Method and system for constructing three-dimensional geological model based on multi-source heterogeneous geological data

Abstract

The application provides a three-dimensional geological model construction method and system based on multi-source heterogeneous geological data, and belongs to the technical field of geological data processing and three-dimensional modeling. The application obtains stratum age data, contact type data and sedimentary facies evolution rate of each drilling point, performs numerical mapping on the age data and the contact type data to obtain an attribute vector, fuses the attribute vector and a gradient function on a time axis to obtain a multi-dimensional time sequence. A cumulative distance measure is calculated by using dynamic time warping, a minimum cost alignment path is searched, a non-continuous interval is extracted to determine a missing position to obtain a matching path. Coordinate points are extracted from the matching path to form an interface coordinate point set, the gradient function is converted into a tangent vector and mapped to a three-dimensional space grid to obtain a geometric constraint field. The interface coordinate point set and the geometric constraint field are used as constraints to construct a smooth functional equation, surface parameters are solved by minimization, and a three-dimensional geological model is constructed by interpolation fitting. The application can construct a three-dimensional geological model that conforms to the logic of geological evolution.

Description

Technical Field

[0001] This application belongs to the field of geological data processing and 3D modeling technology, and in particular relates to a method and system for constructing 3D geological models based on multi-source heterogeneous geological data. Background Technology

[0002] The method for constructing 3D geological models based on multi-source heterogeneous geological data integrates geological information from different sources, such as borehole data, seismic data, and stratigraphic dating data, to build a 3D model that reflects the underground geological structure. This modeling method has significant application value in fields such as oil and gas exploration, mineral resource evaluation, geological hazard prediction, and engineering geological evaluation.

[0003] In existing technologies, 3D geological modeling mainly employs spatial interpolation-based methods, using algorithms such as Kriging interpolation and spline interpolation to spatially expand discrete borehole data. Some methods utilize geostatistical techniques to probabilistically model stratigraphic distributions, or employ implicit function methods to reconstruct the surfaces of stratigraphic interfaces. These methods can achieve good modeling results when dealing with regular geological structures.

[0004] However, existing methods struggle to accurately identify stratigraphic gaps when dealing with sedimentary systems exhibiting complex contact relationships such as overlap, truncation, and unconformity. Furthermore, they lack effective constraints on the temporal logic of geological evolution, making it difficult to achieve topologically consistent connectivity of discrete borehole data in complex tectonic regions. Consequently, modeling results often fail to accurately reflect the geological history. Therefore, existing technologies suffer from the technical problem of insufficient geological model construction due to the difficulty in constructing topologically consistent 3D geological models that conform to geological evolution logic in complex geological tectonic regions. Summary of the Invention

[0005] The purpose of this application is to provide a method and system for constructing a three-dimensional geological model based on multi-source heterogeneous geological data, so as to solve the problem of insufficient geological model construction in the existing technology.

[0006] To address the aforementioned technical problems, in a first aspect, this application provides a method for constructing a three-dimensional geological model based on multi-source heterogeneous geological data, comprising: Acquire age and contact type data of different strata at each borehole point in the target area, as well as the sedimentary facies evolution rate at each borehole point; Numerical mapping was performed on the age data and contact type data to obtain attribute vectors for different strata. By fusing the attribute vectors with the gradient function of the stratum depositional thickness determined according to the sedimentary facies evolution rate on the time axis, a multidimensional time series sequence for each borehole point was obtained. Dynamic time warping is used to calculate the cumulative distance measure between multidimensional time series. Based on the cumulative distance measure, the alignment path with the minimum total cost between different multidimensional time series is searched. By extracting the discontinuous intervals of the alignment path, the missing position of the formation corresponding to each borehole point is determined, and the matching path between formations of different borehole points is obtained. The spatial coordinates of each borehole point in the corresponding stratum are extracted from the matching path to form an interface coordinate point set. The gradient function of the corresponding stratum at the alignment time is converted into a tangent vector at the corresponding spatial coordinate in the interface coordinate point set and mapped to the three-dimensional spatial grid of the target area to obtain the geometric constraint field. A smooth functional equation is constructed using the interface coordinate point set as the position constraint and the geometric constraint field as the gradient guiding term. The surface parameters are obtained by minimizing the smooth functional equation, and the surface parameters are used to perform surface interpolation fitting on multiple strata in the three-dimensional spatial grid to construct a three-dimensional geological model.

[0007] Optionally, numerical mapping is performed between the chronological data and the contact type data to obtain attribute vectors for different strata, including: Extract the time label of the stratum corresponding to each age data. Based on the time label, and according to the preset correspondence between the time label and the time component, obtain the time component of each stratum. The interface property information of each contact type data is extracted. Based on the interface property information, and according to the preset correspondence between the interface property information and the structural type components, the structural type components of each stratum are obtained. Based on the vertical spatial arrangement of the strata, an attribute vector for each stratum is constructed using time and structural type components.

[0008] Optionally, a multidimensional time series sequence for each borehole point is obtained by fusing the attribute vector with a gradient function of the formation depositional thickness determined according to the sedimentary facies evolution rate on the time axis, including: Calculate the slope of the sedimentary facies evolution rate at each time node corresponding to each borehole point, and based on the slope, determine the thickness evolution component at each time node according to the preset mapping relationship between the slope and the thickness evolution component. The thickness evolution components of each time node are arranged and curve-fitted according to the chronological order of the time axis to obtain the gradient function. The time axis is determined based on the continuous distribution span of the time labels of the period data in the time dimension. Based on the time label of the stratum corresponding to the attribute vector, the mapping interval of the attribute vector on the time axis is determined. The target thickness evolution component corresponding to the mapping interval is extracted from the gradient function. A multidimensional feature vector is constructed based on the target thickness evolution component and the corresponding attribute vector. The multidimensional feature vectors are arranged in chronological order along the time axis to construct a multidimensional time sequence for each borehole point.

[0009] Optionally, dynamic time warping is used to calculate the cumulative distance measure between multidimensional time series, including: Calculate the sum of squares of the differences between each component of the multidimensional feature vector in the multidimensional time series of any two borehole points to obtain the point-pair distance value. Arrange all point-pair distance values ​​according to the time position of each multidimensional feature vector on the time axis to obtain the distance matrix. By accumulating the minimum point-to-point distance value from the previous time position in the distance matrix along the time axis and the point-to-point distance value from the current time position, the cumulative distance measure for each time position is obtained.

