze various information based on their experience to draw conclusions. This means that the accuracy of the geological prediction is largely subject to the amount of information collected and analyzed by the engineering and technical personnel and their personal experience, which will lead to the problem of greater human interference factors in the advanced geological prediction work.

[0005] Therefore, it is urgent to establish a method to collect as much quantitative geological information as possible, use information fusion methods to analyze these various information parameters, calculate the risk coefficient of the development of water disaster hazards ahead, and derive the probability of the development of water disaster hazards ahead based on this risk coefficient, thereby reducing interference from some human factors, improving the accuracy of advanced geological forecasts, and achieving the effect of 1+1>2 of the advanced geological forecast working method. Summary of the Invention

[0006] In order to solve the technical problems of tunnel geological prediction, the present invention provides a multi-information fusion method for identifying water disaster hazards in tunnel geological prediction. The following technical solutions are adopted:

[0007] A multi-information fusion method for identifying water disaster hazards in tunnels based on advanced geological prediction includes the following steps:

[0008] Step 1: Collect information on various tunnel geological prediction methods;

[0009] Step 2: extract key information parameters from the information of various tunnel advanced geological prediction methods. The key information parameters include the value of the intact rock mass, the value of the hidden danger point, the distance between the hidden danger midpoint and the tunnel face, and the maximum detection distance.

[0010] Step 3: When it is determined that the difference in distance between the midpoint of the water hazard potential obtained by any two tunnel advanced geological prediction methods and the tunnel face is less than or equal to a set difference threshold, a distance coefficient is calculated;

[0011] Step 4: For each tunnel advanced geological prediction method, calculate the probability coefficient of Class A hidden danger, the probability coefficient of Class B hidden danger, and the comprehensive abnormality probability coefficient respectively;

[0012] Step 5: Calculate the weight coefficient of each tunnel advanced geological prediction method based on the Class A hidden danger probability coefficient, Class B hidden danger probability coefficient, and distance coefficient of each tunnel advanced geological prediction method;

[0013] Step 6: Using the comprehensive abnormal probability coefficient and weight coefficient of each tunnel's advanced geological prediction method, an information fusion method is implemented based on the dynamic weighted average method to calculate the risk coefficient of water disaster hidden danger development ahead;

[0014] Step 7: Determine the development probability of the water disaster hazard ahead based on the value of the water disaster hazard development risk coefficient ahead.

[0015] By adopting the above technical solution, a variety of tunnel advanced geological prediction method information (advanced drilling information, advanced geological prediction information, and ground geophysical information of the tunnel longitudinal axis, etc.) can be collected when implementing advanced geological prediction work. The information fusion scheme formula is implemented based on the dynamic weighted average method, and the weight scores of various information are quantified. The scores are substituted into the formula to calculate the risk coefficient of the development of water disaster hazards ahead. The information fusion method formula is implemented using the dynamic weighted average method: The weighted average method is to assign a weight to each parameter or indicator, and then perform weighted average on the values ​​of each parameter or indicator to obtain a comprehensive value. The dynamic weighted average method used in this method is to dynamically adjust the weight value based on the weighted average according to the on-site geological conditions information and the information of the implemented advanced geological prediction method;

[0016] It solves the problems of scattered reference information, frequent human interference, and inaccurate judgment in the identification of water disaster hazards encountered in the advanced geological prediction of tunnels using geophysical exploration and drilling. It helps to more accurately identify the spatial location characteristics of hazards in water-rich areas such as caves and faults in front of the heading face, quantify the probability of development of water disaster hazards in front of the tunnel, and provide a reference for tunnel excavation construction, so as to reduce the safety risks of tunnel excavation construction.

