A probabilistic evaluation method for geotechnical engineering stability

Abstract

The present invention discloses a probabilistic evaluation method for geotechnical engineering stability, establishes a mapping regression model of geotechnical physical and mechanical parameters and foundation pit deformation displacement values, trains the geotechnical mechanical parameters obtained by orthogonal design, and then obtains the foundation pit deformation displacement values ​​according to the geotechnical physical and mechanical parameters measured on site, and finally evaluates the stability of the foundation pit probabilistically through the stability probability of the foundation pit. The mapping regression model established by the present invention has a shorter calculation time, better calculation efficiency, and higher evaluation result accuracy, and can scientifically and effectively avoid the occurrence of over-protection and instability accidents that are common in the foundation pit construction process, saves manpower and material resources of on-site construction, and can also ensure on-site construction safety, which is more in line with the actual application needs of the construction site.

Description

Technical Field

[0001] The invention relates to the field of rock and soil deformation stability evaluation, and in particular to a probabilistic evaluation method for geotechnical engineering stability. Background Art

[0002] During the construction process, the evaluation of foundation pit stability has always been a difficult problem that the engineering safety department urgently needs to solve. In most construction projects, the stability of the foundation pit is evaluated by whether the deformation displacement of the foundation pit is within the control range. This method not only takes a lot of manpower and material resources, but also has poor evaluation accuracy, which often leads to excessive protection of the foundation pit and instability accidents. In fact, the deformation displacement of the foundation pit has a complex nonlinear mapping relationship with the physical and mechanical parameters of the rock and soil mass, which is difficult to express with a simple calculation equation. Summary of the invention

[0003] The present invention provides a probabilistic evaluation method for geotechnical engineering stability to overcome the above technical problems.

[0004] In order to achieve the above object, the technical solution of the present invention is:

[0005] A probabilistic evaluation method for geotechnical engineering stability comprises the following steps:

[0006] S1. Obtain the range of physical and mechanical parameters of the rock and soil mass at the construction site, and obtain the set of physical and mechanical parameters of the rock and soil mass based on the orthogonal design method;

[0007] S2. According to the set of physical and mechanical parameters of the rock and soil mass, a mapping regression model between the physical and mechanical parameters of the rock and soil mass and the deformation and displacement values ​​of the foundation pit is established to obtain a trained mapping regression model;

[0008] S3: Obtain the physical and mechanical parameters of the rock and soil measured on site, and obtain the deformation and displacement values ​​of the foundation pit on site according to the mapping regression model of the trained physical and mechanical parameters of the rock and soil and the deformation and displacement values ​​of the foundation pit;

[0009] S4. According to the deformation displacement value of the foundation pit at the site, the stability probability of the foundation pit is obtained to evaluate the stability of the foundation pit.

[0010] Furthermore, in S2, the mapping regression model of the physical and mechanical parameters of the rock and soil mass and the deformation and displacement values ​​of the foundation pit is established as follows:

[0011] S21: acquiring a plurality of mapping partitions according to the set of physical and mechanical parameters of the rock and soil body;

[0012] S22: Get the The data of the mapped partitions after the review series processing can be used by the simulation foundation pit model for regression mapping;

[0013] S23: Establishing a first input sequence formula and a second input sequence formula;

[0014] S24: According to the The data processed by the examined series of the mapping partitions that can be used by the simulation foundation pit model for regression mapping, the first input series formula and the second input series formula are used to obtain the data function of the improved simulation foundation pit model;

[0015] S25: Establish a sequence formula to obtain the first The phase transformation coefficient of the deformation displacement value of the foundation pit in each mapping partition;

[0016] S26: According to the The phase transformation coefficient of the deformation displacement value of the mapping partition foundation pit is obtained. The deformation and displacement values ​​of the foundation pit are mapped and divided into different zones to complete the establishment of a mapping regression model for the physical and mechanical parameters of the rock and soil mass and the deformation and displacement values ​​of the foundation pit.

[0017] Furthermore, in S22, the formula used to obtain the data processed by the review series that can be used by the simulation foundation pit model for regression mapping is as follows:

[0018] ,

[0019] in, The data processed by the review series can be used by the simulation foundation pit model for regression mapping; All are the correlation coefficients of the review series; For the The phase transformation coefficient of the deformation displacement value of the foundation pit in each mapping partition; Indicates Physical and mechanical parameters of rock and soil mass in each mapping partition; To examine the errors in the relationship between sequence processing; CC For the sigmoid activation function: .