[0010] Optionally, the alignment path with the minimum total cost between different multidimensional time series is searched based on the cumulative distance metric. The missing locations of the formation corresponding to each borehole point are determined by extracting the discontinuous intervals of the alignment path, thus obtaining the matching path between formations at different borehole points, including: Extract the preceding time position corresponding to the smallest cumulative distance measure along the reverse direction of the time axis as a path node, and arrange all path nodes in chronological order according to the time axis to obtain the aligned path. Calculate the time interval between adjacent path nodes in the alignment path on the time axis, and determine the non-contiguous intervals between adjacent path nodes whose time interval is greater than a preset interval threshold as missing positions. Based on the multidimensional feature vector corresponding to each path node in the alignment path, the stratigraphic relationship between the two corresponding borehole points is established. The time position corresponding to the missing position is marked as the stratigraphic missing identifier in the stratigraphic relationship, and the stratigraphic relationship is arranged according to the order of the path nodes in the alignment path to obtain the matching path.

[0011] Optionally, the spatial coordinates of each borehole point in the corresponding stratum are extracted from the matching path to form an interface coordinate point set. The gradient function of the corresponding stratum at the alignment time is converted into a tangent vector at the corresponding spatial coordinate in the interface coordinate point set and mapped to the three-dimensional spatial mesh of the target region to obtain the geometric constraint field, including: Extract the spatial coordinates of the two borehole points associated with each layer relationship in the matching path in the corresponding stratum, and arrange all spatial coordinates according to the order of the layer relationships in the matching path to obtain the interface coordinate point set; Based on the time position of the path node corresponding to each layer relationship in the matching path on the time axis, the thickness evolution component corresponding to the time position is extracted from the gradient function, the direction vector between adjacent spatial coordinates in the interface coordinate point set is calculated, and the tangent vector of each spatial coordinate is constructed with the thickness evolution component as the amplitude and the direction vector as the direction. Extract the spatial coordinates that are closest to each grid node in the 3D spatial grid from the set of interface coordinate points, and construct a geometric constraint field using the tangent vector corresponding to the spatial coordinates as the directional constraint value of the grid node.

[0012] Optionally, a smooth functional equation is constructed using the interface coordinate point set as the position constraint and the geometric constraint field as the gradient guiding term. The surface parameters are obtained by minimizing the smooth functional equation, and then surface interpolation fitting is performed on multiple strata in the three-dimensional spatial grid using these surface parameters to construct a three-dimensional geological model, including: The position constraint term is the sum of the minimum squared distances from each grid node in the 3D spatial grid to the set of interface coordinate points. The gradient constraint term is the sum of the squared gradient deviations between the direction constraint value of each grid node in the geometric constraint field and the rate of change of direction between adjacent grid nodes. The position constraint term and the gradient constraint term are weighted and summed using preset weight coefficients to construct a smooth functional equation with the position coordinates of the grid nodes in the 3D spatial grid as independent variables. Based on the smooth functional equation, the position adjustment direction and adjustment range of each grid node are determined by calculating the derivative value of the position coordinates of each grid node in the three-dimensional space grid. The position coordinates of the corresponding grid node are iteratively updated along the position adjustment direction and according to the adjustment range until the value of the smooth functional equation reaches the minimum value. The target position coordinates of the grid node are then determined as surface parameters. Connect adjacent mesh nodes corresponding to the surface parameters to obtain surface elements. Stack surface elements of different depths according to the layer relationship in the matching path to obtain a three-dimensional geological model.

[0013] Secondly, this application provides a three-dimensional geological model construction system based on multi-source heterogeneous geological data, including: The acquisition module is used to acquire age data and contact type data of different strata at each borehole point in the target area, as well as the sedimentary facies evolution rate of each borehole point; The generation module is used to numerically map age data and contact type data to obtain attribute vectors of different strata. By fusing the attribute vectors with the gradient function of the stratum depositional thickness determined according to the sedimentary facies evolution rate on the time axis, a multidimensional time series sequence of each borehole point is obtained. The generation module is also used to calculate the cumulative distance measure between multidimensional time series using dynamic time warping, search for the alignment path with the minimum total cost between different multidimensional time series based on the cumulative distance measure, and determine the missing position of the formation corresponding to each borehole point by extracting the discontinuous interval of the alignment path, so as to obtain the matching path between the formations of different borehole points. The mapping module is used to extract the spatial coordinates of each borehole point in the corresponding stratum from the matching path to form an interface coordinate point set. It converts the gradient function of the corresponding stratum at the alignment time into the tangent vector at the corresponding spatial coordinate in the interface coordinate point set and maps it to the three-dimensional spatial grid of the target area to obtain the geometric constraint field. The module is used to construct a smooth functional equation with the interface coordinate point set as the position constraint and the geometric constraint field as the gradient guiding term. The surface parameters are obtained by minimizing the smooth functional equation, and the surface parameters are used to perform surface interpolation fitting on multiple strata in the three-dimensional spatial grid to construct a three-dimensional geological model.

[0014] Thirdly, this application provides an electronic device, comprising: Memory, used to store computer programs; A processor is configured to execute the computer program to implement the steps of the three-dimensional geological model construction method based on multi-source heterogeneous geological data as described in the first aspect above.

[0015] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the steps of the three-dimensional geological model construction method based on multi-source heterogeneous geological data as described in the first aspect above.

[0016] The method for constructing a three-dimensional geological model based on multi-source heterogeneous geological data provided in this application first transforms heterogeneous geological description information into a computable multidimensional time series by numerically mapping age data and contact type data and fusing them with a gradient function. Subsequently, dynamic time warping is used to nonlinearly align the multidimensional time series, which can automatically identify the stratigraphic missing locations between different borehole points in the time dimension, solving the problem that existing methods are unable to accurately locate stratigraphic missing areas under complex contact relationships such as overlap and cut-off.

[0017] Then, by converting the gradient function into a tangent vector and constructing a geometric constraint field, a geological evolution time-series logical constraint is introduced into the surface reconstruction. This enables the minimization solution of the smooth functional equation to complete multi-level co-surface fitting while satisfying the topological consistency requirement, avoiding the modeling distortion caused by the lack of evolutionary logical constraints in existing methods. Therefore, this application effectively solves the technical problem of existing technologies in constructing three-dimensional geological models that conform to geological evolution logic and are topologically consistent in complex geological structures.

[0018] Furthermore, this application first extracts the spatial coordinates of borehole points associated with each stratigraphic relationship from the matching path and arranges them in chronological order according to the stratigraphic relationship to construct a set of interface coordinate points, transforming the originally discretely distributed borehole spatial information into an ordered set of points with stratigraphic logic. Subsequently, by extracting the thickness evolution component corresponding to the time position from the gradient function and constructing a tangent vector with the direction vector between adjacent spatial coordinates as the direction, the geological evolution time series information is transformed into spatial geometric direction constraints, ensuring that the extension trend of the stratigraphic interface is consistent with the geological history evolution process, overcoming the defect of existing methods that suffer from interface reconstruction distortion due to the lack of evolutionary logic constraints.

