Construction project site category discrimination method and system

Abstract

The invention discloses a three-dimensional geological information model-based construction project site category intelligent discrimination method and system, and the method comprises the steps: constructing a three-dimensional geological model through multi-source data fusion, automatically calculating the thickness of a covering layer and an equivalent shear wave velocity, and carrying out the automatic calculation of the thickness of a covering layer through combining a standard discrimination rule and a machine learning algorithm. And precise division and visual output of site categories are realized. According to the method, the problem of weak spatial relevance caused by replacing surfaces with points in a traditional method is solved, and the efficiency and precision of aseismic design are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of construction site classification discrimination methods, and in particular to a large-scale earthquake-resistant site classification automatic discrimination method and system based on a three-dimensional geological information model. Background Art

[0002] In the field of construction engineering, earthquake-resistant protection is a vital step in safeguarding the safety of life and property, and the accurate identification of earthquake-resistant site categories is the logical starting point of the entire earthquake-resistant design system. According to the current "Code for Seismic Design of Buildings" (GB50011-2010, 2016 edition), site classification requires comprehensive consideration of two key parameters: equivalent shear wave velocity and cover thickness. This directly determines the value of seismic motion parameters and the selection of structural earthquake-resistant measures. From an engineering practice perspective, the accuracy of site classification not only affects the seismic safety of building structures, but also affects the full life cycle cost of construction projects. Incorrect classification may lead to overly conservative earthquake-resistant designs or insufficient safety reserves. The former results in a waste of resources, while the latter poses a safety hazard.

[0003] The current manual identification method widely used in the engineering community involves obtaining stratigraphic data at discrete points through drilling, calculating equivalent shear wave velocity based on single-hole shear wave velocity test results, and ultimately classifying sites based on overburden thickness. This "point data" analysis model faces three technical barriers:

[0004] Inadequate spatial representation: Analytical units based on borehole locations make it difficult to construct a continuous site engineering geological model, resulting in significant "point-like discreteness" in the identification results. In complex geological conditions (such as karst areas and near fault zones), sudden changes in stratum properties can occur tens of meters between adjacent boreholes, and traditional methods are unable to capture this spatial heterogeneity.

[0005] Lack of 3D information: Current standards rely on 2D cross-sectional analysis, which essentially simplifies 3D geological space into a collection of vertical 1D data. This ignores the impact of 3D spatial factors such as stratum dip, lens distribution, and interlayer structures on shear wave propagation paths. Measured data from a highway project showed that when stratum dip exceeds 15°, the equivalent shear wave velocity calculated using traditional methods can deviate by 12%-15% from the actual 3D propagation model.

[0006] Delayed response to abnormal geology: For deeply buried local geological anomalies (such as ancient river channel fillings and weak interlayers), the probability of single-hole exposure is less than 30%, and they are often discovered only during the construction phase, forcing the seismic design plan to be reworked.

[0007] The rapid development of 3D geological modeling, BIM, and big data technologies is increasingly being used in engineering, providing new approaches and technical solutions to these problems. A large-scale automatic seismic site classification method based on 3D geological information modeling has emerged, promising to significantly improve the accuracy and efficiency of engineering construction. Summary of the Invention

[0008] The purpose of the present invention is to provide a large-scale seismic site classification automatic discrimination method and system based on three-dimensional geological information model technology to solve the problems of traditional discrimination methods such as using points instead of surfaces, weak spatial correlation, inability to reflect three-dimensional spatial distribution characteristics, and easy neglect of the impact of local geological anomalies. The present invention realizes the automation and intelligence of the site classification process, improves the discrimination accuracy and efficiency, and makes the discrimination results intuitive and visual.

[0009] In order to achieve the above object, the present invention provides a method for distinguishing the type of construction site, which is characterized by comprising the following steps:

[0010] S1: Collect data on the project site's topography, geological drilling, geophysical data (surface wave exploration and inversion of shear wave velocity), borehole shear wave tests, and historical seismic motion records;

[0011] S2, based on the data, conduct multi-source data fusion modeling to construct a continuous three-dimensional terrain surface model, a geological structure model, a cover layer thickness distribution model H(x,y,z) and a shear wave velocity model Vs(x,y,z);

[0012] S3, using the Geological Information Model (GIM) standard, assigns unique identification attribute values ​​to each 3D geological unit, and associates attributes such as shear wave velocity and cover layer distribution thickness;