[0020] Furthermore, in S23, the first input sequence formula and the second input sequence formula are established as follows:

[0021] ,

[0022] ,

[0023] Where: Represents the first input sequence formula, using the sigmoid activation function to record the input information; Represents the second input sequence formula, using the tanh function to record the input information; are the correlation coefficients of the first upper input sequence formula; They are all the correlation coefficients of the second upper series formula; is the relation error of the first input sequence; is the relation error of the second upper input sequence; is a hyperbolic tangent function:

[0024] .

[0025] Furthermore, in S24, the formula used to obtain the data function of the improved simulation foundation pit model is as follows:

[0026] ,

[0027] in, This is the data function of the simulation foundation pit model before improvement. This is the data function of the improved simulation foundation pit model.

[0028] Furthermore, in S25, the formula for the following sequence is established as follows:

[0029] ,

[0030] The formula used to obtain the phase transformation coefficient of the foundation pit deformation displacement value phase of the σth mapping partition is as follows: ,

[0031] Where: The data output for the next sequence; are the correlation coefficients of the following series; is the relational error of the following series; It is the phase transformation coefficient of the foundation pit deformation displacement value phase in the mapping partition.

[0032] Further, in S26, obtaining the The formula used for the deformation displacement value of the foundation pit in each mapping partition is as follows:

[0033] ,

[0034] Where: Represents the correlation coefficient of the regression mapping formula; For the Mapping partition foundation pit deformation displacement value; is the back-calculation error of foundation pit deformation.

[0035] Furthermore, in S3, the method used to obtain the on-site foundation pit deformation displacement value is as follows:

[0036] S31: According to the set of physical and mechanical parameters of the rock and soil mass measured on site, the mean and variance of the physical and mechanical parameters of the rock and soil mass measured on site are obtained; the formula used is as follows:

[0037] ,

[0038] ,

[0039] ,

[0040] ,

[0041] ,

[0042] ,

[0043] ,

[0044] ,

[0045] Where: They are the mean values ​​of cohesion, internal friction angle, Poisson's ratio and elastic modulus measured on site; All are the cohesion values ​​of rock and soil measured on site; All are the internal friction angle values ​​of rock and soil measured on site; All are the Poisson's ratios of rock and soil measured on site; These are the elastic modulus values ​​of rock and soil measured on site; is the total number of geotechnical parameters measured on site; They are the variance of cohesion, the variance of internal friction angle, the variance of Poisson’s ratio and the variance of elastic modulus;

[0046] S32: establishing a normal distribution of the mechanical parameters of the rock and soil mass according to the mean and variance of the physical and mechanical parameters of the rock and soil mass measured on site, and obtaining a sampled set of mechanical parameters of the rock and soil mass by adopting a sampling method with replacement based on the normal distribution of the mechanical parameters of the rock and soil mass;

[0047] Among them, the normal distribution of the mechanical parameters of the rock and soil mass is established as follows:

[0048] ,

[0049] ,

[0050] ,

[0051] ,

[0052] Where: represents the normal distribution of the cohesion measured in the field; Represents the normal distribution of the internal friction angle measured on site; represents the normal distribution of Poisson's ratio measured in the field; Represents the normal distribution of elastic modulus measured in the field;

[0053] S33: Obtain the foundation pit deformation displacement value corresponding to the sampled rock and soil body mechanical parameter set according to the mapping regression model between the rock and soil body physical and mechanical parameters and the foundation pit deformation displacement value.

[0054] Furthermore, in S4, the method for obtaining the stability probability of the foundation pit is as follows:

[0055] S41: based on industry standards, obtaining the number of foundation pit deformation displacement values ​​on site that meet the industry standards;

[0056] S42: Obtaining a probability of foundation pit stability according to the number of foundation pit deformation displacement values ​​on site that meet the industry standard; ,

[0057] Where: is the probability of foundation pit stability; The number of foundation pit deformation displacement values ​​that meet industry standards; R Represents the total number of mechanical parameters of rock and soil after sampling.