[0019] Finally, the tangent vector is mapped to each grid node in the 3D spatial grid as a directional constraint value, forming a geometric constraint field covering the entire modeling region. This provides globally consistent gradient guidance information for subsequent multi-level co-surface fitting, thus laying the foundation for solving the problem of topological consistency connectivity in complex tectonic regions. Therefore, this application effectively improves the evolutionary logic consistency and spatial topological accuracy of geological model construction under complex contact relationships. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 A flowchart illustrating a method for constructing a three-dimensional geological model based on multi-source heterogeneous geological data, provided in an embodiment of this application; Figure 2 A schematic diagram illustrating a method for determining the matching path between different borehole points in an embodiment of this application; Figure 3 A flowchart illustrating a method for constructing a three-dimensional geological model, provided in an embodiment of this application; Figure 4 A schematic diagram of a three-dimensional geological model construction system based on multi-source heterogeneous geological data provided in this application embodiment; Figure 5 This is a schematic diagram of the hardware structure of an electronic device provided in one embodiment of this application. Detailed Implementation

[0022] Existing 3D geological modeling methods face two fundamental shortcomings when dealing with sedimentary systems with complex contact relationships such as overlap, truncation, and unconformity. Firstly, spatial interpolation methods such as Kriging interpolation and implicit functions rely solely on the spatial location information of borehole points for surface fitting, failing to perceive the evolutionary differences of strata over time. This results in the identification of missing strata locations relying on manual interpretation rather than automatic computer positioning, which can easily lead to erroneous stratigraphic correspondences in areas with sparse boreholes.

[0023] On the other hand, existing methods lack quantitative constraints on the temporal logic of geological evolution when constructing stratigraphic interfaces. This results in generated surfaces that only satisfy geometric smoothness conditions rather than geological evolution laws, often leading to distortions such as topological non-closure or cross-layering at overlap and truncation boundaries. The root of this contradiction lies in the fact that existing technologies simplify geological modeling to a purely spatial geometric problem, neglecting the temporal evolution logic of stratigraphic formation and the constraints of sedimentary processes.

[0024] To address the aforementioned issues, this application proposes a method for constructing a three-dimensional geological model based on multi-source heterogeneous geological data. The core of this method lies in the synergistic integration of geological evolution time-series information and spatial geometric information into the modeling constraint system. Specifically, this application fuses age data, contact type data, and sedimentary facies evolution rates into a multi-dimensional time-series sequence. Dynamic time warping is then used to nonlinearly align stratigraphic sequences between different boreholes along the time dimension, thereby automatically identifying missing stratigraphic locations and establishing stratigraphic matching paths.

[0025] Furthermore, the geological evolution rate is transformed into a tangent vector, and a geometric constraint field covering the entire region is constructed. This field serves as a gradient guiding term to drive the minimization solution of the smooth functional equation, achieving multi-layer co-surface fitting. This method transforms geological temporal logic into quantifiable mathematical constraints, upgrading the modeling process from simple spatial interpolation to evolutionary logic restoration under spatiotemporal joint constraints. It solves the problems of stratigraphic misjudgment and topological distortion caused by the lack of temporal constraints in existing technologies, and achieves topological consistency construction of three-dimensional geological models under complex contact relationships.

[0026] To enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some embodiments of the present application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0027] To address the problems of existing technologies, embodiments of this application provide a method, apparatus, device, computer storage medium, and computer program product for constructing a three-dimensional geological model based on multi-source heterogeneous geological data. The method for constructing a three-dimensional geological model based on multi-source heterogeneous geological data provided in this application embodiment is described below.

[0028] Figure 1 This illustration shows a flowchart of a method for constructing a three-dimensional geological model based on multi-source heterogeneous geological data, according to an embodiment of this application. Figure 1 As shown, the method includes: S101. Obtain age data and contact type data of different strata at each borehole point in the target area, as well as the sedimentary facies evolution rate of each borehole point.

[0029] Chronological data refers to geological age information indicating the formation time of strata. It can include the name of the chronostratigraphic unit and the absolute age range, and is used to identify the temporal position of each stratum in geological history. Contact type data refers to structural information describing the interface properties between adjacent strata. It can include contact types such as conformable, parallel unconformity, angular unconformity, overlap, and truncation, and is used to reflect the occurrence of tectonic activity and erosive discontinuities in the stratigraphic sedimentary sequence.

[0030] Sedimentary facies evolution rate is a physical quantity that represents the rate of change in the thickness of sedimentary material accumulation per unit time during different geological periods. Its dimension is m / Ma. The sedimentary facies evolution rate is the ratio of the sedimentary thickness of each stratum to its sedimentary duration and is used to quantify the temporal evolution of sedimentary intensity.

[0031] For the The accumulation rate of the strata. ,in The sedimentary thickness of this stratum and the depth of the bottom interface of this stratum as recorded in the borehole core. Depth of top interface The difference is measured directly. The depositional duration of this stratum and the starting point of the absolute age interval of the stratum's age data. and the finish line The difference is determined, and the unit is Ma.

[0032] Specifically, in the modeling scenario of complex sedimentary systems in sedimentary basins, three types of data—chronological data, contact type data, and sedimentary facies evolution rate—can be obtained by reviewing regional geological survey archives and original borehole records. First, the chronostratigraphic unit names and absolute age ranges of each stratum are extracted from the core description records and vertical borehole depth records at each borehole point, serving as the chronological data for that borehole point. The depths of the top and bottom interfaces corresponding to the vertical sequence are also extracted. Next, the interface property information between adjacent strata is extracted from the stratigraphic contact relationship description records, serving as the contact type data for that borehole point.

[0033] For example, a certain drilling point identified a total of The stratigraphic set, its age dataset is as follows ,in This refers to the chronostratigraphic unit and age range of a certain early stratum. This is the latest set of geological chronology information. Coexistence between strata A set of contact interfaces, and their contact type datasets. ,in This could be an angular unconformity contact between the first and second stratigraphic units. It can be considered as an integrated contact between the second and third strata.

[0034] Simultaneously, for each formation at each borehole point, the top interface depth of that formation is read from the core description record and vertical borehole depth record at each borehole point. Depth of bottom interface The sedimentary thickness of this stratum was obtained. The absolute age range of this stratum was then obtained from its age data to determine the depositional duration. .