[0017] Optionally, the information of various tunnel advance geological prediction methods includes advance drilling information, information of various geophysical prospecting methods, and ground geophysical prospecting information of the tunnel longitudinal axis;

[0018] The information of various tunnel advanced geological prediction methods is as follows:

[0019] Advance drilling method: the complete rock mass section is assigned a value of 1, the water hazard potential section is assigned a value of 0.5, and the distance between the midpoint of the water hazard potential and the tunnel face is , Maximum drilling distance ;

[0020] DC method: average apparent resistivity of intact rock mass , Average resistivity value at water disaster risk areas 2. Distance between the midpoint of the water disaster hazard and the tunnel face , Maximum effective detection distance ;

[0021] Geological radar method: average spectral characteristic value of intact rock mass 、 , average spectrum characteristic value at the flood disaster hazard location 、 , the distance between the midpoint of the water disaster hazard and the tunnel face , Maximum effective detection distance ;

[0022] Transient electromagnetic method: average apparent resistivity of intact rock mass , Average resistivity value at water disaster risk areas 2. Distance between the midpoint of the water disaster hazard and the tunnel face , Maximum effective detection distance ;

[0023] Seismic wave method: average wave velocity of intact rock mass , average wave velocity at water disaster risk areas 2. Distance between the midpoint of the water disaster hazard and the tunnel face , Maximum effective detection distance ;

[0024] The nth advanced geological prediction method, the average wave velocity value of the complete rock mass , average wave velocity at water disaster risk areas 2. Distance between the midpoint of the water disaster hazard and the tunnel face , Maximum effective detection distance ;

[0025] Ground geophysical methods: average parameter values ​​of intact rock mass , average parameter values ​​at water disaster risk areas 2. Distance between the midpoint of the water disaster hazard and the tunnel face , implement the maximum effective detection distance in the advanced detection method .

[0026] Using the above technical solution, assume that n methods are implemented at the same tunnel face mileage during the advanced geological forecasting work, including advance drilling, seismic wave, geological radar, transient electromagnetic, and direct current methods. Furthermore, ground geophysical methods (usually magnetotellurics) are also implemented during the tunnel survey and design phase to obtain the variation characteristics of the physical parameters along the tunnel's longitudinal axis. These n+1 methods generate n+1 sets of geological information corresponding to the detection range ahead of the tunnel face. Based on the numerical distribution characteristics of the geophysical results, information parameters such as the location and magnitude of potential hidden danger points, the magnitude of intact rock mass, the distance of the hidden danger midpoint from the tunnel face, and the maximum detection distance corresponding to each method can be extracted. These information parameters are then set to provide parameters for subsequent calculations.

[0027] Optionally, the difference threshold is set to 20%. When judging the distance between the midpoint of the water hazard hazard obtained by any two tunnel advance geological prediction methods and the tunnel face, When the difference is less than or equal to 20%, the distance coefficient is calculated.

[0028] By adopting the above technical solution, the difference threshold of 20% is an empirical value. If it is too large, it will lead to the fusion of multiple abnormal ranges, making the positioning of geological hazards inaccurate. If it is too small, it will lead to the deterioration of the fusion effect of multiple methods and weaken the probability of the development of each geological hazard.

[0029] Optionally, the distance coefficient is calculated using the following formula:

[0030] ;

[0031] in is the distance coefficient.

[0032] Optional, in step 4

[0033] Assume that the probability coefficient of Class A hidden danger is :

[0034] ;

[0035] Assume that the probability coefficient of Class B hidden danger is :

[0036] ;

[0037] Assume that the comprehensive abnormal probability coefficient :

[0038] , Az and Bz take values ​​of 2 and 3 respectively.

[0039] By employing this technical solution, the Class A hidden danger coefficient reflects the relationship between the abnormal development section and the detection range of the geophysical prospecting method. Given that the resolution and accuracy of geophysical prospecting methods are related to the detection range, the Class A hidden danger coefficient is based on this relationship to assess the likelihood of abnormal hidden danger development.

[0040] The Class B hidden danger coefficient reflects the relationship between the abnormal value and the normal value. It is generally believed that the greater the difference between the abnormal value and the normal value, the greater the probability of the hidden danger developing.

