A tunnel geological parameter inference method based on multi-stage data Bayesian inference

Abstract

The present invention relates to the technical field of tunnel surrounding rock detection, and discloses a tunnel geological parameter inference method based on multi-stage data Bayesian inference, comprising the following steps: step S1, collecting unexcavated TSP data in front of the tunnel face and excavated geological data behind the tunnel face; step S2, dividing the collected data into a data set; step S3, performing conditional probability calculation on the data set to obtain conditional probability; step S4, performing prior probability calculation on the data set using a Bayesian inference model to obtain a prior probability; step S5, performing a first posterior probability calculation on the data set using a Bayesian inference model to obtain a first posterior probability; step S6, using the first posterior probability as a prior probability. The present invention realizes the inference of four indicators of rock hardness, groundwater conditions, integrity and ground stress of the geological data of the unexcavated part, and provides data support for tunnel geological construction.

Description

Technical Field

[0001] The present invention relates to the technical field of tunnel surrounding rock detection, and in particular to a tunnel geological parameter inference method based on multi-stage data Bayesian inference. Background Art

[0002] During the tunnel construction process, the tunnel engineering disasters caused often have a great impact on the tunnel construction, which may slow down the progress of tunnel construction or endanger the safety of construction workers. Therefore, it is very important to predict and identify tunnel geological disasters in advance. This method mainly focuses on the inference of geological parameters of the unexcavated area in front of the tunnel face during tunnel construction and the prediction and identification function of rock burst disasters.

[0003] Among them, rockburst refers to a disastrous phenomenon in which rocks suddenly release energy due to the concentration of ground stress during underground engineering and mining. Its main characteristics include the sudden fracture and fragmentation of rocks, the generation of shock waves, and the sudden eruption and collapse of rocks or ore bodies. The main reason for its occurrence is high ground stress. Under high ground stress and hard rock environment, the stress of the surrounding rock is redistributed after the disturbance of tunnel excavation. When the stress borne by the surrounding rock reaches the limit, it may cause rockburst disasters. Therefore, it can be seen that the main influencing factors of rockburst disasters include the ground stress size of the tunnel area and the rock mass strength of the surrounding rock. Among them, the rock mass strength of the surrounding rock depends on the rock hardness of the rock contained in the surrounding rock itself, the groundwater situation of the surrounding rock, and the integrity of the surrounding rock. Therefore, it is necessary to detect the geological parameters of the surrounding rock of the construction. The TSP method is generally used for data detection. However, the detection length of the TSP method is limited, and it cannot quickly detect the geological data of the entire surrounding rock. For this reason, a tunnel geological parameter inference method based on multi-stage data Bayesian inference is proposed. Summary of the invention

[0004] In order to solve the technical problems existing in the prior art, the present invention provides a tunnel geological parameter inference method based on multi-stage data Bayesian inference.

[0005] The present invention is implemented by the following technical solution: a tunnel geological parameter inference method based on multi-stage data Bayesian inference, comprising the following steps:

[0006] Step S1, collecting unexcavated TSP data in front of the tunnel face and excavated geological data behind the tunnel face;

[0007] Step S2, dividing the collected data into data sets;

[0008] Step S3, performing conditional probability calculation on the data set to obtain conditional probability;

[0009] Step S4, using the Bayesian inference model to calculate the prior probability of the data set to obtain the prior probability;

[0010] Step S5, using the Bayesian inference model to perform a first posterior probability calculation on the data set to obtain a first posterior probability;

[0011] Step S6, taking the first posterior probability as the prior probability, and inputting the data set into the Bayesian inference model to perform the second posterior probability;

[0012] Step S7, repeating steps S5 and S6 to perform multiple calculations using the Bayesian inference model data set;

[0013] Step S8: Select the state with the highest probability as the inference result according to the content calculated in step S7.

[0014] As a further improvement of the above scheme, the unexcavated TSP data in front of the tunnel face in step S1 include longitudinal wave velocity, shear wave velocity, Poisson's ratio, density, static Young's modulus, dynamic Young's modulus and direct wave velocity; the excavated geological data behind the tunnel face include rock hardness, groundwater conditions, integrity and ground stress.