[0058] Beneficial effects: The present invention provides a probabilistic evaluation method for the stability of geotechnical engineering, which establishes a mapping regression model of the physical and mechanical parameters of the rock and soil mass and the deformation and displacement values ​​of the foundation pit, and trains the mechanical parameters of the rock and soil mass obtained by orthogonal design, and then obtains the deformation and displacement values ​​of the foundation pit according to the physical and mechanical parameters of the rock and soil mass measured on site, and finally, the stability of the foundation pit is evaluated probabilistically through the stability probability of the foundation pit. The mapping regression model established by the present invention has a shorter calculation time, better calculation efficiency, and higher evaluation result accuracy, and can scientifically and effectively avoid the occurrence of over-protection and instability accidents that are common in the foundation pit construction process, save manpower and material resources for on-site construction, and can also ensure on-site construction safety, which is more in line with the actual application needs of the construction site. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0060] Figure 1 It is a flow chart of the probabilistic evaluation method of rock and soil engineering stability of the present invention;

[0061] Figure 2 It is a progressive evaluation flow chart in an embodiment of the present invention. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0063] This embodiment discloses a probabilistic evaluation method for geotechnical engineering stability. Figure 1 and Figure 2 As shown, the following steps are included:

[0064] S1. Obtain the range of physical and mechanical parameters of the rock and soil at the construction site, based on the orthogonal design method, to obtain A set of physical and mechanical parameters of rock and soil is formed; the data in the set of physical and mechanical parameters of rock and soil is used as training data for the simulation foundation pit model;

[0065] Specifically, this embodiment selects the rock mass physical and mechanical surrounding rock grade 4 to 5 for orthogonal design. The range of the rock mass physical and mechanical parameters of the construction site in this embodiment is shown in Table 1, where a total of 72 physical and mechanical parameter sets are formed. . The corresponding 72 groups of foundation pit deformation displacement values ​​were obtained by bringing them into the simulation model, as shown in Table 2;

[0066] Table 1

[0067]

[0068] Table 2

[0069]

[0070] In Table 2, the last column "Displacement Value" corresponds to The pit bottom deformation displacement value is derived from the calculation of the simulated pit model. The significance of this embodiment is to simplify the calculation process and quickly obtain a large number of pit deformation displacement values ​​for evaluation.

[0071] In this embodiment, according to the physical and mechanical parameters of the rock and soil at the construction site, appropriate levels of the physical and mechanical parameters of the rock and soil are selected for orthogonal design, and the A set of physical and mechanical parameters of rock and soil ,in, Indicates cohesion; represents the internal friction angle; represents Poisson's ratio; represents the elastic modulus; Represents a set of physical and mechanical parameters of rock and soil; further: A set of physical and mechanical parameters of rock and soil Bring in the simulation foundation pit model to obtain The corresponding displacement value of the bottom of the foundation pit ,in, Represents the set of deformation displacement values ​​of the foundation pit bottom; Indicates W The deformation displacement value of the foundation pit bottom corresponding to the physical and mechanical parameters of the rock and soil mass of the group; the simulation foundation pit model in this embodiment includes but is not limited to the foundation pit model established by computer software such as PFC, flac3D, and ansys.

[0072] S2. According to the set of physical and mechanical parameters of the rock and soil mass, a mapping regression model between the physical and mechanical parameters of the rock and soil mass and the deformation and displacement values ​​of the foundation pit is established to obtain a trained mapping regression model;

[0073] Preferably, the mapping regression model of the physical and mechanical parameters of the rock and soil mass and the deformation displacement value of the foundation pit is established as follows:

[0074] S21: acquiring a plurality of mapping partitions according to the set of physical and mechanical parameters of the rock and soil body;

[0075] Specifically, in this embodiment, the set of physical and mechanical parameters of the rock and soil mass collected at the test point and obtained through orthogonal design is: , the set of foundation pit deformation displacement values ​​calculated by the simulation foundation pit model is .Will , the data forms learning samples, and the number of learning samples obtained is divided into mapping partitions, and after regression fitting each mapping partition, establish their own independent mapping relationship functions. Phase transformation coefficient of the deformation displacement value of the foundation pit in the mapping partition Calculate the final deformation displacement value In this embodiment, the mapping partitions are divided into two types according to actual needs by technicians in the field. Map partitions.