[0041] Optionally, when multiple geophysical methods are used for construction, a given weight coefficient is used to fuse the results of multiple tunnel advance geological prediction methods and calculate the weight coefficient of the corresponding geophysical method. The formula is:

[0042]

[0043] Among them: the independent variable a is the probability coefficient of Class A hidden dangers, the independent variable b is the probability coefficient of Class B hidden dangers, c is the ratio of the distance between the midpoint of the detected water hazard and the tunnel face to the actual maximum detection distance of the corresponding geophysical exploration method, and f(a, b, c) is the weight function.

[0044] Optionally, in step 7, a comprehensive discrimination is performed based on the comprehensive anomaly probability coefficient of a single geophysical prospecting method and the weight coefficient of the corresponding geophysical prospecting method to obtain the risk coefficient Q of the water disaster hazard ahead. The discrimination formula is as follows:

[0045] .

[0046] By adopting the above technical solution, the information fusion method formula is realized by the dynamic weighted average method: the weighted average method is to assign a weight to each parameter or indicator, and then perform weighted average on the values ​​of each parameter or indicator to obtain a comprehensive value.

[0047] It helps to more accurately identify the spatial location characteristics of hidden dangers in water-rich areas such as caves and fractures in front of the tunnel face, quantify the probability of development of water disaster hazards in front of the tunnel, and provide a reference for tunnel excavation construction to reduce safety risks in tunnel excavation construction.

[0048] Optionally, when the midpoint of multiple water hazard hazards is at a distance from the tunnel face When the difference is greater than 20%, it is assumed that there are multiple abnormal points ahead. The average parameter value of the corresponding abnormal position is taken according to the geophysical exploration results. The distance coefficient and comprehensive abnormal probability coefficient of the abnormal points at different positions are calculated. Finally, the water disaster hazard development risk coefficient Q corresponding to the multiple abnormal points is calculated respectively.

[0049] By adopting the above technical solution and following the above steps, multi-source data fusion for tunnel advanced geological prediction based on dynamic weighted average method was realized, and a quantitative index was derived.

[0050] Optionally, the value range of the risk coefficient Q is [0, 1], where a value of 0 represents that there is no water disaster hazard in the detection range ahead, and a value of 1 represents that there is a water disaster hazard in the detection range ahead. When the value is between 0 and 1, it represents that there is a possibility of water disaster hazard developing ahead. The size of the Q value is the possibility of the existence of the hazard.

[0051] In summary, the present invention includes at least one of the following beneficial technical effects:

[0052] The present invention can provide a method for identifying water disaster hazards in tunnel advanced geological forecasting based on multi-information fusion. When implementing the advanced geological forecasting work, information of multiple tunnel advanced geological forecasting methods is collected, and an information fusion scheme formula is implemented based on the dynamic weighted average method. The weight scores of various information are quantified and substituted into the formula for calculation to obtain the development risk coefficient of water disaster hazards ahead.

[0053] The weighted average method assigns a weight to each parameter or indicator, and then performs weighted average on the values ​​of each parameter or indicator to obtain a comprehensive value. The dynamic weighted average method used in this method dynamically adjusts the weight value based on the weighted average according to the on-site geological conditions and the information of the advanced geological prediction method implemented.

[0054] It solves the problems of scattered reference information, frequent human interference, and inaccurate judgment in the identification of water disaster hazards encountered in the advanced geological prediction of tunnels using geophysical exploration and drilling. It helps to more accurately identify the spatial location characteristics of hazards in water-rich areas such as caves and faults in front of the heading face, quantify the probability of development of water disaster hazards in front of the tunnel, and provide a reference for tunnel excavation construction, so as to reduce the safety risks of tunnel excavation construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 This is a flow chart of a multi-information fusion method for identifying hidden dangers of water disasters in tunnels based on advanced geological prediction;

[0056] Figure 2 It is a diagram of the magnetotelluric exploration results and the corresponding advanced geological prediction range in a specific embodiment of the present invention;

[0057] Figure 3 Schematic diagram of the results of advanced geological prediction using transient electromagnetic method in a specific embodiment of the present invention;

[0058] Figure 4 It is a schematic diagram of TSP results in a specific embodiment of the present invention. DETAILED DESCRIPTION

[0059] The present invention will be further described in detail below with reference to the accompanying drawings.