[0013] S4, based on the three-dimensional shear wave velocity model, vertically search for the first stratum interface whose shear wave velocity Vs is not less than a first threshold, calculate the vertical distance H(P) from the surface to the interface, and obtain the output cover layer thickness H according to the built-in algorithm rules;

[0014] S5, for any evaluation grid node P (x, y), extract the thickness d of each soil layer within the calculation depth d0 along the vertical z i and shear wave velocity Calculate the equivalent shear wave velocity v according to the standard formula se ;

[0015] S6, according to the rules of the discriminant table, establish (v se ,H) 2D discriminant matrix, automatically calculates and matches I0, I1, II, III, and IV site categories;

[0016] S7, introduce random forest or support vector machine (SVM) algorithms, use historical earthquake damage data to train models, and correct discrimination bias under complex geological conditions;

[0017] S8, outputting a three-dimensional site category distribution model or a two-dimensional zoning map including site category attributes.

[0018] Furthermore, the multi-source data fusion modeling adopts Kriging interpolation method or neural network algorithm to construct a continuous three-dimensional terrain surface model, geological structure model, cover layer thickness distribution model H(x,y,z) and shear wave velocity model Vs(x,y,z).

[0019] Furthermore, the first threshold is 500 m / s.

[0020] Furthermore, in step S4, the thickness H of the covering layer should generally be determined according to the distance from the ground to the top surface of the soil layer with a shear wave velocity greater than 500 m / s and the shear wave velocity of each underlying rock and soil layer greater than 500 m / s; when there is a soil layer 5 m below the ground with a shear wave velocity greater than 2.5 times the shear wave velocity of each upper soil layer, and the shear wave velocity of this layer and each underlying rock and soil layer is greater than 400 m / s, it can be determined according to the distance from the ground to the top surface of the soil layer; isolated boulders and lenses with a shear wave velocity greater than 500 m / s should be regarded as the surrounding soil layers; volcanic rock hard interlayers in the soil layer should be regarded as rigid bodies, and their thickness should be deducted from the covering soil layer.

[0021] Furthermore, the categories of I0, I1, II, III and IV sites are as follows:

[0022]

[0023] Furthermore, the construction site classification method according to claim 1 is characterized in that, when calculating the equivalent shear wave velocity v se When , the standard formula is

[0024] Furthermore, it is characterized in that the cover layer thickness distribution model H(x, y, z) can be at least one of a linear inclined layer model, a Gaussian distribution model, an exponential decay model, a composite model, or a terrain-based model;

[0025] Linear inclined layer model: This model can be used when the overburden is uniformly inclined: H(x,y,z) = H0 + kx·x + ky·y, where H0 represents the base thickness, kx and ky are the slopes in the x and y directions, respectively.

[0026] Gaussian distribution model: This model is suitable for simulating the covering layer formed by single point source deposition.

[0027]

[0028] Among them, H max Maximum thickness, (x0, y0) is the center point coordinate, σ x and σ y is the standard deviation of the distribution;

[0029] Exponential decay model: This model is more appropriate when simulating cover characteristics that vary with depth;

[0030] H(x,y,z)=H0·e- α·z

[0031] Where H0 represents the reference thickness and α is the attenuation coefficient;

[0032] Composite model: This model can be used if you want to take multiple factors into consideration;

[0033]

[0034] A represents the fluctuation amplitude, H0 represents the reference thickness, λx and λy are the fluctuation wavelengths, and β is the depth attenuation coefficient;

[0035] Terrain-based model: This model can be used when the thickness of the cover is related to the terrain;

[0036] Where H0 represents the reference thickness, is the terrain gradient, and γ is the correlation coefficient.

[0037] Furthermore, it is characterized in that the shear wave velocity model Vs(x,y,z) can be Vs(x,y,z)=V0·[1+A·sin(ω0·x)·cos(ω y ·y)]·[1+B·f(x,y,z)], where V0 is the initial shear wave velocity (m / s), A is the lateral variation amplitude coefficient (dimensionless), ω x ,ω y is the lateral variation angular frequency (1 / m), B is the random disturbance intensity coefficient (dimensionless), and f(x,y,z) is the spatial random field function. x ·x)·cos(ω y y) simulates the directional changes of geological structures. The model f(x,y,z) takes into account local geological anomalies and can be implemented using a Gaussian random field.