[0076] Specifically, in this embodiment, The transformation coefficient of the deformation displacement value of the foundation pit in each mapping partition is:

[0077] (1)

[0078] Where: For the The phase transformation coefficient of the deformation displacement value of the foundation pit in each mapping partition; is an abstract function, is the calculation matrix of foundation pit deformation; Indicates Physical and mechanical parameters of rock and soil mass in each mapping partition; is the error of foundation pit deformation; Indicates the index number of the mapping partition;

[0079] ,

[0080] Where: For the Mapping partition foundation pit deformation displacement value; is the anti-abstraction function; is the back calculation coefficient of foundation pit deformation; is the back-calculation error of foundation pit deformation;

[0081] S22: Get the The data of the mapped partitions after the review series processing can be used by the simulation foundation pit model for regression mapping;

[0082] Preferably, the formula used to obtain the data processed by the review series that can be used by the simulation foundation pit model for regression mapping is as follows:

[0083] Specifically, this embodiment is based on Taking the mapping partition as an example, the physical and mechanical parameter values ​​of the rock and soil measured on site are The phase transformation coefficient of the foundation pit deformation displacement value of the previous mapping partition Substitute the following review series formula:

[0084] (2)

[0085] in, The data processed by the review series can be used by the simulation foundation pit model for regression mapping; , All are the correlation coefficients of the review series; For the The phase transformation coefficient of the deformation displacement value of the foundation pit in each mapping partition; Indicates Physical and mechanical parameters of rock and soil mass in each mapping partition; To examine the errors in the relationship between sequence processing; CCis the sigmoid activation function; its expression is:

[0086] (3)

[0087] Specifically, the review sequence in this embodiment is used to screen and review invalid information in the data input into the simulation foundation pit model, and only valid information is allowed to be input into the simulation foundation pit model for the next step of training;

[0088] S23: Establishing a first input sequence formula and a second input sequence formula;

[0089] Specifically, the Physical and mechanical parameters of rock and soil mass in the mapping partitions With Phase transformation coefficient of the deformation displacement value of the foundation pit in the mapping partition Substitute the following input sequence formula:

[0090] (4)

[0091] (5)

[0092] Where: Represents the first input sequence formula, using the sigmoid activation function to record the input information; Represents the second input sequence formula, using the tanh function to record the input information; , are the correlation coefficients of the first upper input sequence formula; , They are all the correlation coefficients of the second upper series formula; is the relation error of the first input sequence; is the relation error of the second upper input sequence; is a hyperbolic tangent function:

[0093] (6)

[0094] Specifically, the input sequence formula in this embodiment is used to select input information from the first input sequence formula and the second input sequence formula, that is, the number of rock and soil mechanical parameters that can be stored in the current mapping partition;

[0095] S24: obtaining a data function of an improved simulation foundation pit model according to the data that can be used by the simulation foundation pit model for regression mapping after the review sequence processing, the first input sequence formula and the second input sequence formula;

[0096] Specifically, this embodiment uses the data that can be used by the simulation foundation pit model for regression mapping after the review series is processed. And based on the first upper sequence formula And the second upper sequence formula The data obtained can be stored in the simulation foundation pit model to the greatest extent, and the data function of the simulation foundation pit model is processed to improve the data function of the simulation foundation pit model.

[0097] Preferably, the formula used to obtain the data function of the improved simulation foundation pit model is as follows:

[0098] (7)

[0099] in, This is the data function of the simulation foundation pit model before improvement. It is the data function of the improved simulation foundation pit model;

[0100] Specifically, in this embodiment, the data function of the initial simulation model and the transformation coefficient of the initial mapping partition foundation pit deformation displacement value phase are both obtained by initializing the mapping regression model.

[0101] S25: Establish a sequence formula to obtain the first The phase transformation coefficient of the deformation displacement value of the foundation pit in each mapping partition;

[0102] Specifically, the data function of the improved simulation foundation pit model is The set of parameters monitored by the monitoring system of the mapping partition, namely Physical and mechanical parameters of rock and soil mass in the mapping partitions The phase transformation coefficient of the foundation pit deformation displacement value of the previous mapping partition Substitute the following sequence formula:

[0103] (8)

[0104] The formula used to obtain the phase transformation coefficient of the foundation pit deformation displacement value phase of the σth mapping partition is as follows: (9)

[0105] Where: The data output for the next sequence; , are the correlation coefficients of the following series; is the relational error of the following series; for The phase transformation coefficient of the foundation pit deformation displacement value phase in the mapping partition.