[0035] according to Layer-by-layer calculation yields the depositional rate of the borehole at each stratum during the corresponding geological period, serving as the sedimentary facies evolution rate data for that borehole. ,in This represents the accumulation rate during the first set of strata deposition. This represents the depositional rate corresponding to the most recent strata during the depositional period. Ultimately, this applies to the entire target area. Repeat the above process for each borehole point to obtain the corresponding age dataset for each borehole point. Contact type dataset and sedimentary facies evolution rate dataset .

[0036] S102. Numerical mapping is performed on the age data and contact type data to obtain attribute vectors for different strata. By fusing the attribute vectors with the gradient function of the stratum depositional thickness determined according to the sedimentary facies evolution rate on the time axis, a multidimensional time series sequence for each borehole point is obtained.

[0037] Optionally, the process of numerically mapping the age data and contact type data in step S102 to obtain attribute vectors for different strata may specifically include: S1021. Extract the time label of the stratum corresponding to each age data. Based on the time label, according to the preset correspondence between the time label and the time component, obtain the time component of each stratum.

[0038] A time tag is a discrete identifier extracted from chronological data to identify the geological time unit to which a stratum belongs. It can encode the geological time level corresponding to the name of the chronological stratigraphic unit in the chronological data and is used to uniquely determine the position of the stratum in the geological time series.

[0039] Time components refer to numerical variables that can be used in quantitative calculations after converting time labels through a pre-defined correspondence between time labels and time components. They can be the representative absolute age value of the geological time unit corresponding to the time label, expressed in millions of years, and used to establish calculable time coordinates for various strata on a time axis. The pre-defined correspondence between time labels and time components is shown in Table 1 below: Table 1: Correspondence between Time Labels and Time Components

[0040] As shown in Table 1, using common geological time units as examples, Table 1 establishes a mapping relationship between the name of a chronostratigraphic unit and its representative absolute age value. Each time label corresponds to a unique time component value, which is the median of the absolute age range of the corresponding chronostratigraphic unit, expressed in millions of years, serving as the quantitative coordinate of that stratum on the time axis. In practical applications, Table 1 can be expanded according to the specific distribution of strata ages in the target area.

[0041] Specifically, in the scenario of modeling complex sedimentary systems in a sedimentary basin, the age dataset for each borehole point is first obtained from S101. In this process, the names of the chronostratigraphic units corresponding to each stratigraphic unit are extracted one by one and used as the time labels for that stratigraphic unit. Then, according to the preset correspondence in Table 1, each time label is matched with the corresponding representational absolute age value to obtain the time components of each stratigraphic unit.

[0042] For example, a total of [number] borehole points were identified. Stratigraphic and age datasets are as follows ,in The time stamp is Lower Jurassic. The time stamp is Middle Jurassic. The time stamp is the Upper Jurassic period. If the time label is Lower Cretaceous, then referring to Table 1, the corresponding time components are obtained sequentially as follows: , , , , forming a set of time components .

[0043] S1022. Extract the interface property information of each contact type data. Based on the interface property information, and according to the preset correspondence between the interface property information and the structural type components, obtain the structural type components of each stratum.

[0044] Interface property information refers to the qualitative description of the structural features and sedimentary discontinuity of the contact interface between adjacent strata extracted from contact type data. It can be used to identify the type of contact interface such as conformable, parallel unconformity, angular unconformity, overlap, and truncation, and is used to distinguish the intensity of tectonic activity and the state of stratigraphic continuity represented by different contact interfaces during geological evolution.

[0045] The structural type component refers to the integer encoded value obtained by converting the interface property information through a pre-defined correspondence between the interface property information and the structural type component. It can be an ordered integer arranged from low to high according to the complexity of the stratigraphic topology and is used to convert qualitative interface contact features into quantitative expressions that can participate in vector operations.

[0046] The correspondence between the preset interface property information and the structure type components is shown in Table 2 below: Table 2: Correspondence between Interface Property Information and Structure Type Components

[0047] As shown in Table 2, the information on various interface properties is encoded into ordered integer structure type components according to the topological complexity of the stratigraphic contact interfaces, from low to high. Among them, conformable contact indicates continuous and uninterrupted stratigraphic deposition, with the lowest encoding value; overlap and truncation contact indicate that the stratigraphy has spatial gaps or is truncated, with the most complex topological relationships, and the highest encoding value.

[0048] Specifically, the contact type dataset for each borehole point is first obtained from S101. In this process, the interface property information of each contact interface is extracted one by one. Then, according to the preset correspondence in Table 2, each interface property information is matched with the corresponding structural type component.

[0049] For example, there are 3 contact interfaces between 4 formations at a certain borehole point, and the contact type dataset is as follows: ,in The interface properties information indicates that the angle is not integrated. The interface information is integrated. If the interface property information is overridden, then referring to Table 2, the corresponding structural type components are obtained sequentially. , , Forming a set of structural type components .

[0050] S1023. Construct the attribute vector of each stratum according to the vertical spatial arrangement of the strata, based on the time component and the structural type component.

[0051] An attribute vector is a multi-component column vector that is constructed by splicing the temporal component of the stratum and the structural type component of the overlying interface in vertical spatial order for each stratum. It can simultaneously express the temporal location and contact tectonic characteristics of the stratum and is used to uniformly express heterogeneous chronostratigraphic information and interface contact information into a computable vector form.

[0052] Specifically, firstly, based on the vertical spatial arrangement of strata from deep to shallow, the time components of each stratum are systematically aligned with the structural type components of the corresponding interfaces. Next, the time components of each stratum... Its overlying interface's structural type component By splicing together the data, the attribute vector of the stratum is constructed. Specifically, it can be expressed as:

[0053] in, For the first The time component of the stratigraphic sequence For the first strata and the first The structural type component of the interface between strata, for the deepest stratum, its The component can be assigned a default base encoding value of 0.

[0054] This embodiment transforms qualitative heterogeneous geological description information in complex sedimentary systems of sedimentary basins into quantitative vector expressions with time-series attributes, effectively solving the problem of difficulty in merging and calculating multi-source heterogeneous geological data due to format differences.

[0055] Optionally, step S102, which involves fusing the attribute vector with the gradient function of the formation depositional thickness determined according to the sedimentary facies evolution rate on the time axis to obtain a multidimensional time series sequence for each borehole point, may specifically include: S1024. Calculate the slope of the sedimentary facies evolution rate at each time node corresponding to each borehole point, and based on the slope, determine the thickness evolution component at each time node according to the preset mapping relationship between the slope and the thickness evolution component.

