Method for selecting optimal shear wave velocity prediction model according to field demand evaluation

Abstract

The invention belongs to the technical field of geotechnical engineering test and investigation, and provides a method for selecting an optimal shear wave velocity prediction model according to field demand evaluation, which can quantitatively evaluate the accuracy and rationality of a plurality of shear wave velocity prediction models and select the optimal prediction model. A reference value is provided for accurate estimation of the shear wave velocity and geotechnical engineering test and investigation; according to the technical scheme, the method comprises the following steps: evaluating and selecting a shear wave velocity optimal prediction model according to field requirements; the method comprises the following steps: determining a performance angle target area alpha'according to field requirements, respectively calculating mean values of center rates i and deviation angles alpha of various types of models for multiple types of shear wave velocity prediction models, performing comparative analysis to evaluate the accuracy and rationality of the shear wave velocity prediction models, and selecting an optimal shear wave velocity prediction model.

Description

Technical Field

[0001] The invention discloses a method for selecting an optimal prediction model for shear wave velocity according to field demand evaluation, and belongs to the technical field of geotechnical engineering testing and investigation. Background Art

[0002] Shear wave velocity is the most basic characteristic parameter of soil materials. It reflects the hardness of the propagation medium and the dynamic response characteristics of the small strain shear modulus of the soil, and has become one of the widely used physical quantities in earthquake engineering and geotechnical engineering. It is widely used in many aspects such as site superior period determination, earthquake site response analysis, site classification, soil dynamic parameter calculation, liquefaction evaluation, foundation dynamic design, etc. The shear wave velocity is closely related to the shear strength and shear deformation behavior of the soil, which comprehensively reflects the structural properties of the soil. Therefore, the determination or testing of shear wave velocity parameters is particularly important.

[0003] The methods for determining shear wave velocity can usually be divided into two categories: indoor bending element test and in-situ test method. The in-situ test method can give the in-situ engineering characteristics of the soil layer on site, which is more in line with the actual situation. Therefore, the in-situ test method is often used in engineering practice and is widely used in various fields of geotechnical engineering. Since the shear wave velocity requires professional equipment and professionals in the process of testing, which is costly, it is not feasible or economical to directly test the shear wave velocity in some low-risk areas or some projects. Therefore, it is very important to accurately estimate the shear wave velocity. Usually, other in-situ test techniques such as static penetration test and standard penetration test are used for estimation. Therefore, shear wave velocity prediction models of different types are established for different soils or unified soils. However, in the face of numerous models, how to choose the appropriate model formula for direct analysis or what function form to use for the construction of a new site-specific model is the core issue that geotechnical engineers are most concerned about. Therefore, new evaluation or discrimination performance indicators are of great significance. The high cost of existing in-situ shear wave velocity test determination, how to select existing shear wave velocity prediction models, and how to evaluate the accuracy or performance of existing shear wave velocity prediction models or constructed shear wave velocity prediction models have become issues that need to be urgently addressed. Summary of the invention

[0004] In order to solve one of the above-mentioned technical defects, the present invention provides a method for selecting the optimal prediction model of shear wave velocity according to field demand evaluation, which can quantitatively evaluate the accuracy and rationality of multiple shear wave velocity prediction models and select the optimal prediction model, providing reference value for accurate estimation of shear wave velocity and geotechnical engineering testing and investigation.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is: a method for selecting the optimal prediction model of shear wave velocity according to the field demand evaluation, determining the performance angle target area α' according to the field demand, and calculating the mean values ​​of the center rate i and the deviation angle |α| of each type of model for multiple categories of shear wave velocity prediction models, and performing comparative analysis to evaluate the accuracy and rationality of the shear wave velocity prediction model, and selecting the optimal prediction model of shear wave velocity. The specific steps are as follows:

[0006] S1, determine the performance angle target area α' according to the actual needs, that is, the range of allowable deviation of the model prediction results;

[0007] S2, using the in-situ test method to carry out tests in multiple boreholes on the site, each borehole is divided into multiple main layers along the depth direction according to the hole depth, and the basic parameters and shear wave velocity of each main layer are obtained;

[0008] S3, interpreting derived parameters according to the basic parameters of each main layer to obtain a one-to-one corresponding shear wave velocity and static penetration parameter data set;

[0009] S4, substitute the static penetration parameter data set of each main layer into various models, and calculate the corresponding shear wave velocity prediction value V under various models. sE , forming the corresponding shear wave velocity prediction value V under various model formulas sE and the measured shear wave velocity V sM Data pair, plotting the predicted shear wave velocity V sE —Measured value of shear wave velocity V sM Draw V on the coordinate point diagram sE =V sM Reference lines;

[0010] S5, the predicted shear wave velocity V obtained above sE and the measured shear wave velocity V sM Data pairs, calculate the performance angle index α of each main layer under each model:

[0011]