[0106] Specifically, the following series formula in this embodiment is used to calculate and output the updated phase transformation coefficient and data function;

[0107] S26: According to the The phase transformation coefficient of the deformation displacement value of the mapping partition foundation pit is obtained For the The deformation and displacement values ​​of the foundation pit are mapped and divided into different zones, and the mapping regression model between the physical and mechanical parameters of the rock and soil mass and the deformation and displacement values ​​of the foundation pit is established;

[0108] Preferably, obtaining The formula used for the deformation displacement value of the foundation pit in each mapping partition is as follows:

[0109] (10)

[0110] Where: Represents the correlation coefficient of the regression mapping formula; For the Mapping partition foundation pit deformation displacement value; is the back-calculation error of foundation pit deformation.

[0111] S3: Obtain the physical and mechanical parameters of the rock and soil measured on site, and obtain the deformation and displacement values ​​of the foundation pit on site according to the mapping regression model of the trained physical and mechanical parameters of the rock and soil and the deformation and displacement values ​​of the foundation pit;

[0112] This embodiment establishes a normal distribution for the mechanical parameters of rock and soil measured on site and performs sampling, and imports the sampling results into a mapping regression model; the details are as follows:

[0113] S31: according to the set of physical and mechanical parameters of the rock and soil mass measured on site, the physical and mechanical parameters of the rock and soil mass collected on site are substituted into the following formula to obtain the mean and variance of the physical and mechanical parameters of the rock and soil mass measured on site;

[0114] (11)

[0115] (12)

[0116] (13)

[0117] (14)

[0118] (15)

[0119] (16)

[0120] (17)

[0121] (18)

[0122] Where: They are the mean values ​​of cohesion, internal friction angle, Poisson's ratio and elastic modulus measured on site; All are the cohesion values ​​of rock and soil measured on site; All are the internal friction angle values ​​of rock and soil measured on site; All are the Poisson's ratios of rock and soil measured on site; These are the elastic modulus values ​​of rock and soil measured on site; is the total number of geotechnical parameters measured on site; They are the variance of cohesion (A), the variance of internal friction angle (B), the variance of Poisson’s ratio (C), and the variance of elastic modulus (D);

[0123] S32: According to the mean and variance of the physical and mechanical parameters of the rock and soil mass measured on site, a normal distribution of the mechanical parameters of the rock and soil mass is established, and based on the normal distribution of the mechanical parameters of the rock and soil mass, a random sampling method with replacement is used to select R times, obtained R Mechanical parameter set of rock and soil after group sampling ;

[0124] Among them, the normal distribution probability of rock and soil mechanical parameters is as follows:

[0125] ,

[0126] ,

[0127] ,

[0128] ,

[0129] Where: represents the normal distribution of the cohesion measured in the field;

[0130] S33: According to the R The mapping regression model between the set of rock and soil mechanical parameters after group sampling and the physical and mechanical parameters of the rock and soil and the deformation displacement value of the foundation pit is obtained. R The deformation displacement value of the foundation pit corresponding to the set of mechanical parameters of the rock and soil mass after group sampling;

[0131] Specifically, in this embodiment, R Mechanical parameter set of rock and soil after group sampling Bring it into the mapping regression model of the physical and mechanical parameters of the rock and soil mass and the deformation and displacement values ​​of the foundation pit, and bring it in one group at a time, and repeat it over and over again. R times, obtain R The deformation displacement value of the foundation pit corresponding to the set of mechanical parameters of the rock and soil mass after group sampling .

[0132] Specifically, this embodiment takes the 72nd set of data as an example, and transforms the parameter values ​​and the foundation pit deformation displacement values ​​measured on site into a matrix and introduces it into the review series formula:

[0133] ,

[0134] Substitute the transformation coefficient of the historical monitoring data set and the foundation pit deformation displacement value into the upper input series formula:

[0135] ,

[0136] ,

[0137] Data processed using censored series and the input sequence to store the data into the model to the maximum extent , Process the data function of the model and improve the data function of the model:

[0138] ,

[0139] Substitute the improved data function, the historical monitoring data set and the phase transformation coefficient of the foundation pit deformation displacement value into the following series function formula:

[0140] ,

[0141] ,

[0142] Finally, the predicted displacement value is obtained

[0143] ;

[0144] Establish the normal distribution probability of each geotechnical parameter , and randomly selected by sampling with replacement R times, obtained R A set of mechanical parameters of rock and soil ;

[0145] The sampled R A set of mechanical parameters of rock and soil Bring it into the mapping regression model of the trained rock and soil physical and mechanical parameters and foundation pit deformation displacement values, and bring in one group each time, and repeat it over and over again. R times, obtainR The deformation displacement value of the foundation pit corresponding to the set of mechanical parameters of the rock and soil mass after group sampling is: .