[0060] The embodiment of the present invention discloses a method for distinguishing hidden dangers of water disasters in tunnels through advanced geological prediction based on multi-information fusion.

[0061] Reference Figures 1-4 A multi-information fusion method for identifying water disaster hazards in tunnels based on advanced geological prediction includes the following steps:

[0062] Step 1: Collect information on various tunnel geological prediction methods;

[0063] Step 2: extract key information parameters from the information of various tunnel advanced geological prediction methods. The key information parameters include the value of the intact rock mass, the value of the hidden danger point, the distance between the hidden danger midpoint and the tunnel face, and the maximum detection distance.

[0064] Step 3: When it is determined that the difference in distance between the midpoint of the water hazard potential obtained by any two tunnel advanced geological prediction methods and the tunnel face is less than or equal to a set difference threshold, a distance coefficient is calculated;

[0065] Step 4: For each tunnel advanced geological prediction method, calculate the probability coefficient of Class A hidden danger, the probability coefficient of Class B hidden danger, and the comprehensive abnormality probability coefficient respectively;

[0066] Step 5: Calculate the weight coefficient of each tunnel advanced geological prediction method based on the Class A hidden danger probability coefficient, Class B hidden danger probability coefficient, and distance coefficient of each tunnel advanced geological prediction method;

[0067] Step 6: Using the comprehensive abnormal probability coefficient and weight coefficient of each tunnel's advanced geological prediction method, an information fusion method is implemented based on the dynamic weighted average method to calculate the risk coefficient of water disaster hidden danger development ahead;

[0068] Step 7: Determine the development probability of the water disaster hazard ahead based on the value of the water disaster hazard development risk coefficient ahead.

[0069] When implementing advanced geological forecasting, various tunnel advanced geological forecasting methods (e.g., advanced drilling information, advanced geological forecasting information, and ground geophysical information on the tunnel's longitudinal axis) can be collected. An information fusion scheme formula is implemented based on the dynamic weighted averaging method. The weight scores of various information are quantified and substituted into the formula to calculate the risk coefficient for the development of water hazards ahead. The information fusion formula is implemented using the dynamic weighted averaging method: The weighted averaging method assigns a weight to each parameter or indicator, then performs a weighted average of the values ​​to obtain a composite value. This method uses the dynamic weighted averaging method, based on this weighted average, to dynamically adjust the weight values ​​based on the on-site geological conditions and the implemented advanced geological forecasting method.

[0070] It solves the problems of scattered reference information, frequent human interference, and inaccurate judgment in the identification of water disaster hazards encountered in the advanced geological prediction of tunnels using geophysical exploration and drilling. It helps to more accurately identify the spatial location characteristics of hazards in water-rich areas such as caves and faults in front of the heading face, quantify the probability of development of water disaster hazards in front of the tunnel, and provide a reference for tunnel excavation construction, so as to reduce the safety risks of tunnel excavation construction.

[0071] Information on various tunnel advance geological prediction methods includes advance drilling information, information on various geophysical prospecting methods, and ground geophysical prospecting information on the tunnel longitudinal axis;

[0072] The information of various tunnel advanced geological prediction methods is as follows:

[0073] Advance drilling method: the complete rock mass section is assigned a value of 1, the water hazard potential section is assigned a value of 0.5, and the distance between the midpoint of the water hazard potential and the tunnel face is , Maximum drilling distance ;

[0074] DC method: average apparent resistivity of intact rock mass , Average resistivity value at water disaster risk areas 2. Distance between the midpoint of the water disaster hazard and the tunnel face , Maximum effective detection distance ;

[0075] Geological radar method: average spectral characteristic value of intact rock mass 、 , average spectrum characteristic value at the flood disaster hazard location 、 , the distance between the midpoint of the water disaster hazard and the tunnel face , Maximum effective detection distance ;