[0012] According to the performance angle target area α' given in step S1, the performance angle target area α' is the shear wave velocity prediction value V sE —Measured value of shear wave velocity V sM The coordinate point V on the graph sE =V sM The reference line is offset to both sides by an angle of α' with the origin of the coordinate system as the center. When the absolute value of the performance angle index α is less than the angle α', it represents the corresponding shear wave velocity prediction value V sE and the measured shear wave velocity V sM The data points of the data pair are within the performance perspective target region α';

[0013] S6. Determine the number of data points within the performance angle target area α' according to step S5, and calculate the ratio of data points that meet the performance angle target area α' under each model, that is, the center rate i:

[0014]

[0015] The central rates i corresponding to the various models are compared and analyzed, and the prediction models with central rates i lower than 90% are screened out, and the remaining prediction models are the better prediction models;

[0016] S7. Calculate the mean deviation angles |α| of all main layers under each model, compare the mean deviation angles |α| of all better prediction models, and select the corresponding model with the smallest mean deviation angle |α| as the optimal prediction model.

[0017] The in-situ test method in step S1 is a static penetration test or a standard penetration test.

[0018] The static penetration test in step S1 adopts the following two methods:

[0019] a) Perform a shear wave velocity test on the static penetration test hole at the same time, and use the shear wave velocity test hole data obtained from the shear wave velocity test as the shear wave velocity test value used in subsequent calculations;

[0020] b) If the shear wave velocity test is not carried out simultaneously on the static penetration test hole, the shear wave velocity test hole data of the nearest borehole to the static penetration test hole is selected as the shear wave velocity test hole data for subsequent calculations, and the distance between the nearest borehole and the hole is less than 0.5m.

[0021] In step S1, each borehole is divided into a plurality of main layers according to the hole depth in the depth direction, and each main layer is further divided into a plurality of sub-layers. Each main layer is tested for a corresponding shear wave velocity, and each sub-layer is tested for a corresponding set of basic parameters. In step S2, the basic parameters of all sub-layers in each main layer are averaged, which is the basic parameter corresponding to the shear wave velocity of the main layer.

[0022] The in-situ test method in step S1 is specifically as follows:

[0023] The sublayer is set to 0.05m and the main layer to 1m. A set of basic parameter data of static penetration is collected every 0.05m along the depth of each borehole. The penetration is paused every 1m to start collecting and recovering shear wave velocity parameter data.

[0024] The basic parameters in step S1 include the cone tip resistance q t , side wall friction f s .

[0025] The derived parameters in step S2 include the average friction ratio R f , normalized friction ratio F r , pore pressure parameter B q , Soil Classification Index I c , Normalized cone tip resistance Q tn .

[0026] The specific interpretation of the derived parameters in step S2 is as follows:

[0027] Friction ratio R f The expression is:

[0028] R f =f s / q t ×100% (1)

[0029] Normalized friction ratio F r The expression is:

[0030] F r =f s / (q t -σ v0 ) (2)

[0031] In the formula, σ v0 is the overburden pressure on the soil layer;

[0032] Pore ​​pressure parameter B q The expression is:

[0033] B q =(u2-u0) / (q t -σ v0 ) (3)

[0034] Where u0 is the hydrostatic pressure;

[0035] Soil Classification Index I c The calculation method is as follows:

[0036]

[0037] In the formula, Q tn is the normalized cone tip resistance, which is expressed as:

[0038]

[0039] In the formula, σ′ v0 is the effective overburden pressure; n is the stress index;

[0040] Specifically, n = 0.381I c +0.05(σ′ v0 / p a )-0.15≤1; where Ic is the soil classification index; p a is the reference pressure, which is atmospheric pressure.

[0041] After the basic parameters and shear wave velocity of each main layer are obtained in step S1, the data is preprocessed first, including eliminating abnormal points based on mathematical statistical analysis theory, and obtaining corresponding data groups of multiple shear wave velocities and static penetration parameters for model analysis as data sets for model analysis or construction.

[0042] The mean of the deviation angle |α| may also be the mean of |sinα| or the mean of |tanα|. In step S7, the corresponding model with the smallest mean of |sinα| or |tanα| is selected as the optimal prediction model.