[0015] As a further improvement of the above solution, in step S2, the data set is divided by combining the excavated geological data parameters behind the tunnel face with the unexcavated TSP data in front of the tunnel face;

[0016] Among them, the rock hardness, groundwater conditions, integrity and ground stress in the excavated geological data parameters behind the tunnel face are used as references, and the data sets are divided into three groups: D1, D2 and D3. The selection of the types of unexcavated TSP data in front of the tunnel face corresponding to the excavated geological data parameter indicators behind the tunnel face is as follows:

[0017] Rock hardness: D1: dynamic Young's modulus, D2: longitudinal wave velocity, D3: density;

[0018] Groundwater conditions: D1: transverse wave velocity, D2: longitudinal wave velocity, D3: direct wave velocity;

[0019] Completeness: D1: longitudinal wave velocity, D2: Poisson's ratio, D3: static Young's modulus;

[0020] Geostress: D1: direct wave velocity, D2: Poisson's ratio, D3: density.

[0021] As a further improvement of the above solution, step S3 includes the following steps:

[0022] Step A1: Frequency distribution:

[0023] Determine the frequency of occurrence of the excavated geological data parameters behind the tunnel face and the corresponding unexcavated TSP data in front of the tunnel face;

[0024] Step A2: Proportional conversion:

[0025] Convert the joint frequencies to probabilities to get the conditional probabilities.

[0026] As a further improvement of the above scheme, in step S4, the prior probability calculation of the geological data parameter indicators that have been excavated behind the tunnel face is implemented based on the geological data parameter indicators that have been excavated behind the tunnel face collected in the data sets D1, D2 and D3, so as to obtain the prior probability.

[0027] As a further improvement of the above solution, in step S5, the data set D1 is initially updated using a Bayesian inference method to obtain the first posterior probability data.

[0028] As a further improvement of the above solution, in step S6, a second posterior calculation is performed on the data set D2 using the first posterior probability as the prior probability, thereby obtaining second posterior probability data.

[0029] As a further improvement of the above solution, the unexcavated TSP data in front of the tunnel face in step S1 is collected using the TSP method.

[0030] As a further improvement of the above solution, the Bayesian inference model includes:

[0031]

[0032] in, : Posterior probability, the probability that the parameter θ is the true value after given the data D;

[0033] : Likelihood function, which indicates the probability of data D appearing when parameter θ is true;

[0034] : Prior probability refers to the original belief about parameter θ before observing the current data;

[0035] : The marginal likelihood, also called the normalizing constant or evidence factor, is the overall probability of the observation D occurring.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] After completing the statistics of the prior probability and the conditional probability, the present invention uses the method of using the posterior data as the prior data of the next step, and each time the observed D1, D2, and D3 states are input, the prior probability is combined with the influence of the three physical states to generate the posterior probability, and then the generated posterior probability is used as the prior probability, and the observed D1, D2, and D3 states are input multiple times to update the prior probability multiple times, thereby realizing the dynamic update process of the prior probability combined with the influence of the observed state; and according to the distribution of the posterior probability that is updated each time, the state with the highest probability in the rock hardness state is selected as the inference result;

[0038] The geological data of the excavated part of the tunnel is processed as a priori, that is, a general judgment of the rear of the tunnel. Then, TSP advanced geophysical exploration is used to collect TSP data behind the heading face, and the collected TSP data is input into the inference model as observation data. The input observation data will affect the priori according to the conditional probability relationship between TSP data and geological data to obtain the posterior, and the posterior is the inference result of the geological data of the unexcavated section corresponding to the input TSP data; then the updated posterior is used as the priori, and the subsequent TSP data is continuously input, so as to continuously update the posterior and obtain the geological data inference result, realizing the calculation of four indicators of rock hardness, groundwater conditions, integrity and ground stress of the geological data of the unexcavated part, providing data support for tunnel geological construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 A schematic diagram of the principle of a tunnel geological parameter inference method based on multi-stage data Bayesian inference provided by the present invention;