[0146] S4. According to the deformation and displacement values ​​of the foundation pit at the site, statistical collation and induction are performed, and the stability probability of the foundation pit is obtained with reference to domestic industry standards, so as to evaluate the stability of the foundation pit; the output results of the mapping regression model of the deformation and displacement values ​​of the foundation pit are statistically analyzed, and with reference to relevant domestic standards, the ratio of the foundation pits that meet the standards to all the foundation pits is calculated;

[0147] S41: based on industry standards, obtaining the number of foundation pit deformation displacement values ​​on site that meet the industry standards;

[0148] Specifically, this embodiment is based on relevant industry standards. R The deformation displacement value of the foundation pit corresponding to the set of mechanical parameters of the rock and soil mass after group sampling Find the foundation pit deformation displacement values ​​that meet industry standards and count their number .

[0149] S42: Obtain the probability of foundation pit stability according to the number of foundation pit deformation displacement values ​​on site that meet the industry standards; the formula used is as follows:

[0150] (19)

[0151] Where: is the probability of foundation pit stability; The number of foundation pit deformation displacement values ​​that meet industry standards;

[0152] At this point, the probability of foundation pit stability is obtained. According to the above steps, the deformation displacement value of the foundation pit can be directly, quickly and accurately calculated and evaluated based on the currently measured mechanical parameters of the rock and soil [cohesion (A), internal friction angle (B), Poisson's ratio (C) and elastic modulus (D)]. It no longer relies on the on-site measured deformation displacement value of the foundation pit for evaluation, and can evaluate the stability of the foundation pit more accurately, scientifically and quickly.

[0153] In order to further prove the accuracy of the model, 100,000 sets of rock and soil physical and mechanical parameters were selected, and their parameter values ​​are shown in Table 3.

[0154] Table 3

[0155]

[0156] Table 4 summarizes the prediction errors of the 100,000 test sample groups shown in Table 3. It can be seen that the evaluation results of this embodiment differ from the actual results by only 2.6%, with a high accuracy rate, which can meet the needs of foundation pit stability evaluation at the construction site.

[0157] Table 4

[0158]

[0159] Specifically, in order to obtain a scientific and accurate probabilistic evaluation method for foundation pit stability and solve the possible over-protection and instability accidents during foundation pit construction, this embodiment uses the mechanical parameters of the rock and soil body obtained by orthogonal design to obtain the mapping relationship between the physical and mechanical parameters of the rock and soil body and the deformation and displacement values ​​of the foundation pit, and then performs a sampling with replacement according to the normal distribution of the physical and mechanical parameters of the rock and soil body measured on site, and brings the sampling results into the trained mapping model to obtain the deformation and displacement values ​​of the foundation pit, and statistically output the distribution characteristics of the values, obtain the stability probability of the foundation pit, and perform a probabilistic evaluation of the stability of the foundation pit. The mapping regression model established by the present invention has a shorter calculation time, better calculation efficiency, and higher evaluation result accuracy. It can scientifically and effectively avoid the occurrence of over-protection and instability accidents that are common in the foundation pit construction process, save the manpower and material resources of the on-site construction, and also ensure the safety of the on-site construction, which is more in line with the actual application needs of the construction site. Compared with judging the stability of the foundation pit based on whether the deformation and displacement of the on-site foundation pit are in a controllable range, this embodiment effectively saves manpower and material resources on the construction site, and provides a fast, accurate, and scientific probabilistic evaluation method for the construction site.