[0043] The method for selecting the optimal prediction model for shear wave velocity based on field demand evaluation provided by the present invention effectively solves the problems of high cost, poor construction period, and how to select a shear wave velocity estimation method or prediction model in existing in-situ shear wave velocity determination. It can make comparisons within different coordinate ranges, and accurately and comprehensively evaluate the performance of the established model and the concentration and discreteness of the prediction range, thereby providing an important basis for rapid and accurate shear wave velocity estimation and having good engineering application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The present invention will be further described in detail below in conjunction with the accompanying drawings;

[0045] Figure 1 is a flow chart of the method of the present invention;

[0046] Figure 2 is the shear wave velocity prediction value V of each model in the embodiment of the present invention sE —Measured value of shear wave velocity V sM Coordinate point diagram of ;

[0047] Figure 3 1 is a schematic diagram of the range of the performance angle target area α' of Embodiment A of the present invention on each model coordinate point diagram;

[0048] Figure 4 1 is a schematic diagram of the range of the performance angle target area α' of Embodiment B of the present invention on each model coordinate point diagram;

[0049] Figure 5 It is a comparison chart between the optimized prediction model and the 7# prediction model under the conditions of Example B of the present invention. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in combination with the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0051] The method for selecting the optimal shear wave velocity prediction model according to the field demand evaluation of the present invention is to calculate the performance angle index α and the center rate i of each type of shear wave velocity prediction model respectively for multiple types of shear wave velocity prediction models, and to compare and analyze to evaluate the accuracy and rationality of the shear wave velocity prediction model, and select the optimal shear wave velocity prediction model, including the following: Figure 1 The steps shown in the figure are described in detail below.

[0052] S1, determine the performance angle target area α' according to the actual needs, that is, the range of allowable deviation of the model prediction results;

[0053] S2, the in-situ test method is used to carry out tests in multiple boreholes on the site. Each borehole is divided into multiple main layers along the depth direction according to the hole depth, and the basic parameters and shear wave velocity of each main layer are obtained. The basic parameters include the cone tip resistance q t , side wall friction f s ; In order to make the collected data more accurate and reasonable, each main layer can be divided into multiple sub-layers, each main layer is tested for a corresponding shear wave velocity, and each sub-layer is tested for a corresponding set of basic parameters. The basic parameters of all sub-layers in each main layer are averaged, which is the basic parameter corresponding to the shear wave velocity of the main layer. For example: set the sub-layer to 0.05m and the main layer to 1m, collect a set of basic parameter data of static penetration every 0.05m along the depth of each borehole, pause penetration every 1m to start collecting and recovering shear wave velocity parameter data, and take the average of the 20 groups of basic parameters within 1m of each main layer as the basic parameter corresponding to the shear wave velocity of the main layer. The basic parameters in the present invention are not limited to the above-mentioned cone tip resistance q t , side wall friction f s , but is determined based on the parameters required by the prediction model to be evaluated.

[0054] After obtaining the basic parameters and shear wave velocity of each main layer and sublayer, the data is preprocessed first, including the elimination of abnormal points based on mathematical and statistical analysis theory, to obtain corresponding data groups of multiple shear wave velocities and static penetration parameters for model analysis, which are used as data sets for model analysis or construction.

[0055] In-situ testing methods can be static penetration tests or standard penetration tests.

[0056] The in-situ test method in step S1 adopts the following two methods:

[0057] a) Perform a shear wave velocity test on the static penetration test hole at the same time, and use the shear wave velocity test hole data obtained from the shear wave velocity test as the shear wave velocity test value used in subsequent calculations;

[0058] b) If the shear wave velocity test is not carried out simultaneously on the static penetration test hole, the shear wave velocity test hole data of the nearest borehole to the static penetration test hole is selected as the shear wave velocity test hole data for subsequent calculations, and the distance between the nearest borehole and the hole is less than 0.5m.

[0059] S3, according to the basic parameters of each main layer, the derived parameters are interpreted to obtain a one-to-one corresponding shear wave velocity and static penetration parameter data set; the derived parameters include the average friction ratio R f , normalized friction ratio F r , pore pressure parameter B q , Soil Classification Index I c , Normalized cone tip resistance Q tn The derived parameters in the present invention are not limited to the average friction ratio R f , normalized friction ratio F r , pore pressure parameter B q , Soil Classification Index I c , Normalized cone tip resistance Q tn , but is determined based on the parameters required by the prediction model to be evaluated.

[0060] Derived parameters are interpreted as follows:

[0061] Friction ratio R f The expression is:

[0062] R f =f s / q t ×100% (1)

[0063] Normalized friction ratio F r The expression is:

[0064] F r =f s / (q t -σ v0 ) (2)

[0065] In the formula, σ v0 is the overburden pressure on the soil layer;

[0066] Pore ​​pressure parameter B q The expression is:

[0067] B q=(u2-u0) / (q t -σ v0 ) (3)

[0068] Where u0 is the hydrostatic pressure;

[0069] Soil Classification Index I c Proposed by Robertson, the calculation method is as follows:

[0070]

[0071] In the formula, Q tn is the normalized cone tip resistance, which is expressed as:

[0072]

[0073] In the formula, σ′ v0 is the effective overburden pressure; n is the stress index; specifically n = 0.381I c +0.05(σ′ v0 / p a )-0.15≤1; where I c is the soil classification index; p a is the reference pressure, which is atmospheric pressure.