[0040] Figure 2 A relationship diagram of the indoor strength results of the surrounding rock of the tunnel face in Example 6 provided by the present invention and the rebound hammer measurement value converted according to the empirical formula;

[0041] Figure 3 A schematic flow chart of a tunnel geological parameter inference method based on multi-stage data Bayesian inference provided by the present invention. DETAILED DESCRIPTION

[0042] The present invention is further described below in conjunction with the accompanying drawings and specific implementation methods. It should be noted that, under the premise of no conflict, the various embodiments or technical features described below can be arbitrarily combined to form a new embodiment.

[0043] Example 1

[0044] Please combine Figure 1 and Figure 3 , a tunnel geological parameter inference method based on multi-stage data Bayesian inference in this embodiment includes the following steps:

[0045] Step S1, collecting unexcavated TSP data in front of the tunnel face and excavated geological data behind the tunnel face;

[0046] Step S2, dividing the collected data into data sets;

[0047] Step S3, performing conditional probability calculation on the data set to obtain conditional probability;

[0048] Step S4, using the Bayesian inference model to calculate the prior probability of the data set to obtain the prior probability;

[0049] Step S5, using the Bayesian inference model to perform a first posterior probability calculation on the data set to obtain a first posterior probability;

[0050] Step S6, taking the first posterior probability as the prior probability, and inputting the data set into the Bayesian inference model to perform the second posterior probability;

[0051] Step S7, repeating steps S5 and S6 to perform multiple calculations using the Bayesian inference model data set;

[0052] Step S8: Select the state with the highest probability as the inference result according to the content calculated in step S7.

[0053] Example 2

[0054] Collect the unexcavated TSP data in front of the tunnel face and the excavated geological data behind the tunnel face. The unexcavated TSP data in front of the tunnel face include longitudinal wave velocity, shear wave velocity, Poisson's ratio, density, static Young's modulus, dynamic Young's modulus and direct wave velocity; the excavated geological data behind the tunnel face include rock hardness, groundwater conditions, integrity and ground stress;

[0055] The basic principle of TSP method, i.e. earthquake prediction or advanced geological prediction ahead of the tunnel, is as follows:

[0056] Excitation holes are arranged within a certain range of the side wall near the tunnel face. Seismic waves are artificially excited in the holes, and the generated seismic waves propagate in the tunnel surrounding rock in the form of spherical waves. When the wave impedance of the surrounding rock changes (for example, when encountering karst, fault or rock layer interface), part of the seismic waves will be reflected back, and the other part will continue to propagate forward; the reflected seismic waves are received by high-precision receivers and transmitted to the host to form seismic wave records. The returned seismic wave records can be used to calculate the longitudinal wave velocity, shear wave velocity, direct wave velocity, etc. within the coverage of the TSP advanced geological forecast; based on the returned seismic wave information, the geological conditions in front of the tunnel face can be judged.

[0057] Example 3

[0058] The data set is divided based on the excavated geological data parameters behind the tunnel face and the unexcavated TSP data in front of the tunnel face;

[0059] Among them, the rock hardness, groundwater conditions, integrity and ground stress in the excavated geological data parameters behind the tunnel face are used as references, and the data sets are divided into three groups: D1, D2 and D3. The selection of the types of unexcavated TSP data in front of the tunnel face corresponding to the excavated geological data parameter indicators behind the tunnel face is as follows:

[0060] Rock hardness: D1: dynamic Young's modulus, D2: longitudinal wave velocity, D3: density;

[0061] Groundwater conditions: D1: transverse wave velocity, D2: longitudinal wave velocity, D3: direct wave velocity;

[0062] Completeness: D1: longitudinal wave velocity, D2: Poisson's ratio, D3: static Young's modulus;

[0063] Geostress: D1: direct wave velocity, D2: Poisson's ratio, D3: density.