[0038] Another aspect of the present invention provides a system for implementing the above method, characterized in that it includes:

[0039] Data fusion modeling module, used to import multi-source geological data and fuse them to build a 3D model;

[0040] Parameter calculation module, used to automatically calculate equivalent shear wave velocity and cover thickness;

[0041] Intelligent identification module, integrating standard rule algorithms to output site categories;

[0042] The machine learning module is used to introduce intelligent algorithms to correct discrimination bias under complex geological conditions; the visualization module is used to generate three-dimensional models and two-dimensional GIS zoning maps.

[0043] Furthermore, the multi-source geological data that can be imported by the data fusion modeling module include topographic data, geological drilling data, geophysical data (surface wave exploration inversion shear wave velocity), borehole shear wave test data and historical seismic record data.

[0044] Furthermore, the three-dimensional model generated by the visualization module can intuitively display the distribution characteristics of the site categories in three-dimensional space, and the two-dimensional GIS zoning map can display the planar zoning of the site categories.

[0045] The present invention has at least the following beneficial effects: the patent of the present invention makes full use of BIM technology, realizes the deep coupling of three-dimensional geological information model and standard algorithm for the first time, realizes the automation of the whole process from data acquisition to result output, breaks through the traditional two-dimensional data processing mode, and converts the seismic discrimination parameter (v se , H) is converted into a three-dimensional continuous variable in space to quantify the impact of geological spatial heterogeneity on site classification. By integrating multi-source geological data to construct a three-dimensional shear wave velocity and cover layer thickness model, the equivalent shear wave velocity and cover layer thickness are automatically calculated, and intelligent site classification is achieved based on seismic standards, and finally two- and three-dimensional visualization results are output. The patent quantifies the spatial heterogeneity of geological parameters based on a continuous three-dimensional model, which can directly reflect the three-dimensional spatial distribution characteristics, and introduces a machine learning algorithm to optimize the discrimination model, taking into account the impact of local geological anomalies on site classification, significantly improving the discrimination accuracy. It has the advantages of automated and easy-to-implement operation process, intuitive and visual discrimination results, precise analysis results, and high discrimination efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is an implementation flow chart of the construction site classification identification method. DETAILED DESCRIPTION

[0047] The following embodiments of the present invention are described in further detail with reference to the accompanying drawings and examples. The following examples are used to illustrate the present invention but are not intended to limit the scope of the present invention.

[0048] It will be understood that when used in this specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and / or components, but does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or collections.

[0049] To simplify the drawings, only the parts relevant to the present invention are schematically shown in each figure. They do not represent the actual structure of the product. Furthermore, to simplify the drawings and facilitate understanding, in some figures, only one of the components with the same structure or function is schematically depicted or labeled. As used herein, "one" not only means "only one" but also "more than one."

[0050] It should be further understood that the term "and / or" used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0051] In the embodiments shown in the accompanying drawings, directional indications (such as up, down, left, right, front, and rear) used to explain the structure and movement of various components of the present invention are not absolute but relative. These descriptions are applicable when the components are in the positions shown in the accompanying drawings. If the descriptions of the positions of these components are changed, the directional indications will also change accordingly.

[0052] In addition, in the description of the present application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0053] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the specific embodiments of the present invention will be described below with reference to the accompanying drawings. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings and other embodiments can be obtained based on these drawings without inventive work.

[0054] Figure 1 This is a flowchart of the implementation of the construction site classification method of this embodiment. The specific process is:

[0055] 1. Data collection (step S1)

[0056] Collect topographic data of the project site, including topographic contour maps and topographic survey data, to accurately reflect the topographic and geomorphological characteristics of the site. Collect geological drilling data, including the location, depth, and soil stratification of the boreholes, and obtain the geological structure information of the site through data from multiple boreholes. Collect geophysical data and use surface wave exploration technology to obtain shear wave velocity inversion results to provide data support for building a shear wave velocity model. Collect borehole shear wave test data and obtain the shear wave velocity values ​​of each soil layer by performing shear wave velocity tests on the boreholes. Collect historical seismic motion records, including the magnitude, epicenter location, and seismic motion parameters of earthquakes that have occurred in the area, for subsequent machine learning model training and correction of discrimination bias.