[0056] The slope of change refers to the ratio of the difference between adjacent time points in the sedimentary facies evolution rate dataset to the time interval. It is used to describe the local variation trend of sedimentation rate over time and reflects the degree of acceleration or deceleration of the sedimentary dynamic environment during that geological period.

[0057] The thickness evolution component refers to a numerical variable that quantifies the local evolution intensity of deposition thickness at each time point, obtained by transforming the change slope through a preset mapping relationship between the change slope and the thickness evolution component. The preset mapping relationship between the change slope and the thickness evolution component is shown in the following formula:

[0058] in, For the first The slope of change at each time point For the corresponding thickness evolution components, The preset dimensional conversion scaling factor and the unit is This is used to unify the dimensions of the changing slope into the length dimension of the thickness evolution. The preset regional baseline deposition thickness constant is expressed in units of... .

[0059] The specific calibration method is as follows: First, select several boreholes with complete stratigraphic development and core records within the target area as reference boreholes. For each stratigraphic unit in the reference boreholes, directly measure the measured sedimentary thickness from the measured depth difference between the top and bottom interfaces of the core, and use this as the measured thickness evolution component at that time point. The unit is meters, and the corresponding slope of change is calculated using the difference method in this step. Thus obtain Group sample pairs .

[0060] Then, with the changing slope The independent variable and the measured thickness evolution component are... As the dependent variable, the least squares method is used to apply the linear mapping. Perform a fitting to make the sum of squared residuals Take the minimum value, and thus solve the problem simultaneously. and Its closed-form solution is:

[0061] in, , These represent the sample means of the slope of change and the measured thickness evolution component, respectively. The total number of sample pairs extracted from the reference borehole. (Calculated) and It is applicable to all boreholes within the same sedimentary system in the target area.

[0062] Specifically, firstly, the sedimentary facies evolution rate dataset for each borehole point obtained in S101 is analyzed. Perform the difference operation between adjacent nodes to calculate the first node. The slope of change at each time point:

[0063] The slope at the endpoints is calculated using one-sided forward and backward differences, specifically expressed as the first endpoint... and tail endpoint Next, the slope of change at each time point is substituted into the above mapping relationship to obtain the thickness evolution component at each time point. Forming a thickness evolution component sequence .

[0064] S1025. Arrange the thickness evolution components of each time node in chronological order according to the time axis and obtain the gradient function by curve fitting. The time axis is determined by the continuous distribution span of the time labels of the period data in the time dimension.

[0065] The time axis refers to a continuous time coordinate axis formed by the time labels of the age data of each borehole point in the target area, from the earliest geological age to the latest geological age. It can be an ordered time series in millions of years and is used as a reference benchmark to unify the time coordinates of different strata of each borehole point to ensure that the multidimensional time series of different borehole points are comparable in the time dimension.

[0066] The gradient function is a continuously differentiable function obtained by curve fitting, taking the thickness evolution component as the function value and each time node on the time axis as the independent variable. It is used to extend the discrete thickness evolution component into a continuous description of thickness evolution that can be evaluated at any time on the time axis. The general form of the gradient function is defined as follows:

[0067] in, For time variables on the time axis, For the first The thickness evolution components at each time point. For the first cubic spline basis functions centered at time nodes This represents the total number of time points involved in the fitting process.

[0068] Specifically, firstly, based on the entire target area The minimum and maximum values ​​of the time components corresponding to the time tags in the age data of each borehole point are used to determine the start and end range of the time axis, forming a continuous time coordinate axis. Next, according to the chronological order of the time axis from earliest to newest, the thickness evolution component sequence obtained in S1024 is... With the corresponding set of time components Arranged in an ordered manner, discrete data pairs are formed for time nodes and thickness evolution components. .

[0069] Then, using the aforementioned discrete data pairs as input, and employing natural boundary conditions (i.e., setting the second derivative of the spline curve to zero at both ends), the basis function coefficients are solved using a cubic spline curve fitting method to obtain the gradient function at the borehole point. .

[0070] S1026. Based on the time label of the stratum corresponding to the attribute vector, determine the mapping interval of the attribute vector on the time axis, extract the target thickness evolution component corresponding to the mapping interval from the gradient function, and construct a multidimensional feature vector based on the target thickness evolution component and the corresponding attribute vector.

[0071] The mapping interval refers to the continuous time period corresponding to the absolute age interval of the stratum determined on the time axis based on the time label of the stratum corresponding to the attribute vector. It is used to locate the function interval corresponding to the deposition period of the stratum in the gradient function and serves as the time range for extracting the target thickness evolution component.

[0072] The target thickness evolution component refers to the representative function value extracted by integrating and averaging the gradient function over a mapping interval of a certain stratum. It can be the mean of the gradient function over that mapping interval and is used to condense the thickness evolution intensity information corresponding to the depositional period of the stratum in the continuous gradient function into a single value. The multidimensional feature vector is an extended vector formed by concatenating the target thickness evolution component with the corresponding attribute vector, and is used to simultaneously carry three types of information: time location, contact tectonic features, and thickness evolution intensity.

[0073] Specifically, firstly, based on the time label corresponding to each stratum in S1021, the absolute age range of that stratum is extracted from the chronological data, and the continuous time period corresponding to that age range on the time axis is determined as the mapping range of the attribute vector. ,in and These represent the start and end time coordinates of the age interval of this stratum. Next, the mapped interval is substituted into the gradient function fitted in S1025. The mean value of the integral over this interval is calculated as the target thickness evolution component. The specific calculation process is shown in the following formula:

[0074] in, For the first The evolution component of the target thickness corresponding to the strata. and These are the start and end time coordinates of the mapping interval for this stratum. Then, With the corresponding attribute vector To splice together and construct the first Multidimensional feature vectors of the strata .

[0075] S1027. Arrange the multidimensional feature vectors in chronological order according to the time axis to construct a multidimensional time sequence for each borehole point.

[0076] A multidimensional time series refers to an ordered vector sequence formed by arranging the multidimensional feature vectors corresponding to all strata at each borehole point in chronological order from earliest to newest along the time axis. It is a computable discrete sequence that includes three types of information: time location, contact structural features, and thickness evolution intensity, and has time-series attributes. It is used to uniformly transform the heterogeneous geological description information of the borehole point into a standardized multidimensional time series expression.

[0077] Specifically, firstly, based on the multidimensional feature vectors corresponding to each stratum in S1026... Medium time component Since a larger absolute age value indicates an earlier geological era, all multidimensional feature vectors are sorted in descending order according to the timeline from earliest to newest. Then, the sorted multidimensional feature vectors are concatenated sequentially to construct the multidimensional time series sequence of the borehole site. Ultimately, for the entire target area... Repeat the processing steps S1021 to S1027 for each borehole point to obtain the multidimensional time series sequence corresponding to each borehole point. .