[0064] Example 4

[0065] Conditional probability statistics (taking the conditional probability of calculating rock hardness as an example):

[0066] Regarding the conditional probability (likelihood probability) of rock hardness, three types of unexcavated TSP data in front of the tunnel face related to rock hardness are considered, including the dynamic Young's modulus, longitudinal wave velocity, and Poisson's ratio of the unexcavated TSP data in front of the tunnel face. When considering the conditional probability distribution between the three physical properties and rock hardness, the following steps are used to determine the conditional probability:

[0067] (1) Data collection and status definition:

[0068] Data Collection: TSP observational data including rock hardness and corresponding dynamic Young's modulus, longitudinal wave velocity and Poisson's ratio are collected.

[0069] State definition: define the state for each variable; the dynamic Young's modulus is divided into five states according to the specific numerical range of the collected data, which can be briefly described as low, lower, medium, higher, and high, to facilitate the statistics of conditional probability.

[0070] (2) Frequency analysis and frequency calculation:

[0071] Frequency distribution:

[0072] For each rock hardness state, the frequencies of different combinations of the three attributes are counted, such as the number of occurrences of dynamic Young's modulus under different rock hardness states and the number of occurrences of different rock hardness states.

[0073] Ratio conversion:

[0074] Convert the joint occurrence frequency into probability. The conditional probability is expressed as the probability of other attribute states appearing under different states of one attribute. The probability calculation formula is as follows:

[0075]

[0076] By calculating the conditional probability, the conditional probability matrix of rock hardness and dynamic Young's modulus, longitudinal wave velocity, and Poisson's ratio is generated. An example of conditional probability is shown in Table 1:

[0077] Table 1 is the conditional probability table of rock hardness and dynamic Young's modulus

[0078]

[0079] Example 5

[0080] Probability update process (taking rock hardness as an example):

[0081] (1) Confirmation of prior probability

[0082] The prior probability is confirmed by known information. Prior probability is an important concept in the Bayesian inference process. It expresses the view of a thing in the form of probability. At this time, according to the excavated geological data parameter indicators behind the tunnel face collected in the data sets D1, D2 and D3, the prior probability calculation of the excavated geological data parameter indicators behind the tunnel face is realized, and the probability of rock hardness under different hardness conditions is calculated, so as to obtain the prior probability of rock hardness. ;

[0083] The specific calculation method is to statistically analyze the geological parameters of the excavated section behind the tunnel face. Taking rock hardness as an example, the rock hardness data within 50m behind the tunnel face is recorded and statistically analyzed. One piece of data is recorded every 2m. The frequency of occurrence of each rock hardness is recorded, and the probability of occurrence of each rock hardness is calculated. The calculation method is as follows:

[0084]

[0085] After the calculation is completed, some rock hardness states may not have appeared, resulting in a calculated probability of 0. When the prior probability is 0, the posterior probability will always be updated to 0, which does not meet the inference requirements. Therefore, the rock hardness state with a probability of zero needs to be adjusted. Increase the probability of all rock hardness states with a probability of 0 by 2%, and redistribute them according to the increased probability distribution so that the sum of the rock hardness probabilities is 1, so that it meets the probability update requirements;

[0086] When the prior probability of rock hardness is calculated to be: P(hard rock)=0.2, P(harder rock)=0.4, P(softer rock)=0.1, P(soft rock)=0.2, P(extremely soft rock)=0.1, it means that under the current circumstances, the probability of harder rock appearing is the highest, while the probability of extremely soft rock appearing is the lowest.

[0087] (2) Conditional probability calculation:

[0088] The conditional probability calculation of rock hardness is realized based on the collected data sets D1, D2 and D3, so as to obtain the conditional probability matrix value, which also represents the probability of observing a specific dynamic Young's modulus state under a certain rock hardness.

[0089] (3) Calculation of evidence factor:

[0090] The evidence factor is calculated by combining the likelihood of all hardness states, i.e. the marginal probability of the dynamic Young's modulus in all hardness states. If the state of the observed Young's modulus is low, the calculation method is as follows:

[0091]

[0092] The evidence factor is a normalization constant used to ensure that the sum of the posterior probabilities is 1. By calculating the likelihood of the dynamic Young's modulus under each condition and weighting the prior distribution, the marginal probability of the dynamic Young's modulus is obtained.