[0074] S4, substitute the static penetration parameter data set of each main layer into various models, and calculate the corresponding shear wave velocity prediction value V under various models. sE , forming the corresponding shear wave velocity prediction value V under various model formulas sE and the measured shear wave velocity V sM Data pair, plotting the predicted shear wave velocity V sE —Measured value of shear wave velocity V sM Draw V on the coordinate point diagram sE =V sM Reference lines.

[0075] S5, the predicted shear wave velocity V obtained above sE and the measured shear wave velocity V sM Data pairs, calculate the performance angle index α of each main layer under each model:

[0076]

[0077] According to the performance angle target area α' given in step S1, the performance angle target area α' is the shear wave velocity prediction value V sE —Measured value of shear wave velocity V sM The coordinate point V on the graph sE =V sMThe reference line is offset to both sides by an angle of α' with the origin of the coordinate system as the center. When the absolute value of the performance angle index α is less than the angle α', it represents the corresponding shear wave velocity prediction value V sE and the measured shear wave velocity V sM The data points of the data pair are within the performance perspective target region α';

[0078] S6. Determine the number of data points within the performance angle target area α' according to step S5, and calculate the ratio of data points that meet the performance angle target area α' under each model, that is, the center rate i:

[0079]

[0080] The central rates i corresponding to the various models are compared and analyzed, and the prediction models with central rates i lower than 90% are screened out, and the remaining prediction models are the better prediction models.

[0081] S7. Calculate the mean deviation angles |α| of all main layers under each model, compare the mean deviation angles |α| of all better prediction models, and select the corresponding model with the smallest mean deviation angle |α| as the optimal prediction model.

[0082] The present invention is applicable to the evaluation and selection of various shear wave velocity parameter prediction models, for example:

[0083] If the existing CPT-V s The correlation model can be used to calculate the corresponding shear wave velocity prediction value under each model, forming the shear wave velocity prediction value V under each model formula. sE and the measured shear wave velocity V sM Corresponding data pairs, the method of the present invention is used according to actual needs in the existing CPT-V s Select the optimal prediction model from the correlation models;

[0084] If you rebuild CPT-V s Then, different types of fitting functions need to be used to fit the CPT-V s , calculate the corresponding shear wave velocity prediction value under each model, and form the shear wave velocity prediction value V under each model formula established sE and the measured shear wave velocity V sM For the corresponding data pair, the method of the present invention is used in combination with the existing CPT-V s The correlation model of CPT-V was evaluated and reconstructed. s The rationality of the correlation model and the reconstruction of CPT-V according to actual needs s The correlation model and the existing CPT-V sSelect the best prediction model from the correlation models, or determine the optimization effect of reconstructing the CPT-Vs correlation model;

[0085] Existing CPT-V s The categories of correlation models mainly include linear, power function, and exponential function. The existing CPT-V s Among the parameters used in the correlation model, the static penetration test uses basic parameters or derived parameters, and the shear wave velocity test uses basic parameters or derived parameters. The parameters in the model may involve one or more parameters.

[0086] The above prediction models can all be used to evaluate and select the optimal prediction model using the present invention.

[0087] The preferred embodiments of the present invention will be described in detail below in combination with example data. The preferred embodiments are only for illustrating the present invention, rather than limiting the protection scope of the present invention.

[0088] The research object is a waste Yellow River floodplain site with non-cohesive soil (sand-silt mixture). The 10 existing CPT-V s Evaluate and select the optimal prediction model among the shear wave velocity prediction models.

[0089] Table 1 Summary of existing shear wave velocity prediction models

[0090]

[0091] Two field requirements of Example A and Example B are formulated respectively, and the most suitable shear wave velocity prediction model is selected for them.

[0092] According to step S1 of the method for selecting the optimal shear wave velocity prediction model according to field demand evaluation of the present invention, the performance angle target area α', i.e., the range of allowable deviation of the model prediction result, is determined according to field demand:

[0093] Embodiment A: The specified performance angle target area α' is -10° to 10°;

[0094] Embodiment B: The performance angle target region α' is specified to be -5° to 5°.

[0095] Summarizing all prediction models #1-10, the parameters that need to be substituted are: q t 、f s , z, σ v0 ,σ′ v0 , B q ,I c 、p a , Q tn 、e.

[0096] Where: q t is the cone tip resistance of pore pressure static penetration, f s is the side wall friction, z is the depth, σ v0 ,σ′ v0 are the total overburden pressure and the effective overburden pressure, B q is the pore pressure parameter, I c is the soil classification index, p a Reference pressure = 100 kPa, Q tn is the normalized cone tip resistance, e is a natural constant, and V calculated in Table 1 s That is the shear wave velocity prediction value V required in the present invention. sE .

[0097] V s1 is the normalized shear wave velocity, which is expressed as follows:

[0098]

[0099] Where V s For the shear wave velocity tested, the #9 and #10 prediction models can be based on V s1 And reverse the above formula to get V s , that is, the predicted shear wave velocity V sE .