[0078] This embodiment transforms the discrete sedimentary evolution information of each borehole point in the complex sedimentary system of the sedimentary basin into a standardized computable sequence that takes into account time location, contact tectonic features, and thickness evolution intensity.

[0079] S103. Calculate the cumulative distance measure between multidimensional time series using dynamic time warping. Search for the alignment path with the minimum total cost between different multidimensional time series based on the cumulative distance measure. Determine the missing position of the formation corresponding to each borehole point by extracting the discontinuous intervals of the alignment path, and obtain the matching path between formations of different borehole points.

[0080] Figure 2A schematic diagram of a method for determining the matching path between formations at different borehole points, provided in one embodiment of this application, is shown. Figure 2 As shown, drilling point The multidimensional time series 21 is arranged along the horizontal time axis as follows Drilling point The multidimensional time series 22 is arranged along the vertical time axis as follows The point-to-point distance values ​​corresponding to the pairwise combinations of the multidimensional feature vectors at each time position of the two multidimensional time series are accumulated position by position to form a cumulative distance matrix 23. The row index of the cumulative distance matrix 23 corresponds to the borehole point. The time position index of the multidimensional feature vector, and the column index corresponding to the borehole point. The temporal position index of the multidimensional feature vector.

[0081] Starting from the termination time position of the cumulative distance matrix 23, backtracking along the time axis position by position to the preceding time position with the smallest cumulative distance measure, we obtain the alignment path 24 formed by sequentially connecting path nodes 25. Each path node 25 is recorded as... Alignment path 24 is the nonlinear correspondence between two multidimensional time series that minimizes the total cost in the time dimension.

[0082] The time interval between two adjacent path nodes 25 in alignment path 24 on the time axis Greater than the preset interval threshold At this time, a discontinuous interval with a significantly larger span than the intervals between other adjacent path nodes is formed between adjacent path nodes 25. This discontinuous interval is the missing position 26. The missing position 26 indicates that there are geological phenomena such as stratum missingness, erosion, or overburden between the two borehole points during this time period. The alignment path 24 is then embedded with a stratum missing marker 27 at the missing position 26. To make the alignment path 24 break at that point to explicitly mark the time position of the formation discontinuity between the two borehole points, the entire path from the start time position to the end time position and carrying the formation missing marker 27 constitutes the matching path 28.

[0083] Optionally, the process of calculating the cumulative distance measure between multidimensional time series using dynamic time warping in step S103 may specifically include: S1031. Calculate the sum of squares of the differences between each component of each multidimensional feature vector in the multidimensional time series of any two borehole points to obtain the point-to-point distance value. Arrange all point-to-point distance values ​​according to the time position of each multidimensional feature vector on the time axis to obtain the distance matrix.

[0084] Point-to-point distance refers to the first value of a sequence in a multidimensional time series of any two borehole points. The multidimensional feature vector and the first sequence of another sequence The sum of squares of the differences between the components of the multidimensional feature vectors is used to measure the overall difference between the two multidimensional feature vectors in three dimensions: time location, contact structure features, and thickness evolution intensity.

[0085] A distance matrix is ​​a two-dimensional numerical matrix formed by arranging the point-to-point distance values ​​of all multidimensional feature vector pairs between any two borehole point multidimensional time series according to their respective time positions on the time axis. It is used to comprehensively record the similarity information of two sequences at any combination of time positions.

[0086] Specifically, in the modeling scenario of complex sedimentary systems in sedimentary basins, all data first obtained from S1027... Multidimensional time series of drilling points Choose any two drilling points. and Corresponding multidimensional time series and .

[0087] Next, regarding The Middle multidimensional feature vectors and The Middle multidimensional feature vectors Calculate the corresponding point-to-point distance value ,in, , , Drilling points No. The time component, structure type component, and target thickness evolution component of the multidimensional feature vector. , , Drilling point No. The corresponding components of each multidimensional feature vector. Then, according to the time position of each of the two sequences on the time axis, all... The distance values ​​of each pair of points are arranged into a distance matrix:

[0088] in, The row index corresponds to The time position index and column index of the multidimensional feature vector are corresponding to... The time position index and elements of the multidimensional feature vector Indicates that the two sequences are in the first position. With the Distance values ​​between points of the multidimensional feature vector at each time location.

[0089] S1032. Along the time axis, accumulate the point-to-point distance value with the smallest value of the previous time position in the distance matrix and the point-to-point distance value of the current time position to obtain the cumulative distance measure for each time position.

[0090] Cumulative distance measure refers to the cumulative path cost obtained by gradually adding the point-to-point distance value at the current position to the minimum cumulative value calculated in the previous time position along the time axis under the dynamic time warping framework. It is used to quantify the overall similarity cost of two multidimensional time series under a specific time position alignment scheme.

[0091] Specifically, firstly, using the distance matrix For input, construct and Cumulative distance matrix of the same dimension Its elements Indicates the time position of the two sequences and The cumulative path cost at alignment point. Then, following the timeline from earliest to newest, the cumulative distance measure is calculated time-by-time according to the following recursive relationship:

[0092] in, Current time position , Cumulative distance measurement at location This represents the distance between the corresponding point pairs in the distance matrix. The boundary condition is set to the value with the minimum cumulative cost among the three possible preceding time positions. , , .

[0093] This embodiment calculates the comprehensive differences between the time position, contact tectonic features and thickness evolution intensity of each multidimensional feature vector pair in the multidimensional time series of any two borehole points in the complex sedimentary system of the sedimentary basin, constructs a distance matrix, and then accumulates it position by position along the time axis to obtain the cumulative distance matrix, thereby transforming the multidimensional heterogeneous differences between stratigraphic sequences of different borehole points into quantifiable cumulative path costs.

[0094] Optionally, step S103, which involves searching for the alignment path with the minimum total cost among different multidimensional time series based on the cumulative distance metric, and determining the missing location of the formation corresponding to each borehole point by extracting the discontinuous intervals of the alignment path, can specifically include: S1033. Extract the preceding time position corresponding to the smallest cumulative distance measure along the reverse direction of the time axis as a path node, and arrange all path nodes in chronological order according to the time axis to obtain the aligned path.

[0095] A path node refers to the index coordinate of the preceding time position with the smallest cumulative distance measure at each time point during reverse backtracking of the cumulative distance matrix. It can be the index pair of elements traversed sequentially along the optimal path in the cumulative distance matrix and is used to record the matching relationship of each time position when two multidimensional time series achieve optimal cost alignment in the time dimension. The alignment path refers to the ordered index sequence formed by arranging all path nodes in ascending order of time along the time axis and is used to describe the overall nonlinear correspondence between the two borehole point multidimensional time series in the time dimension.