[0093] (4) Calculation of posterior probability:

[0094] According to the calculated prior probability, conditional probability and evidence factor, the posterior probability can be updated by inputting the data set D1. The update algorithm is as follows:

[0095]

[0096] in, To import the dataset The calculated posterior probability is is the prior probability, For the dataset The calculated conditional probability is is the evidential factor;

[0097] (5) Introduce the second data set D2 and calculate the new posterior probability:

[0098] The posterior probability of the previous update As a priori, the calculation method is as follows:

[0099]

[0100] in, To import the dataset The calculated posterior probability is is the prior probability, For the dataset The calculated conditional probability is is the evidential factor;

[0101] Here, we assume that the introduced data set D2 is the longitudinal wave velocity, and the observed longitudinal wave velocity state is medium, then the evidence factor The algorithm is as follows.

[0102]

[0103] In the above formula,

[0104] D2 is the observed longitudinal wave velocity state,

[0105] D1 is the observed dynamic Young's modulus state,

[0106] is the conditional probability that the observed P-wave velocity state is between medium and rock hardness,

[0107] is the posterior probability of rock hardness obtained during the first data update,

[0108] Through the calculation of the above formula, the posterior probability of introducing the D2 data set for secondary update is obtained .

[0109] (6) Introduce the third data set D3 and calculate the new posterior probability:

[0110] The posterior probability of the previous update As a priori, the calculation method is as follows:

[0111]

[0112] in, To import the dataset The calculated posterior probability is is the prior probability, For the dataset The calculated conditional probability is is the evidential factor;

[0113] Here, we assume that the introduced data set D3 is density, and assume that the observed density state is high, then the evidence factor The algorithm is as follows:

[0114]

[0115] In the above formula,

[0116] D3 is the observed Poisson's ratio state,

[0117] D2 is the observed longitudinal wave velocity state,

[0118] D1 is the observed density state,

[0119] is the conditional probability between the observed Poisson's ratio state being high and the rock hardness,

[0120] is the posterior probability of rock hardness obtained during the second data update,

[0121] By calculating the above formula, we can get the posterior probability of introducing the D3 dataset for three updates. .

[0122] At this point, a complete posterior probability update based on three data sets D1, D2, and D3 has been completed, and the generated posterior probability It represents the new probability distribution formed by the changes in concepts and probabilities after the prior probability is observed continuously for three data sets related to rock hardness.

[0123] Example 6

[0124] Geological data collection standards:

[0125] (1) Rock hardness classification and on-site rapid collection standards

[0126] In addition to measuring the lithology, structure, and genesis, the qualitative judgment of rock hardness also includes the degree of weathering and the development of joints and fissures. If all the above factors are taken into account during on-site judgment, it will not only take a lot of time, but also the factors have cross-relationships, and the feasibility of comprehensive evaluation is not high. In order to obtain the hardness of the surrounding rock intuitively and quickly on site, a geological hammer is used to knock on the rock, and the hammer sound and hammer reaction are recorded. The above factors are comprehensively considered and analyzed to verify each other and finally determine the hardness of the rock. The specific distinction method is shown in Table 2.

[0127] Table 2 shows the rock hardness identification method:

[0128] Rock hardness Uniaxial compressive strength (Mpa) Knocking sound Hammer reaction Hard Rock >60 Crisp Rebound, shock Hard rock 30-60 Crisper Slight rebound, slight vibration Soft rock 15-30 Not crisp No rebound, easy to break Soft Rock 5-15 Hoarseness No rebound, easy to break extremely soft rock <5 Hoarseness No rebound, deep dents, can be crushed by hand

[0129] When the accuracy of on-site data is high or the hammering phenomenon is difficult to distinguish, quantitative methods are used to judge. At present, the commonly used quantitative judgment methods in tunnels are mainly rebound hammer and point load method. Compared with the latter, rebound hammer is easy to carry, does not require sampling, and does not disturb the surrounding rock, so it is suitable for most tunnels.