[0076] Transient electromagnetic method: average apparent resistivity of intact rock mass , Average resistivity value at water disaster risk areas 2. Distance between the midpoint of the water disaster hazard and the tunnel face , Maximum effective detection distance ;

[0077] Seismic wave method: average wave velocity of intact rock mass , average wave velocity at water disaster risk areas 2. Distance between the midpoint of the water disaster hazard and the tunnel face , Maximum effective detection distance ;

[0078] The nth advanced geological prediction method, the average wave velocity value of the complete rock mass , average wave velocity at water disaster risk areas 2. Distance between the midpoint of the water disaster hazard and the tunnel face , Maximum effective detection distance ;

[0079] Ground geophysical methods: average parameter values ​​of intact rock mass , average parameter values ​​at water disaster risk areas 2. Distance between the midpoint of the water disaster hazard and the tunnel face , implement the maximum effective detection distance in the advanced detection method .

[0080] Assume that n methods are implemented at the same tunnel face mileage during advanced geological forecasting, including advance drilling, seismic wave, geological radar, transient electromagnetic, and direct current methods. Furthermore, ground geophysical exploration (usually magnetotelluric) was also implemented during the tunnel survey and design phase to capture the variations in physical properties along the tunnel's longitudinal axis. These n+1 methods yield n+1 sets of geological information corresponding to the detection range ahead of the tunnel face. Based on the numerical distribution of the geophysical results, information parameters such as the location and magnitude of potential hidden danger points, the magnitude of intact rock mass, the distance of the hidden danger's midpoint from the tunnel face, and the maximum detection distance corresponding to each method can be extracted. These information parameters are then used to provide parameters for subsequent calculations.

[0081] Assuming the difference threshold is 20%, when judging the distance between the midpoint of the water hazard hazard obtained by any two tunnel advanced geological prediction methods and the tunnel face, When the difference is less than or equal to 20%, the distance coefficient is calculated.

[0082] The difference threshold of 20% is an empirical value. If it is too large, it will lead to the fusion of multiple anomaly ranges, making the positioning of geological hazards inaccurate. If it is too small, the fusion effect of multiple methods will be poor, weakening the probability of the development of each geological hazard.

[0083] The distance coefficient is calculated using the following formula:

[0084] ;

[0085] in is the distance coefficient.

[0086] In step 4

[0087] Assume that the probability coefficient of Class A hidden danger is :

[0088] ;

[0089] Assume that the probability coefficient of Class B hidden danger is :

[0090] ;

[0091] Assume that the comprehensive abnormal probability coefficient :

[0092] , Az and Bz take values ​​of 2 and 3 respectively.

[0093] The Class A hidden danger coefficient reflects the relationship between the abnormal development section and the detection range of the geophysical exploration method. Given that the resolution and accuracy of geophysical exploration methods are related to the detection range, the Class A hidden danger coefficient is based on this relationship to assess the possibility of abnormal hidden danger development.

[0094] The Class B hidden danger coefficient reflects the relationship between the abnormal value and the normal value. It is generally believed that the greater the difference between the abnormal value and the normal value, the greater the probability of the hidden danger developing.

[0095] In the case of using multiple geophysical methods for construction, a given weight coefficient is used to integrate the results of multiple tunnel advance geological prediction methods and calculate the weight coefficient of the corresponding geophysical method. The formula is:

[0096] .

[0097] Among them: the independent variable a is the probability coefficient of Class A hidden danger, the independent variable b is the probability coefficient of Class B hidden danger, and c is the ratio of the distance between the midpoint of the detected water disaster hidden danger and the tunnel face to the actual maximum detection distance of the corresponding geophysical exploration method. , f(a, b, c) is the weight function.

[0098] In step 7, a comprehensive judgment is made based on the comprehensive anomaly probability coefficient of a single geophysical prospecting method and the weight coefficient of the corresponding geophysical prospecting method to obtain the risk coefficient Q of the water disaster hazard ahead. The judgment formula is as follows:

[0099] .

[0100] The information fusion method is realized by the dynamic weighted average method: the weighted average method is to assign a weight to each parameter or indicator, and then perform weighted average on the values ​​of each parameter or indicator to obtain a comprehensive value.