[0096] Specifically, firstly, the cumulative distance matrix obtained in S1032... For input, start from the end time position Starting from this point, we backtrack along the timeline, sequentially from one point in time to the next, at each point... From its three possible preceding positions , , Extract the index pair corresponding to the one with the smallest cumulative distance measure as the path node, and continue backtracking until the starting position is reached. .

[0097] Next, all the backtracked path nodes are rearranged in chronological order from earliest to newest, forming a sequence like this: Figure 2 Alignment path marked as 24 Each path node 25 records as This is used to record the time indexes of the two sequences during the alignment step. The total number of nodes in alignment path 24 is given by the fact that alignment path 24 passes through the cumulative distance measure with the smallest value in the cumulative distance matrix 23.

[0098] S1034. Calculate the time interval between adjacent path nodes in the alignment path on the time axis, and determine the non-continuous interval between adjacent path nodes whose time interval is greater than the preset interval threshold as the missing position.

[0099] The time interval refers to the absolute value of the difference between the time components of the multidimensional feature vectors corresponding to two adjacent path nodes in the alignment path. It is used to measure the span of two adjacent path nodes on the time axis and to reflect whether there is a sudden jump in time in the alignment path within this interval.

[0100] Missing locations refer to the non-contiguous intervals between adjacent path nodes in the alignment path where the time interval is greater than a preset interval threshold. These locations are used to identify spatial positions between two borehole points where geological phenomena such as stratigraphic missingness, erosion, or overburden exist within that time period. The preset interval thresholds are shown in Table 3 below. Table 3: Preset Interval Threshold Comparison Table

[0101] As shown in Table 3, Table 3 sets corresponding preset interval thresholds according to the stratigraphic age span of the target area. Preset interval threshold The value is determined by taking the median of the absolute values ​​of the differences in time components between adjacent formations at all borehole points in the target area. As a characteristic time interval between adjacent strata under normal regional sedimentary conditions, and let .

[0102] The values ​​in Table 3 are obtained by statistically analyzing and merging typical time intervals between adjacent strata within different stratigraphic time spans according to this rule. The larger the stratigraphic time span, the larger the characteristic time intervals between adjacent strata are usually, and the higher the corresponding preset interval threshold. In practical applications, Table 3 can be adjusted based on the statistical results of the time intervals between adjacent strata in the target area.

[0103] Specifically, firstly, the alignment path 24 obtained in S1033 is... Adjacent path nodes and Extract the time components of the corresponding time positions of the two drilling points respectively, and calculate the time interval. ,in For the first With the The time interval between 25 path nodes and .

[0104] Next, based on the stratigraphic age span of the target area, the preset interval threshold is determined according to Table 3. To satisfy all Adjacent path node pairs The discontinuous intervals between them are determined as follows: Figure 2 The missing position is marked as 26.

[0105] S1035. Based on the multi-dimensional feature vector corresponding to each path node in the alignment path, establish the stratigraphic relationship between the two corresponding borehole points, mark the time position corresponding to the missing position as the stratigraphic missing identifier in the stratigraphic relationship, and arrange the stratigraphic relationship according to the order of the path nodes in the alignment path to obtain the matching path.

[0106] Stratification relationship refers to a structured record established based on each path node in the alignment path, describing the matching status of corresponding strata between two borehole points on the time axis. It can include time position subscript pairs of path nodes, corresponding multidimensional feature vectors, and stratum connectivity status markers, and is used to clearly record which strata have corresponding relationships between two borehole points and which time positions have missing strata.

[0107] Stratigraphic missing markers are special symbols assigned to the time interval corresponding to the missing location in stratigraphic relationships. They can be preset empty layer marker values ​​and are used to explicitly mark the time location of missing strata in the stratigraphic relationship sequence. Matching paths refer to the ordered structured sequence formed by arranging all stratigraphic relationships according to the order of the path nodes in the alignment path. They are used to fully describe the layer-by-layer correspondence and missing layer distribution between two borehole points from the earliest to the latest strata.

[0108] Specifically, firstly, based on the alignment path obtained in S1033 For each path node Extract the multidimensional feature vectors of the two drilling points at their corresponding time positions. and Establish a record of the stratigraphic relationship between the two borehole points at this path node. ,in Mark the stratigraphic connectivity at this path node.

[0109] Next, based on all the missing positions 26 identified in S1034, for the path nodes 25 in the alignment path 24 located within the interval of the missing positions 26, their corresponding hierarchical relationships are recorded. Assigning stratigraphic missing identifier 27 This indicates that the formation corresponding to this time location is discontinuous between the two borehole points; for path node 25 at non-missing location 26, Assign a normal connectivity marker. Then, align the path nodes in path 24. The order in which all hierarchical relationship records are arranged is as follows: Figure 2 Matching path marked as 28 .

[0110] This embodiment targets any two borehole points in a complex sedimentary system of a sedimentary basin. Based on the cumulative distance matrix, it backtracks to obtain the alignment path with the minimum total cost. Then, it combines the preset interval threshold in Table 3 to identify discontinuous intervals in the path to locate the missing stratigraphic position. Finally, it integrates the path nodes and missing markers into an ordered matching path, realizing an accurate description of the nonlinear correspondence of stratigraphic sequence at different spatial locations.

[0111] S104. Extract the spatial coordinates of each borehole point in the corresponding stratum from the matching path to form an interface coordinate point set. Convert the gradient function of the corresponding stratum at the alignment time into the tangent vector at the corresponding spatial coordinate in the interface coordinate point set and map it to the three-dimensional spatial grid of the target area to obtain the geometric constraint field.

[0112] Optionally, step S104, which involves extracting the spatial coordinates of each borehole point in the corresponding stratum from the matching path to form an interface coordinate point set, converting the gradient function of the corresponding stratum at the alignment time into a tangent vector at the corresponding spatial coordinate in the interface coordinate point set, and mapping it to the three-dimensional spatial mesh of the target region to obtain the geometric constraint field, may specifically include: S1041. Extract the spatial coordinates of the two borehole points associated with each layer relationship in the matching path in the corresponding stratum, and arrange all spatial coordinates according to the order of the layer relationships in the matching path to obtain the interface coordinate point set.

[0113] Interface coordinate point set refers to the ordered three-dimensional coordinate set formed by arranging the three-dimensional spatial coordinates of the two borehole points corresponding to each layer relationship in the matching path according to the order of the layer relationship, and is used to locate the actual position of each layer interface in three-dimensional space.