[0130] as follows Figure 2 As shown in the figure, there is a relationship between the indoor strength results of the surrounding rock of the tunnel face in the western mountainous area and the rebound tester measurement values ​​converted according to the empirical formula. It shows that the conversion according to this formula can more accurately reflect the actual situation and be applied to rapid identification on site.

[0131] In the western region, most tunnels have hard surrounding rocks, and the rebound values ​​can be converted according to the above relationship.

[0132] (2) Groundwater status classification and on-site rapid collection standards

[0133] In addition to considering the characteristics of rock and structural surface, the surrounding rock classification also needs to determine the surrounding rock occurrence environment. If the surrounding rock is hard and complete Grade I and Grade II surrounding rock, groundwater has little effect on the surrounding rock grade; however, when the rock mass grade is poor, and multiple groups of joints and fissures are developed inside, and the rock type is a softer rock with water seepage, the above conditions can create advantages for groundwater flow, accelerate the deterioration of the face, and groundwater will pose a greater threat to stability. Therefore, it is necessary to judge the groundwater situation, and conduct on-site observations on the water discharge conditions of the surrounding rock face after tunnel excavation, mainly recording the water discharge location and water volume, so as to determine the groundwater status. The qualitative judgment of groundwater should be based on the most unfavorable situation for tunnel stability. The groundwater field observation and recording indicators are shown in Table 3.

[0134] Table 3 shows the groundwater field observation and record indicators:

[0135]

[0136] Table 3 gives a general description of the location and amount of groundwater exposure in detail. In the classification of tunnel surrounding rock, the determination of groundwater is not difficult. Therefore, it is only necessary to record it truthfully and analyze its most unfavorable situation to correct the surrounding rock grade. Generally, only tunnels with poor surrounding rock grades need to be corrected. In addition, if accurate quantitative data of groundwater is required, it can be measured according to Table 4.

[0137] Table 4 is a quantitative judgment table for groundwater on-site:

[0138] Groundwater conditions Water pressure p / Mpa 10m long hole water output Q (L / min*10m) dry 0 0 Wet or dripping ≤0.1 ≤25 Rain or line 0.1-0.5 25-125 Surge >0.5 >125

[0139] (3) Rock mass integrity classification and on-site rapid collection standards

[0140] The integrity of the rock mass is determined by the thickness of the rock layer, the development of the structural surface, the degree of rock mass bonding, and the type of rock mass structure. Therefore, it is necessary to make a quick judgment on several other indicators on site before comprehensively evaluating the integrity of the surrounding rock.

[0141] In addition to collecting and evaluating the degree of integration of rock mass structure and structural surfaces on site, it is also necessary to measure and record the average spacing of each group of structural surfaces, and make a comprehensive assessment of the rock mass integrity based on the above factors. The specific method for determining the rock mass integrity is shown in Table 5.

[0142] Table 5 is the basis for on-site determination of rock mass integrity:

[0143]

[0144] The quantitative acquisition method of rock mass integrity index in tunnel is usually based on RQD, integrity coefficient And the rock mass volume joint number . It refers to the number of joints contained in a unit volume of rock mass. Compared with the previous two, it does not require drilling cores or using geophysical acoustic waves, and is simpler and more convenient for tunnels with limited conditions.

[0145] At the construction site, you only need to measure the distance between the layout measurement line and the structural surface to obtain Results. When calculating the statistics of on-site geological indicators, joints and fissures that do not affect the stability of the rock mass are no longer taken into account. Joints and fissures with no fixed number of groups, uneven distribution, and good fissure ductility (>1m) should be included in the measurement. The specific calculation formula is as follows:

[0146]

[0147] in: Representatives along The total number of effective joints distributed in the normal direction of the group structural surface; For vertical The measurement wiring length of the group joints (usually ≥5m); is the number of random irregular cracks per unit volume ( ).

[0148] For the calculated The completeness result is obtained by looking up the value in Table 6.