[0101] It helps to more accurately identify the spatial location characteristics of hidden dangers in water-rich areas such as caves and fractures in front of the tunnel face, quantify the probability of development of water disaster hazards in front of the tunnel, and provide a reference for tunnel excavation construction to reduce safety risks in tunnel excavation construction.

[0102] When the midpoint of multiple water disaster hazards is far from the tunnel face When the difference is greater than 20%, it is assumed that there are multiple abnormal points ahead. The average parameter value of the corresponding abnormal position is taken according to the geophysical exploration results. The distance coefficient and comprehensive abnormal probability coefficient of the abnormal points at different positions are calculated. Finally, the water disaster hazard development risk coefficient Q corresponding to the multiple abnormal points is calculated respectively.

[0103] Through the above steps, multi-source data fusion for tunnel advanced geological prediction based on dynamic weighted average method was realized, and a quantitative index was obtained.

[0104] The value range of the risk coefficient Q is [0, 1]. A value of 0 means that there is no water disaster risk in the detection range ahead, and a value of 1 means that there is a water disaster risk in the detection range ahead. A value between 0 and 1 means that there is a possibility of water disaster risk ahead. The size of the Q value is the possibility of the risk existing.

[0105] The following uses a specific embodiment to illustrate the implementation principle of the multi-information fusion method for identifying water disaster hazards in tunnel advanced geological forecasting of the present invention:

[0106] Take a limestone tunnel in southwestern Hubei as an example: During the survey and design phase of the tunnel, ground geophysical exploration was carried out using the magnetotelluric method to obtain the apparent resistivity distribution of the longitudinal section of the tunnel axis. At the same time, during the construction phase, the transient electromagnetic method and the TSP advanced seismic wave detection method were implemented to obtain the apparent resistivity distribution information within 80 meters in front of the tunnel face and the longitudinal and transverse wave velocity information within 80 meters in front of the tunnel face, respectively. The specific results are shown in the figure. Figure 2-4 ;

[0107] There are two types of distance values ​​between the midpoint of the hidden danger and the tunnel face: one is the hidden danger discovered by geophysical exploration and the other is the hidden danger discovered by drilling.

[0108] The hidden dangers discovered by geophysical exploration are the abnormal range areas obtained by the detection results, such as the low resistivity abnormal area within the generally high resistivity range, or the low velocity abnormal area within the generally high wave velocity range. These abnormal areas are often the locations where geological hidden dangers develop, and these abnormal areas are mostly a range. The midpoint of the range is the distance value from the midpoint of the hidden danger discovered by geophysical exploration to the tunnel face, such as Figure 2 (a) Length of the midpoint of the segment from the tunnel face, Figure 3 The box on the right is circled in the section with high water hazard risk.

[0109] The hidden dangers discovered during drilling are mainly manifested as abnormal sections such as crushing, water gushing, caves, and mud inclusions within the intact rock mass. The distance from the center point of this section to the tunnel face is the "distance value from the midpoint of the hidden danger discovered during drilling to the tunnel face."

[0110] According to the numerical distribution characteristics of the geophysical exploration results, information parameters such as the numerical value of the possible hidden danger point location, the numerical value of the intact rock mass, the distance between the hidden danger midpoint and the tunnel face, and the maximum detection distance corresponding to each method are extracted. The information parameters are as follows:

[0111] Transient electromagnetic method: The average apparent resistivity of intact rock mass is 345Ω·m, the average resistivity at the water hazard risk point is 149.5Ω·m, the distance between the midpoint of the water hazard risk and the tunnel face is 65m, and the maximum effective detection distance is 80m.

[0112] TSP advanced seismic wave detection method: the average wave velocity of intact rock mass is 5100m / s, the average wave velocity at the flood hazard site is 4600m / s, the distance between the midpoint of the flood hazard site and the tunnel face is 63m, and the maximum effective detection distance is 80m;

[0113] Ground geophysical exploration method (magnetotelluric method): the average parameter value of the intact rock mass is 962Ω·m, the average parameter value at the water hazard risk point is 696Ω·m, the distance between the midpoint of the water hazard risk and the tunnel face is 70m, and the maximum effective detection distance for the advanced detection method is 80m.