[0100] where z and its corresponding q t 、f s is the basic parameter that can be directly measured, σ v0 ,σ′ v0 It can also be obtained directly, p a For reference pressure = 100 kPa, the derived parameter to be interpreted is B q ,I c , Q tn .

[0101] According to step S2 of the method for selecting the optimal prediction model of shear wave velocity according to field demand evaluation of the present invention, a large number of pore pressure static penetration tests (combined shear wave velocity tests) were carried out on the site of the abandoned Yellow River alluvial plain with non-cohesive soil (sand-powder mixture) as the research object. The surface layer of the site is recent artificial miscellaneous fill and cultivated soil, and the main layers below are silt, silt sand, fine sand, and the groundwater level is 3.8m. The test uses an original American multifunctional digital vehicle-mounted seismic wave pore pressure static penetration test system, the components of which are as follows: a static penetration system and a drilling vehicle, which have conventional static penetration, pore pressure and seismic wave test function modules, the probe meets international standards, uses an E4FCS real-time data acquisition computer system, and uses CONEPLOT and CLEANUP as the data processing software, which can directly provide CPTU basic test data and shear wave velocity curves along depth.

[0102] During the test, the probe penetrates evenly at a penetration rate of 20 mm / s, and a set of readings is tested every 0.05 m (5 cm) along the depth. The penetration is stopped every 1 m along the depth, and a seismic wave shear wave velocity test is performed and the shear wave velocity data is collected at the same time, and then the penetration is continued. After each drilling test is completed, the shear wave velocity and basic parameters of static penetration testing, including cone tip resistance q, can be obtained for each borehole along the depth direction within the set penetration depth. c , side wall friction f s After obtaining the basic parameters and shear wave velocity of each main layer and sublayer, the data is preprocessed first, including the removal of abnormal points based on mathematical statistical analysis theory. Specifically, starting from the cone tip resistance, the inaccurate test data caused by thin interlayers are removed. In addition, based on the 3sigma theory principle, the abnormal points outside the range of (mean-3sigma, mean+3sigma) are considered as abnormal points. Combined with the abnormal points in the boxplot box plot, they are finally removed to obtain the corresponding data groups of multiple shear wave velocities and static penetration parameters for model analysis, which are used as the data set for model analysis or construction.

[0103] Obtain the drilling depth and its corresponding main layer basic parameter data set and the shear wave velocity measured value V sM In order to make the data more accurate, each main layer is divided into multiple sub-layers for basic parameter data measurement, and the average value is taken. The surface fill soil is within 1m of the test site, and the shear wave velocity test value is an abnormal value and needs to be eliminated. The drilling depth and its corresponding main layer basic parameter data group are obtained as shown in Table 2.

[0104] Table 2 Drilling depth and its corresponding main layer basic parameter data set

[0105] Z(m) <![CDATA[q t (MPa)]]> <![CDATA[f s (kPa)]]> <![CDATA[σ v0 (kPa)]]> <![CDATA[σ′ v0 (kPa)]]> <![CDATA[V sM (m / s)]]> 2 3.98 35.3548 35.3 35.3 174.39 3 1.89 17.28 53.13 47.735 156.82 4 1.73 15.05 70.70 55.51 166.85 5 1.55 10.94 88.35 63.36 169.29 6 5.7 66.82 106.33 71.535 212.11 7 3.54 43.60 125.10 80.51 224.15 8 4.93 56.42 143.53 89.135 223.13 9 7.76 101.53 161.53 97.335 230.38 10 5.68 87.18 179.53 105.535 221.41 11 5.67 64.01 197.53 113.735 196.04 12 11.13 134.89 215.53 121.935 250.03 13 6.14 73.70 233.53 130.135 212.64 14 7.96 131.06 251.53 138.335 234.2 15 13.32 191.56 269.53 146.535 264.31 16 7.3 98.19 287.53 154.735 230.13 17 8.35 124.02 305.53 162.935 238.8

[0106] According to step S3 of the method for selecting the optimal prediction model of shear wave velocity according to field demand evaluation of the present invention, the derived parameters required are interpreted according to the basic parameters of each main layer;

[0107] According to the above prediction models, the required derived parameters are selected as B q ,I c , Q tn According to the given derived parameter calculation formula, the derived parameter data group is calculated as shown in Table 3.

[0108] Table 3 Drilling depth and its corresponding main layer derived parameter data set

[0109]

[0110]

[0111] According to step S4 of the method for selecting the optimal shear wave velocity prediction model according to the field demand evaluation of the present invention, according to the shear wave velocity prediction model formula in Table 1 above, each corresponding shear wave velocity prediction value under each model can be calculated to form the shear wave velocity prediction value V of each main layer under each model formula. sE and the measured shear wave velocity V sM The corresponding data pairs are shown in Table 4.