[0160] Specifically, this embodiment realizes the probabilistic evaluation of foundation pit deformation stability from the overall perspective. The overall evaluation and analysis process is completely implemented based on code programming. It is an intelligent digital evaluation method that avoids the large amount of human and financial resources consumed by traditional evaluation methods and improves the accuracy of calculation results. It also includes the following beneficial effects:

[0161] (1) This embodiment can quickly and accurately map the deformation and displacement values ​​of the foundation pit. There is a mapping relationship between the mechanical parameters of the rock and soil mass and the deformation and displacement values ​​of the foundation pit, but this relationship is often implicit and relatively complex. The model established in this embodiment can make full use of multiple sets of parameter sets obtained by sampling within the normal distribution range of each parameter to perform a large number of mapping calculations, and quickly and accurately obtain a large number of foundation pit deformation and displacement values;

[0162] (2) The analysis of foundation pit stability in this embodiment is highly reliable. The implementation process of this embodiment is simple. First, the instability deformation displacement standard of the foundation pit is determined according to the relevant industry production rules and regulations, and the large number of foundation pit deformation displacements output by the model are classified and counted according to the instability standard displacement value (that is, the ratio of the displacement value output by the model that is less than the instability standard to the total number of displacements output by the model). Finally, the stability of the foundation pit is determined according to the ratio. The evaluation method of this embodiment is more scientific and accurate than the traditional evaluation method, and effectively prevents excessive protection of the foundation pit and the occurrence of instability accidents at the construction site. While saving construction costs, it also ensures the safety of construction on site.

[0163] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A probabilistic evaluation method for geotechnical engineering stability, characterized in that: The steps include: S1. Obtain the range of physical and mechanical parameters of the rock and soil mass at the construction site, and obtain the set of physical and mechanical parameters of the rock and soil mass based on the orthogonal design method; S2. According to the set of physical and mechanical parameters of the rock and soil mass, a mapping regression model between the physical and mechanical parameters of the rock and soil mass and the deformation and displacement values ​​of the foundation pit is established to obtain a trained mapping regression model; In S2, the mapping regression model between the physical and mechanical parameters of the rock and soil mass and the deformation and displacement values ​​of the foundation pit is established as follows: S21: acquiring a plurality of mapping partitions according to the set of physical and mechanical parameters of the rock and soil body; S22: Obtaining data of the σth mapping partition that has been processed by the review series and can be used by the simulation foundation pit model for regression mapping; In S22, the formula used to obtain the data processed by the review series that can be used by the simulation foundation pit model for regression mapping is as follows: E σ =CC(Q E *V σ-1 +Y E *α σ +ω E ) Among them, E σ It is the data processed by the review series and can be used by the simulation foundation pit model for regression mapping; Q E , Y E are the correlation coefficients of the review series; V σ-1 is the phase transformation coefficient of the foundation pit deformation displacement value phase in the σ-1th mapping partition; α σ represents the physical and mechanical parameters of the rock mass in the σth mapping partition; ω E To examine the error of the sequence processing relationship; CC is the sigmoid activation function: ； S23: Establishing a first input sequence formula and a second input sequence formula; S24: obtaining a data function of an improved simulation foundation pit model according to the data that can be used by the simulation foundation pit model for regression mapping after the review series of the σ-th mapping partition, the first input series formula and the second input series formula; S25: establishing a numerical series formula to obtain the phase transformation coefficient of the deformation displacement value phase of the σth mapping partition foundation pit according to the data function of the improved simulation foundation pit model; S26: according to the phase transformation coefficient of the foundation pit deformation displacement value phase of the σth mapping partition, the foundation pit deformation displacement value of the σth mapping partition is obtained to complete the establishment of a mapping regression model for the physical and mechanical parameters of the rock and soil body and the foundation pit deformation displacement value; S3: Obtain the physical and mechanical parameters of the rock and soil measured on site, and obtain the deformation and displacement values ​​of the foundation pit on site according to the mapping regression model of the trained physical and mechanical parameters of the rock and soil and the deformation and displacement values ​​of the foundation pit; In S3, the method used to obtain the on-site foundation pit deformation displacement value is as follows: S31: According to the set of physical and mechanical parameters of the rock and soil mass measured on site, the mean and variance of the physical and mechanical parameters of the rock and soil mass measured on site are obtained; the formula used is as follows: Where: are the mean values ​​of cohesion, internal friction angle, Poisson's ratio and elastic modulus measured on site; x A1 、x A2 ...x Ap All are the cohesion values ​​of rock and soil measured on site; x B1 、x B2 ...x Bp The internal friction angles of rock and soil measured on site are all values; x C1 、x C2 ...x Cp are the Poisson's ratios of rock and soil measured on site; x D1 、x D2 ...x Dp are the elastic modulus values ​​of rock and soil measured on site; p is the total number of mechanical parameters of rock and soil measured on site; s A 2 、s B 2 、s C 2 、s D 2 They are the variance of cohesion, the variance of internal friction angle, the variance of Poisson’s ratio and the variance of elastic modulus; S32: establishing a normal distribution of the mechanical parameters of the rock and soil mass according to the mean and variance of the physical and mechanical parameters of the rock and soil mass measured on site, and obtaining a sampled set of mechanical parameters of the rock and soil mass by adopting a sampling method with replacement based on the normal distribution of the mechanical parameters of the rock and soil mass; Among them, the normal distribution of the mechanical parameters of the rock and soil mass is established as follows: Where: Φ A represents the normal distribution of the cohesion measured on site; Φ B Represents the normal distribution of the internal friction angle measured on site; Φ C Represents the normal distribution of Poisson's ratio measured on site; Φ D Represents the normal distribution of elastic modulus measured in the field; S33: obtaining the foundation pit deformation displacement value corresponding to the sampled rock and soil body mechanical parameter set according to the mapping regression model between the rock and soil body physical and mechanical parameters and the foundation pit deformation displacement value; S4. Obtaining the stability probability of the foundation pit according to the deformation displacement value of the foundation pit at the site, so as to evaluate the stability of the foundation pit; In S4, the method for obtaining the stability probability of the foundation pit is as follows: S41: based on industry standards, obtaining the number of foundation pit deformation displacement values ​​on site that meet the industry standards; S42: Obtaining a probability of foundation pit stability according to the number of foundation pit deformation displacement values ​​on site that meet the industry standard; μ=λ / R Where: μ is the probability of foundation pit stability; λ is the number of foundation pit deformation displacement values ​​that meet industry standards; R represents the total number of rock and soil mechanical parameters after sampling.