[0057] 2. Multi-source data fusion modeling (step S2)

[0058] Using the collected topographic data, geological borehole data, geophysical data, and borehole shear wave test data, we used Kriging interpolation to construct a continuous three-dimensional terrain surface model, a geological structure model, an overburden thickness distribution model H(x,y,z), and a shear wave velocity model Vs(x,y,z). For example, for the shear wave velocity model, we interpolated the shear wave velocity data from each borehole and geophysical point using Kriging to obtain a continuous shear wave velocity distribution for the entire site.

[0059] 3. Assign attribute values ​​(step S3)

[0060] According to the Geological Information Model (GIM) standard, each 3D geological unit is assigned a unique identification attribute value, such as GIM_001 or GIM_002, and attributes such as shear wave velocity and cover thickness are associated with the corresponding geological unit. This approach enables structured management of geological data, facilitating subsequent query and analysis.

[0061] 4. Automatic identification of cover thickness (step S4)

[0062] Based on the constructed 3D shear wave velocity model, a vertical search is performed for each geological unit to find the first stratigraphic interface with a shear wave velocity of 500 m / s or greater. For example, at an assessment point P(x,y), the search begins from the surface and downward. When a stratigraphic interface with a shear wave velocity of 500 m / s is encountered, the vertical distance H(P) from the surface to that interface is calculated. The overburden thickness H at that point is then determined based on the provisions of the Code for Seismic Design of Buildings (GB50011) and the built-in algorithm. The thickness H of the covering layer should generally be determined according to the distance from the ground to the top surface of the soil layer with a shear wave velocity greater than 500m / s and the shear wave velocity of the underlying rock and soil layers greater than 500m / s; when there is a soil layer 5m below the ground with a shear wave velocity greater than 2.5 times the shear wave velocity of the soil layers above it, and the shear wave velocity of this layer and the underlying rock and soil layers are all greater than 400m / s, it can be determined according to the distance from the ground to the top surface of this soil layer; isolated rocks and lenses with a shear wave velocity greater than 500m / s should be regarded as the surrounding soil layers; hard interlayers of volcanic rocks in the soil layer should be regarded as rigid bodies, and their thickness should be deducted from the covering soil layer.

[0063] 5. Automatic calculation of equivalent shear wave velocity (step S5)

[0064] For any evaluation grid node P(x,y), determine the calculation depth d0. Extract the thickness d of each soil layer within the calculation depth d0 along the vertical z direction. i and shear wave velocity According to the standard formula Calculate the equivalent shear wave velocity v se

[0065] 6. Site Category Identification (Step S6)

[0066] According to the GB50011 discrimination table rules, establish (v se , H) Two-dimensional discriminant matrix.

[0067]

[0068] Assume that the equivalent shear wave velocity v at the evaluation point is se =197.37m / s, the thickness of the covering layer H = 15m. By querying the discriminant matrix, it is determined that the site category of this point is Class II site.

[0069] 7. Correction of discrimination deviation (step S7)

[0070] A random forest algorithm was introduced, and the model was trained using collected historical earthquake damage data. For example, historical earthquake damage cases under similar geological conditions in the region and surrounding areas were selected, and the site classification results were compared with the actual earthquake damage to construct a training dataset. By training the random forest model, the patterns of site classification deviation under complex geological conditions were learned, thereby correcting the discrimination results. For complex geological areas in this embodiment, such as those with interlayers, the accuracy of the discrimination results was improved through model correction.

[0071] 8. Result output (step S8)

[0072] The visualization module generates a 3D site classification distribution model and a 2D zoning map for the project site. The 3D model allows for a visual representation of the spatial distribution of different site classifications, such as the range and location of Class I and Class II sites. The 2D zoning map provides a flat representation of the site classifications, facilitating planning and design.

[0073] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A construction site classification method, characterized in that: The following steps are involved: S1: Collect data on the project site's topography, geological drilling, geophysical data (surface wave exploration and inversion of shear wave velocity), borehole shear wave tests, and historical seismic motion records; S2, based on the data, conduct multi-source data fusion modeling to construct a continuous three-dimensional terrain surface model, a geological structure model, a cover layer thickness distribution model H(x,y,z) and a shear wave velocity model Vs(x,y,z); S3, using the Geological Information Model (GIM) standard, assigns unique identification attribute values ​​to each 3D geological unit, and associates attributes such as shear wave velocity and cover layer distribution thickness; S4, based on the three-dimensional shear wave velocity model, vertically search for the first stratum interface whose shear wave velocity Vs is not less than a first threshold, calculate the vertical distance H(P) from the surface to the interface, and obtain the output cover layer thickness H according to the built-in algorithm rules; S5, for any evaluation grid node P (x, y), extract the thickness di and shear wave velocity of each soil layer within the calculation depth d0 along the vertical z Calculate the equivalent shear wave velocity v according to the standard formula se ; S6, according to the rules of the discriminant table, establish (v se ,H) 2D discriminant matrix, automatically calculates and matches I0, I1, II, III, and IV site categories; S7, introduce random forest or support vector machine (SVM) algorithms, use historical earthquake damage data to train models, and correct discrimination bias under complex geological conditions; S8, outputting a three-dimensional site category distribution model or a two-dimensional zoning map including site category attributes.