[0149] Table 6 is the corresponding relationship between the number of rock mass volume joints and the degree of integrity:

[0150]

[0151] (4) Classification of geostress states and standards for rapid on-site collection

[0152] In addition to groundwater, the initial geostress in the geological environment also restricts the stability of the tunnel surrounding rock. Based on the previous research results, a mapping relationship between the surrounding rock characterization phenomena and internal geostress is established.

[0153] The ground stress is estimated by observing the degree of fit of the rock structure surface exposed to the outside. Under low ground stress conditions, the degree of fit of the structure surface is manifested as an open structure surface, mud in the joint surface, and loose rock structure; while under high ground stress conditions, it is manifested as a tightly closed structure surface and a dense structure.

[0154] During the excavation process, the stress conditions inside the surrounding rock are estimated based on the stability of the surrounding rock, that is, the disaster phenomena that are prone to occur. In a low-stress environment, the hard rock in the tunnel is prone to collapse and block falling, and the soft rock is prone to local deformation. In a high-stress environment, the hard rock in the tunnel is prone to rock burst disasters, and the soft rock generally has large deformations and core cake phenomena.

[0155] After the blasting is completed, the magnitude of the ground stress is determined based on the stress adjustment of the rock mass. In a low ground stress environment, the stress adjustment inside the surrounding rock is not active, which is manifested as sinking and convergence deformation inside the tunnel; in a high ground stress environment, the stress adjustment inside the surrounding rock is active, which is manifested as cracking caused by stress concentration, and even in the form of rock burst. The specific on-site rapid judgment criteria are shown in Table 7.

[0156] Table 7 is the field judgment table of initial ground stress of surrounding rock:

[0157] Ground stress state Chimerism Surrounding rock stability Stress adjustment degree Low ground stress Joints are open, mud is contained in the joint plane, there are secondary minerals or cavities, and water seepage in the surrounding rock Loose surrounding rock, landslides, falling blocks Stress adjustment is not active, which is manifested as the subsidence and convergence deformation of the cave wall rock mass. Moderate ground stress The joints are generally closed or slightly open, with no filling or siliceous filling. The hard rock wall is partially peeled off; the soft rock wall is partially deformed The stress adjustment is not active, which is manifested as peeling or local uplift of the foundation pit. High ground stress The joints are tightly closed, and there is extrusion of weak interlayers. Hard rock may experience rock bursts, with many new cracks; soft rock cores may become caked, with obvious displacement of the cave wall rock mass. Active stress adjustment is manifested by cracking, ejection and large displacement caused by stress concentration. Extremely high ground stress The joints are tightly closed, there is weak interlayer extrusion, and the permeability and connectivity are poor. Hard rock often experiences rock bursts; soft rock has obvious large deformation phenomena Stress adjustment is active in the form of rock ejection, explosion and continuous large deformation

[0158] After completing the statistics of the prior probability and the conditional probability, the present invention uses the method of using the posterior data as the prior data of the next step, and each time the observed D1, D2, and D3 states are input, the prior probability is combined with the influence of the three physical states to generate the posterior probability, and then the generated posterior probability is used as the prior probability, and the observed D1, D2, and D3 states are input multiple times to update the prior probability multiple times, thereby realizing the dynamic update process of the prior probability combined with the influence of the observed state; and according to the distribution of the posterior probability that is updated each time, the state with the highest probability in the rock hardness state is selected as the inference result;

[0159] The geological data of the excavated part of the tunnel is processed as a priori, that is, a general judgment of the rear of the tunnel. Then, TSP advanced geophysical exploration is used to collect TSP data behind the heading face, and the collected TSP data is input into the inference model as observation data. The input observation data will affect the priori according to the conditional probability relationship between TSP data and geological data to obtain the posterior, and the posterior is the inference result of the geological data of the unexcavated section corresponding to the input TSP data; then the updated posterior is used as the priori, and the subsequent TSP data is continuously input, so as to continuously update the posterior and obtain the geological data inference result, realizing the calculation of four indicators of rock hardness, groundwater conditions, integrity and ground stress of the geological data of the unexcavated part, providing data support for tunnel geological construction.