2. A probabilistic evaluation method for geotechnical engineering stability according to claim 1, characterized in that: In S23, the first input sequence formula and the second input sequence formula are established as follows: F σ =CC(Q F *V σ-1 +Y F *α σ +ω F ) T σ =tanh(Q T *V σ-1 +Y T *α σ +ω T ) Where: F σ represents the first input sequence formula, using the sigmoid activation function to record the input information; T σ Represents the second input sequence formula, using the tanh function to record the input information; Q F , Y F are the correlation coefficients of the first input sequence formula; Q T , Y T are the correlation coefficients of the second upper series formula; ω F is the relation error of the first input sequence; ω T is the relational error of the second upper input sequence; tanh(·) is the hyperbolic tangent function: tanh(·)=sinh(·)*cos -1 h(·)。 3. A probabilistic evaluation method for geotechnical engineering stability according to claim 1, characterized in that: In S24, the formula used to obtain the data function of the improved simulation foundation pit model is as follows: N σ =E σ *N σ-1 +F σ *T σ Among them, N σ-1 is the data function of the simulation foundation pit model before improvement, N σ This is the data function of the improved simulation foundation pit model.

4. A probabilistic evaluation method for geotechnical engineering stability according to claim 1, characterized in that: In S25, the formula for the next sequence is established as follows: J σ =CC(Q J *In σ-1 +Y J *α σ +ω J ) The formula used to obtain the phase transformation coefficient of the foundation pit deformation displacement value phase of the σth mapping partition is as follows: V σ =J σ *fishy(N) σ ) Where: J σ The data output for the next sequence; Q J , Y J are the correlation coefficients of the following series; ω J is the relation error of the following series; V σ is the phase transformation coefficient of the foundation pit deformation displacement value phase of the σth mapping partition.

5. A probabilistic evaluation method for geotechnical engineering stability according to claim 1, characterized in that: In S26, the formula used to obtain the deformation displacement value of the foundation pit of the σth mapping partition is as follows: K σ =CC(Q K *V σ +ω K ) Where: Q K Represents the correlation coefficient of the regression mapping formula; K σ is the deformation displacement value of the foundation pit in the σth mapping partition; ω K is the back-calculation error of foundation pit deformation.

Images