[0112] Table 4 V of each prediction model sE 、V sM Data pair

[0113]

[0114] Plot the predicted shear wave velocity V for each model sE —Measured value of shear wave velocity V sM The coordinate point diagram is as follows Figure 2 As shown, the deviation of the data pair can be observed intuitively on the coordinate point diagram, and V can be drawn on the point diagram. sE =V sM Reference lines.

[0115] According to step S5 of the method for selecting the optimal prediction model of shear wave velocity according to field demand evaluation of the present invention, the shear wave velocity prediction value V obtained above is sE and the measured shear wave velocity V sM Data pairs, calculate the performance angle index α of each main layer under each model:

[0116]

[0117] The final results are shown in Table 5:

[0118] Table 5 Performance angle index of each main layer of each prediction model (unit: °)

[0119] Z(m) #1 #2 #3 #4 #5 #6 #7 #8 #9 #10 2 -3.40 4.26 5.01 9.91 24.42 10.82 1.12 3.92 8.88 8.86 3 -8.82 4.93 4.86 15.34 27.71 2.71 -1.03 12.13 15.47 15.77 4 -4.38 1.49 1.44 10.84 24.14 0.77 -2.96 7.94 11.37 12.35 5 -6.59 6.63 6.35 12.66 25.48 2.83 -3.89 10.85 15.13 16.63 6 -7.33 2.97 4.72 9.92 21.43 10.28 0.87 2.71 1.41 5.85 7 -10.77 1.42 0.34 13.23 23.78 2.29 -4.19 7.47 6.41 10.39 8 -9.40 0.44 1.92 10.97 21.90 6.98 -1.15 4.57 3.19 8.56 9 -7.86 3.11 5.66 9.07 19.64 12.07 1.96 0.73 3.34 3.35 10 -8.53 3.36 5.76 9.22 19.28 9.07 0.92 1.04 3.63 2.66 11 -5.16 4.93 6.58 5.49 16.87 12.31 4.37 1.28 3.42 3.73 12 -7.48 2.36 5.31 8.31 18.59 14.47 3.08 0.45 5.56 3.22 13 -7.02 3.50 5.43 6.66 17.37 11.15 3.12 0.53 3.56 4.40 14 -8.15 4.07 7.31 7.66 17.07 11.97 2.86 1.76 8.29 0.04 15 -4.04 6.03 9.74 4.61 14.47 18.13 3.59 5.48 12.78 3.28 16 -8.21 2.95 5.47 7.08 16.99 11.23 2.82 1.07 5.63 3.44 17 -8.37 3.21 6.26 7.13 16.60 12.00 3.06 1.86 7.67 1.77

[0120] According to the performance angle target area α' given in step S1, the performance angle target area α' is the shear wave velocity prediction value V sE —Measured value of shear wave velocity V sM The coordinate point V on the graph sE =V sM The reference line is offset to both sides by an angle of α' with the origin of the coordinate system as the center. When the absolute value of the performance angle index α is less than the angle α', it represents the corresponding shear wave velocity prediction value V sE and the measured shear wave velocity V sM The data points of the data pair are within the performance angle target region α';

[0121] The following examples A and B are given to determine the difference of the target area α' of the performance angle according to the actual needs:

[0122] Example A: The performance angle target area α' is set to -10° to 10°, that is, V sE =V sM The reference line is the area formed by shifting 10° to both sides with the coordinate origin as the center, such as Figure 3 It can be seen intuitively as shown;

[0123] Embodiment B: The performance angle target area α' is set to -5° to 5°, that is, V sE =V sM The reference line is the area formed by shifting 5° to both sides with the coordinate origin as the center, such as Figure 4 It can be seen intuitively as shown;

[0124] According to step S6 of the method for selecting the optimal prediction model of shear wave velocity according to field demand evaluation of the present invention, the number of data points in the performance angle target area α' is determined according to step S5, and the ratio of data points satisfying the performance angle target area α' under each model is calculated, that is, the center rate i:

[0125]

[0126] The number of data points and the center rate i statistics in each model in Examples A and B within the performance angle target area α' are shown in Table 6:

[0127] Table 6 Number of data points and center rate i in the performance angle target area α' for each prediction model

[0128]

[0129] The central rates i corresponding to the various models are compared and analyzed, and the prediction models with central rates i lower than 90% are screened out, and the remaining prediction models are the better prediction models;

[0130] From the above, we can see that:

[0131] Under the requirement that the performance angle target area α' of Example A is set to -10° to 10°, prediction models #4, #5, #6, #9, and #10 with a center rate i lower than 90% are screened out, and the remaining prediction models #1, #2, #3, #7, and #8 are better prediction models;

[0132] In Example B, when the performance angle target area α' is set to -5° to 5°, prediction models #1, #3, #4, #5, #6, #8, #9, and #10 with center rates i lower than 90% are screened out, and the remaining prediction models #2 and #7 are preferred prediction models.