[0160] The above-mentioned embodiments are only preferred embodiments of the present invention and cannot be used to limit the scope of protection of the present invention. Any non-substantial changes and substitutions made by technicians in this field on the basis of the present invention shall fall within the scope of protection required by the present invention.

Claims

1. A tunnel geological parameter inference method based on multi-stage data Bayesian inference, characterized in that: The following steps are involved: Step S1, collecting unexcavated TSP data in front of the tunnel face and excavated geological data behind the tunnel face; Step S2, dividing the collected data into data sets; Step S3, performing conditional probability calculation on the data set to obtain conditional probability; Step S4, using the Bayesian inference model to calculate the prior probability of the data set to obtain the prior probability; Step S5, using the Bayesian inference model to perform a first posterior probability calculation on the data set to obtain a first posterior probability; Step S6, taking the first posterior probability as the prior probability, and inputting the data set into the Bayesian inference model to perform the second posterior probability; Step S7, repeating steps S5 and S6 to perform multiple calculations using the Bayesian inference model data set; Step S8, selecting the state with the highest probability as the inference result according to the content calculated in step S7; The unexcavated TSP data in front of the tunnel face in step S1 include longitudinal wave velocity, shear wave velocity, Poisson's ratio, density, static Young's modulus, dynamic Young's modulus and direct wave velocity; the excavated geological data behind the tunnel face include rock hardness, groundwater conditions, integrity and ground stress; In step S2, the data set is divided by combining the excavated geological data parameters behind the tunnel face with the unexcavated TSP data in front of the tunnel face; Among them, the rock hardness, groundwater conditions, integrity and ground stress in the excavated geological data parameters behind the tunnel face are used as references, and the data sets are divided into three groups: D1, D2 and D3. The selection of the types of unexcavated TSP data in front of the tunnel face corresponding to the excavated geological data parameter indicators behind the tunnel face is as follows: Rock hardness: D1: dynamic Young's modulus, D2: longitudinal wave velocity, D3: density; Groundwater conditions: D1: transverse wave velocity, D2: longitudinal wave velocity, D3: direct wave velocity; Completeness: D1: longitudinal wave velocity, D2: Poisson's ratio, D3: static Young's modulus; Geostress: D1: direct wave velocity, D2: Poisson's ratio, D3: density; The step S3 comprises the following steps: Step A1: Frequency distribution: Determine the frequency of occurrence of the excavated geological data parameters behind the tunnel face and the corresponding unexcavated TSP data in front of the tunnel face; Step A2: Proportional conversion: Convert the joint occurrence frequency into probability to obtain the conditional probability; In step S4, the prior probability of the excavated geological data parameter index behind the tunnel face is calculated based on the collected geological data parameter index behind the tunnel face, so as to obtain the prior probability.

2. A tunnel geological parameter inference method based on multi-stage data Bayesian inference as claimed in claim 1, characterized in that: In step S5, the initial update is performed using the Bayesian inference method to obtain the first posterior probability data.

3. The tunnel geological parameter inference method based on multi-stage data Bayesian inference as claimed in claim 1, characterized in that: In step S6, the second posterior calculation is performed using the first posterior probability as the prior probability, thereby obtaining the second posterior probability data.

4. The tunnel geological parameter inference method based on multi-stage data Bayesian inference as claimed in claim 1, characterized in that: In step S1, the unexcavated TSP data in front of the tunnel face is collected using the TSP method.

5. The tunnel geological parameter inference method based on multi-stage data Bayesian inference as claimed in claim 1, characterized in that: The Bayesian inference model includes: ； in, : Posterior probability, the probability that the parameter θ is the true value after given the data D; : Likelihood function, which indicates the probability of data D appearing when parameter θ is true; : Prior probability refers to the original belief about parameter θ before observing the current data; : The marginal likelihood, also called the normalizing constant or evidence factor, is the overall probability of the observation D occurring.

Images