2. The method according to claim 1, characterized in that The multi-source data fusion modeling adopts Kriging interpolation method or neural network algorithm to construct a continuous three-dimensional terrain surface model, a geological structure model, a cover layer thickness distribution model H(x,y,z) and a shear wave velocity model Vs(x,y,z).

3. The method according to claim 1, characterized in that The first threshold is 500 m / s.

4. The construction site classification method according to claim 1, characterized in that: In step S4, the thickness H of the cover layer should generally be determined based on the distance from the ground to the top surface of a soil layer with a shear wave velocity greater than 500 m / s and the shear wave velocity of each underlying rock and soil layer greater than 500 m / s. When there is a soil layer 5 m below the ground with a shear wave velocity greater than 2.5 times the shear wave velocity of each upper soil layer, and the shear wave velocity of this layer and each underlying rock and soil layer is greater than 400 m / s, the thickness H of the cover layer can be determined based on the distance from the ground to the top surface of the soil layer. Boulders and lenses with a shear wave velocity greater than 500 m / s should be treated as the surrounding soil layers. Hard interlayers of volcanic rock in the soil layer should be treated as rigid bodies, and their thickness should be deducted from the cover soil layer.

5. The construction site classification method according to claim 1, characterized in that: In calculating the equivalent shear wave velocity v se When , the standard formula is 6. The construction site classification method according to claim 1, characterized in that: The overburden thickness distribution model H(x, y, z) may be at least one of a linear inclined layer model, a Gaussian distribution model, an exponential decay model, a composite model, or a terrain-based model; Linear Sloping Layer Model: H(x,y,z)=H0+kx·x+ky·y, where H0 represents the base thickness, kx and ky are the slopes in the x and y directions, respectively; Gaussian distribution model: Among them, H max Maximum thickness, (x0, y0) is the center point coordinate, σ x and σ y is the standard deviation of the distribution; Exponential decay model: H(x,y,z)=H0·e -α·z Where H0 represents the reference thickness and α is the attenuation coefficient; Composite Model: A represents the fluctuation amplitude, H0 represents the reference thickness, λx and λy are the fluctuation wavelengths, and β is the depth attenuation coefficient; Terrain-based models: Where H0 represents the base thickness, is the terrain gradient, and γ is the correlation coefficient.

7. The construction site classification method according to claim 1, characterized in that: The shear wave velocity model Vs(x,y,z) can be: Vs(x,y,z)=V0·[1+A·sin(ω x ·x)·cos(ω y ·y)]·[1+B·f(x,y,z)], where V0 is the initial Initial shear wave velocity, in m / s, A is the lateral variation amplitude coefficient, ω x ,ω y is the lateral variation angular frequency, B is the random disturbance intensity coefficient, and f(x, y, z) is the spatial random field function.

8. A system for implementing the method according to any one of claims 1 to 7, characterized in that: include: Data fusion modeling module, used to import multi-source geological data and fuse them to build a 3D model; Parameter calculation module, used to automatically calculate equivalent shear wave velocity and cover thickness; Intelligent identification module, integrating standard rule algorithms to output site categories; Machine learning module, used to introduce intelligent algorithms to correct discrimination bias under complex geological conditions; Visualization module, used to generate three-dimensional models and two-dimensional GIS zoning maps.

9. The system according to claim 8, characterized in that The multi-source geological data that can be imported into the data fusion modeling module include topographic data, geological drilling data, geophysical data, drilling shear wave test data and historical seismic record data.

10. The system according to claim 8, wherein: The three-dimensional model generated by the visualization module can intuitively display the distribution characteristics of site categories in three-dimensional space, and the two-dimensional GIS zoning map can display the planar zoning of site categories.

Images