[0133] According to step S7 of the method for selecting the optimal prediction model of shear wave velocity according to field demand evaluation of the present invention, the mean values ​​of the deviation angles |α| of all main layers under each model are calculated, as shown in Table 6, the mean values ​​of the deviation angles |α| of all the better prediction models are compared, and the corresponding model with the smallest mean value of the deviation angles |α| is selected as the optimal prediction model;

[0134] From the above, we can see that:

[0135] In Example A, among the better prediction models #1, #2, #3, #7, and #8, the one with the smallest mean value of the deviation angle |α| is the prediction model #7, so the prediction model #7 is the optimal prediction model required by Example A;

[0136] In Example B, among the better prediction models #2 and #7, the one with the smallest mean value of the deviation angle |α| is the prediction model #7, so #7 is the optimal prediction model required by Example B.

[0137] If the site geological conditions are highly specific, it is necessary to rebuild the CPT-V s The correlation model is then used for prediction. A variety of relatively applicable prediction models can be selected. After the optimal model is selected by the above method, the coefficients of the prediction model are optimized by the actual measured parameter data, and then a new prediction model is reconstructed. The method provided by the present invention is used for comparative evaluation to determine whether the reconstructed prediction model is better than the optimal prediction model. If it is better, it can be applied to the site. Taking the above embodiment B as an example, the optimal prediction model #7 is optimized by the given basic parameters and derived parameter data to reconstruct a new optimized prediction model:

[0138] V s =4.384q t 0.192 I c 0.234 z 0.0596

[0139] There are many ways and methods for optimizing the prediction model in the prior art, but the present invention only evaluates and selects the model, so the steps and methods for optimizing the model are not described and limited in detail here.

[0140] The shear wave velocity prediction value V obtained according to the optimized prediction model sE and the measured shear wave velocity V sM The corresponding data pairs are shown in Table 7:

[0141] Table 7 V of optimized prediction model sE 、V sM Data pair

[0142]

[0143]

[0144] Plot the shear wave velocity prediction value V of the optimized prediction model sE —Measured value of shear wave velocity V sM Then compare the coordinate point diagram with the prediction model 7# Figure 5 As shown, the performance angle index α of each main layer of the optimization prediction model is calculated as shown in Table 8:

[0145] Table 8 Performance angle index α of each main layer of the optimized prediction model

[0146]

[0147] According to Table 8, when the performance angle target area α' of Example B is set to -5°~5°, the center rate i is 100%, which meets the requirement that the center rate i is greater than 90%, and the average value of the deviation angle |α| is less than 7#, which means that when the actual demand performance angle target area α' is set to -5°~5°, the comprehensive evaluation optimization prediction model is better than the 7# prediction model. When the performance angle target area α' is set to -5°~5°, the optimization prediction model and the 7# prediction model are better than the 7# prediction model. Figure 5 The distribution of data points can be seen intuitively.

[0148] The mean of the deviation angle |α| may also be the mean of |sinα| (mean of the sine of the deviation angle) or the mean of |tanα| (mean of the tangent of the deviation angle). The mean of |sinα| is also counted in Tables 6 and 8. In step S7, the corresponding model with the smallest mean of |sinα| or |tanα| is selected as the optimal prediction model, which is beneficial for use in engineering calculations.

[0149] 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 method for selecting an optimal shear wave velocity prediction model based on field demand evaluation, characterized in that: The performance angle target area α' is determined according to the actual needs. For multiple types of shear wave velocity prediction models, the mean values ​​of the center rate i and deviation angle |α| of each type of model are calculated respectively. The accuracy and rationality of the shear wave velocity prediction model are evaluated by comparative analysis, and the optimal shear wave velocity prediction model is selected. The specific steps are as follows: S1, determine the performance angle target area α' according to the actual needs, that is, the range of allowable deviation of the model prediction results; S2, using the in-situ test method to carry out tests in multiple boreholes on the site, each borehole is divided into multiple main layers along the depth direction according to the hole depth, and the basic parameters and shear wave velocity of each main layer are obtained; S3, interpreting derived parameters according to the basic parameters of each main layer to obtain a one-to-one corresponding shear wave velocity and static penetration parameter data set; S4, substitute the static penetration parameter data set of each main layer into various models, and calculate the corresponding shear wave velocity prediction value V under various models. sE , forming the corresponding shear wave velocity prediction value V under various model formulas sE and the measured shear wave velocity V sM Data pair, plotting the predicted shear wave velocity V sE —Measured value of shear wave velocity V sM Draw V on the coordinate point diagram sE =V sM Reference lines; S5, the predicted shear wave velocity V obtained above sE and the measured shear wave velocity V sM Data pairs, calculate the performance angle index α of each main layer under each model: According to the performance angle target area α' given in step S1, the performance angle target area α' is the shear wave velocity prediction value V sE —Measured value of shear wave velocity V sM The coordinate point V on the graph sE =V sM The reference line is offset to both sides by an angle of α' with the origin of the coordinate system as the center. When the absolute value of the performance angle index α is less than the angle α', it represents the corresponding shear wave velocity prediction value V sE and the measured shear wave velocity V sM The data points of the data pair are within the performance angle target region α'; S6. Determine the number of data points within the performance angle target area α' according to step S5, and calculate the ratio of data points that meet the performance angle target area α' under each model, that is, the center rate i: The central rates i corresponding to the various models are compared and analyzed, and the prediction models with central rates i lower than 90% are screened out, and the remaining prediction models are the better prediction models; S7. Calculate the mean deviation angles |α| of all main layers under each model, compare the mean deviation angles |α| of all better prediction models, and select the corresponding model with the smallest mean deviation angle |α| as the optimal prediction model.