[0114] Under this condition, the specific steps of the dynamic weighted average method to realize information fusion are as follows:

[0115] When the distances between the midpoints of the water hazard hazards obtained by various methods and the tunnel face are not much different from each other (less than 10%), the distance coefficient LD is obtained as 66m.

[0116] The anomaly probability coefficients of each method are obtained: for the transient electromagnetic method, the probability coefficient of Class A hidden dangers is YA1=0.93, the probability coefficient of Class B hidden dangers is YB1=0.71, and the comprehensive anomaly probability coefficient is YZ1=0.86; for the TSP advanced seismic wave detection method, the probability coefficient of Class A hidden dangers is YA2=0.81, the probability coefficient of Class B hidden dangers is YB2=0.13, and the comprehensive anomaly probability coefficient is YZ2=0.58; for the ground geophysical exploration method (magnetotelluric method), the probability coefficient of Class A hidden dangers is YA3=0.77, the probability coefficient of Class B hidden dangers is YB3=0.35, and the comprehensive anomaly probability coefficient is YZ3=0.63.

[0117] The weight coefficients of each method are obtained: the weight coefficient for the transient electromagnetic method is k1=0.90, the weight coefficient for the TSP advanced seismic wave detection method is k2=0.92, and the weight coefficient for the ground geophysical method (magnetotelluric method) is k3=0.86.

[0118] Based on the comprehensive anomaly probability coefficient of a single geophysical method and the weight coefficient of the corresponding geophysical method, the risk coefficient for the development of water disaster hazards ahead is Q=0.71.

[0119] The front water disaster risk coefficient was calculated to be 0.71, indicating that there is a high probability of water disaster hazards developing 66 meters in front of the tunnel face.

[0120] The above are all preferred embodiments of the present invention, and are not intended to limit the scope of protection of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A multi-information fusion method for identifying water disaster hazards in tunnels based on advanced geological forecasting, characterized in that: The following steps are involved: Step 1: Collect information on various tunnel geological prediction methods; Step 2: extract key information parameters from the information of various tunnel advanced geological prediction methods. The key information parameters include the value of the intact rock mass, the value of the hidden danger point, the distance between the hidden danger midpoint and the tunnel face, and the maximum detection distance. Step 3: When it is determined that the difference in distance between the midpoint of the water hazard potential obtained by any two tunnel advanced geological prediction methods and the tunnel face is less than or equal to a set difference threshold, a distance coefficient is calculated; Step 4: For each tunnel advanced geological prediction method, calculate the probability coefficient of Class A hidden danger, the probability coefficient of Class B hidden danger, and the comprehensive abnormality probability coefficient respectively; Step 5: Calculate the weight coefficient of each tunnel advanced geological prediction method based on the Class A hidden danger probability coefficient, Class B hidden danger probability coefficient, and distance coefficient of each tunnel advanced geological prediction method; Step 6: Using the comprehensive abnormal probability coefficient and weight coefficient of each tunnel's advanced geological prediction method, an information fusion method is implemented based on the dynamic weighted average method to calculate the risk coefficient of water disaster hidden danger development ahead; Step 7: judging the development probability of the water disaster hidden danger ahead according to the value of the water disaster hidden danger development risk coefficient ahead; The distance coefficient is calculated using the following formula: ； in is the distance coefficient; In step 4 Assume that the probability coefficient of Class A hidden danger is : ； Assume that the probability coefficient of Class B hidden danger is : ； Assume that the comprehensive abnormal probability coefficient : , Az and Bz take values ​​of 2 and 3 respectively.