2. The method for selecting the optimal shear wave velocity prediction model according to field demand evaluation according to claim 1, characterized in that: The in-situ test method in step S1 is a static penetration test or a standard penetration test.

3. The method for selecting the optimal shear wave velocity prediction model according to field demand evaluation according to claim 2, characterized in that: The static penetration test in step S1 adopts the following two methods: a) Perform a shear wave velocity test on the static penetration test hole at the same time, and use the shear wave velocity test hole data obtained from the shear wave velocity test as the shear wave velocity test value used in subsequent calculations; b) If the shear wave velocity test is not carried out simultaneously on the static penetration test hole, the shear wave velocity test hole data of the nearest borehole to the static penetration test hole is selected as the shear wave velocity test hole data for subsequent calculations, and the distance between the nearest borehole and the hole is less than 0.5m.

4. The method for selecting the optimal shear wave velocity prediction model according to field demand evaluation according to claim 1, characterized in that: In step S1, each borehole is divided into a plurality of main layers according to the hole depth in the depth direction, and each main layer is further divided into a plurality of sub-layers. Each main layer is tested for a corresponding shear wave velocity, and each sub-layer is tested for a corresponding set of basic parameters. In step S2, the basic parameters of all sub-layers in each main layer are averaged, which is the basic parameter corresponding to the shear wave velocity of the main layer.

5. The method for selecting the optimal shear wave velocity prediction model according to field demand evaluation according to claim 4, characterized in that: The in-situ test method in step S1 is specifically as follows: The sublayer is set to 0.05m and the main layer to 1m. A set of basic parameter data of static penetration is collected every 0.05m along the depth of each borehole. The penetration is paused every 1m to start collecting and recovering shear wave velocity parameter data.

6. The method for selecting the optimal shear wave velocity prediction model according to field demand evaluation according to claim 1, characterized in that: The basic parameters in step S1 include the cone tip resistance q t , side wall friction f s .

7. The method for selecting the optimal shear wave velocity prediction model according to field demand evaluation according to claim 1, characterized in that: The derived parameters in step S2 include the average friction ratio R f , normalized friction ratio F r , pore pressure parameter B q , Soil Classification Index I c , Normalized cone tip resistance Q tn .

8. The method for selecting the optimal shear wave velocity prediction model according to field demand evaluation according to claim 7, characterized in that: The specific interpretation of the derived parameters in step S2 is as follows: Friction ratio R f The expression is: R f =f s / q t ×100% (1) Normalized friction ratio F r The expression is: F r =f s / (q t -σ v0 ) (2) In the formula, σ v0 is the overburden pressure on the soil layer; Pore ​​pressure parameter B q The expression is: B q =(u2-u0) / (q t -s v0 ) (3) Where u0 is the hydrostatic pressure; Soil Classification Index I c The calculation method is as follows: In the formula, Q tn is the normalized cone tip resistance, which is expressed as: In the formula, σ′ v0 is the effective overburden pressure; n is the stress index; Specifically, n = 0.381I c +0.05(σ′ v0 / p a )-0.15≤1; where I c is the soil classification index; p a is the reference pressure, which is atmospheric pressure.

9. The method for selecting the optimal shear wave velocity prediction model according to field demand evaluation according to claim 1 or 4, characterized in that: After the basic parameters and shear wave velocity of each main layer are obtained in step S1, the data is preprocessed first, including eliminating abnormal points based on mathematical statistical analysis theory, and obtaining corresponding data groups of multiple shear wave velocities and static penetration parameters for model analysis as data sets for model analysis or construction.

10. The method for selecting the optimal shear wave velocity prediction model according to field demand evaluation according to claim 1, characterized in that: The mean of the deviation angle |α| may also be the mean of |sinα| or the mean of |tanα|. In step S7, the corresponding model with the smallest mean of |sinα| or |tanα| is selected as the optimal prediction model.

Images