2. The method for identifying water disaster hazards in tunnels based on advanced geological prediction using multi-information fusion according to claim 1, characterized in that: Information on various tunnel advance geological prediction methods includes advance drilling information, information on various geophysical prospecting methods, and ground geophysical prospecting information on the tunnel longitudinal axis; The information of various tunnel advanced geological prediction methods is as follows: Advance drilling method: the complete rock mass section is assigned a value of 1, the water hazard potential section is assigned a value of 0.5, and the distance between the midpoint of the water hazard potential and the tunnel face is , Maximum drilling distance ; DC method: average apparent resistivity of intact rock mass , Average resistivity value at water disaster risk areas 2. Distance between the midpoint of the water disaster hazard and the tunnel face , Maximum effective detection distance ; Geological radar method: average spectral characteristic value of intact rock mass 、 , average spectrum characteristic value at the flood disaster hazard location 、 , the distance between the midpoint of the water disaster hazard and the tunnel face , Maximum effective detection distance ; Transient electromagnetic method: average apparent resistivity of intact rock mass , Average resistivity value at water disaster risk areas 2. Distance between the midpoint of the water disaster hazard and the tunnel face , Maximum effective detection distance ; Seismic wave method: average wave velocity of intact rock mass , average wave velocity at water disaster risk areas 2. Distance between the midpoint of the water disaster hazard and the tunnel face , Maximum effective detection distance ; The nth advanced geological prediction method, the average wave velocity value of the complete rock mass , average wave velocity at water disaster risk areas 2. Distance between the midpoint of the water disaster hazard and the tunnel face , Maximum effective detection distance ; Ground geophysical methods: average parameter values ​​of intact rock mass , average parameter values ​​at water disaster risk areas 2. Distance between the midpoint of the water disaster hazard and the tunnel face , implement the maximum effective detection distance in the advanced detection method .

3. The multi-information fusion method for identifying water disaster hazards in tunnels based on advanced geological forecasting according to claim 2 is characterized by: Assuming the difference threshold is 20%, when judging the distance between the midpoint of the water hazard hazard obtained by any two tunnel advanced geological prediction methods and the tunnel face, When the difference is less than or equal to 20%, the distance coefficient is calculated.

4. The multi-information fusion method for identifying water disaster hazards in tunnels based on advanced geological forecasting according to claim 3 is characterized by: In the case of using multiple geophysical methods for construction, a given weight coefficient is used to integrate the results of multiple tunnel advance geological prediction methods and calculate the weight coefficient of the corresponding geophysical method. The formula is: ； Among them: the independent variable a is the probability coefficient of Class A hidden dangers, the independent variable b is the probability coefficient of Class B hidden dangers, c is the ratio of the distance between the midpoint of the detected water hazard and the tunnel face to the actual maximum detection distance of the corresponding geophysical exploration method, and f(a, b, c) is the weight function.

5. The method for identifying water disaster hazards in tunnels based on advanced geological forecasting using multi-information fusion according to claim 4 is characterized by: In step 7, a comprehensive judgment is made based on the comprehensive anomaly probability coefficient of a single geophysical prospecting method and the weight coefficient of the corresponding geophysical prospecting method to obtain the risk coefficient Q of the water disaster hazard ahead. The judgment formula is as follows: 。 6. The multi-information fusion method for identifying water disaster hazards in tunnels based on advanced geological forecasting according to claim 5 is characterized by: When the midpoint of multiple water disaster hazards is far from the tunnel face When the difference is greater than 20%, it is assumed that there are multiple abnormal points ahead. The average parameter value of the corresponding abnormal position is taken according to the geophysical exploration results. The distance coefficient and comprehensive abnormal probability coefficient of the abnormal points at different positions are calculated. Finally, the water disaster hazard development risk coefficient Q corresponding to the multiple abnormal points is calculated respectively.

7. The multi-information fusion method for identifying water disaster hazards in tunnels based on advanced geological forecasting according to claim 6 is characterized by: The value range of the risk coefficient Q is [0, 1]. A value of 0 means that there is no water disaster risk in the detection range ahead, and a value of 1 means that there is a water disaster risk in the detection range ahead. A value between 0 and 1 means that there is a possibility of water disaster risk ahead. The size of the Q value is the possibility of the risk existing.

Images