[0114] Spatial coordinates refer to the three-dimensional position coordinates of each borehole point at the corresponding top interface of the formation. They can include the horizontal position coordinates of the borehole point and the depth coordinates of the top interface of the formation, and are used to uniquely determine the position of the formation interface in the three-dimensional space of the target area.

[0115] Specifically, in the modeling scenario of complex sedimentary systems in sedimentary basins, the matching path is first obtained from S1035. In the middle, each hierarchical relationship record is traversed one by one. Mark the connectivity state of the strata. Path nodes marked as normally connected Extract matching paths respectively In the time position and The three-dimensional spatial coordinates of the top interface of the stratum corresponding to the location.

[0116] Next, under the assumption of vertical drilling, the drilling point is... In the The three-dimensional spatial coordinates of the target strata at each time location are as follows: Drilling point In the The three-dimensional spatial coordinates of the target strata at each time location are as follows: ,in The coordinates of the borehole's constant position in the horizontal plane. This provides the depth coordinates of the top interfaces of different strata. Then, it records the stratigraphic relationships according to the matching path. The extracted spatial coordinates are arranged in an ordered manner according to their sequence to form a set of interface coordinate points. .

[0117] S1042. Based on the time position of the path node corresponding to each layer relationship in the matching path on the time axis, extract the thickness evolution component of the corresponding time position from the gradient function, calculate the direction vector between adjacent spatial coordinates in the interface coordinate point set, and construct the tangent vector of each spatial coordinate with the thickness evolution component as the amplitude and the direction vector as the direction.

[0118] A direction vector is a unit vector obtained by normalizing the three-dimensional vector formed by the difference between two adjacent spatial coordinates in the interface coordinate point set, and is used to represent the local extension direction of the stratigraphic interface at that spatial location. A tangent vector is a three-dimensional vector constructed with the thickness evolution component extracted from the gradient function as the amplitude and the corresponding direction vector as the direction, and is used to simultaneously carry two types of geometric constraint information: the extension direction of the stratigraphic interface at that spatial coordinate and the intensity of sedimentary thickness evolution.

[0119] Specifically, firstly, the matching path in S1035... Each normally connected path node Based on its corresponding time component on the time axis The gradient function obtained from S1025 Extract the thickness evolution component at the corresponding time position. ,in For the first The function value of the gradient function at the corresponding time point of each path node on the time axis.

[0120] Next, the interface coordinate point set adjacent spatial coordinates and Calculate the direction vector ,in For the first Normalized direction vectors at spatial coordinates Let Euclidean norm be the vector. Then, let the thickness evolution component be considered. Amplitude and direction vectors For each spatial coordinate, construct the tangent vector. ,in For the first Tangent vectors at spatial coordinates.

[0121] S1043. Extract the spatial coordinates that are closest to each grid node in the three-dimensional spatial grid from the set of interface coordinate points, and construct a geometric constraint field using the tangent vector corresponding to the spatial coordinates as the directional constraint value of the grid node.

[0122] A three-dimensional spatial grid refers to a regular discretized grid structure that covers the underground space of a target area. It can be a structured grid composed of several equally spaced grid nodes arranged in a row-column layer direction in three-dimensional space, and its resolution is preset according to the spatial range of the target area and the modeling accuracy requirements.

[0123] The directional constraint value refers to the three-dimensional vector assigned to each grid node to constrain the interpolation direction of the surface. It can be the tangent vector corresponding to the spatial coordinate closest to the grid node in the interface coordinate point set, and is used to provide gradient guidance information on the local extension direction of the formation interface for each grid node during the surface reconstruction process.

[0124] A geometric constraint field is a global spatial field formed by establishing a mapping relationship between each grid node in the three-dimensional spatial grid of the target region and its corresponding directional constraint value. It is used to provide continuously distributed directional constraint information on the three-dimensional spatial grid of the entire target region.

[0125] Specifically, firstly, based on the spatial range and modeling accuracy requirements of the target area, a structured three-dimensional spatial grid covering the underground space of the target area is constructed in three-dimensional space: in the horizontal direction, grid rows and columns are evenly divided within the horizontal boundary of the target area according to a preset horizontal resolution.

[0126] In the vertical Z-axis direction, the initial layer of the grid is determined using the measured depth coordinates of each reference borehole at the corresponding top interface of the formation as known control values. Inverse distance weighted interpolation is used to interpolate the initial depth of this layer at each node of the horizontal grid. Let the total number of grid nodes be... , No. The spatial coordinates of the grid nodes are as follows: ,in Next, for each grid node... The interface coordinate point set obtained from S1041 In the middle, through calculation Extract the nearest spatial coordinate from the Euclidean distance between each spatial coordinate. .

[0127] Then, with corresponding tangent vector As a grid node The initial directional constraint values ​​at each location are used to perform a global smooth transition on the 3D spatial mesh using inverse distance weighted interpolation, ensuring the spatial first-order differentiable continuity of the orientation field. Then, a mapping relationship is established between each mesh node and its smoothed directional constraint value to construct the geometric constraint field. .

[0128] In this embodiment, the three-dimensional spatial coordinates of the strata corresponding to each borehole point are extracted from the matching path in chronological order to form an interface coordinate point set. Then, the function value of the gradient function at the corresponding time and the direction vector between adjacent coordinates are combined to construct a tangent vector carrying sedimentary evolution intensity information. Finally, the tangent vector is assigned to each grid node in the three-dimensional spatial grid through nearest neighbor mapping to form a geometric constraint field, thus realizing the effective transformation from discrete borehole matching information to global three-dimensional spatial geometric constraints.

[0129] S105. A smooth functional equation is constructed using the interface coordinate point set as the position constraint and the geometric constraint field as the gradient guiding term. The surface parameters are obtained by minimizing the smooth functional equation. The surface parameters are then used to perform surface interpolation fitting on multiple strata in the three-dimensional spatial grid to construct a three-dimensional geological model.

[0130] Optionally, step S105, which constructs a smooth functional equation using the interface coordinate point set as position constraints and the geometric constraint field as the gradient guiding term, obtains surface parameters by minimizing the smooth functional equation, and uses the surface parameters to perform surface interpolation fitting on multiple strata in the three-dimensional spatial grid to construct a three-dimensional geological model, may specifically include: Figure 3 A flowchart illustrating a method for constructing a three-dimensional geological model according to an embodiment of this application is shown. Figure 3 As shown, when modeling multiple strata, the interface coordinate point set is used as a rigid spatial position constraint, while a geometric constraint field is introduced as a gradient guiding term for surface extension.

[0131] Specifically, in the stage of constructing the smooth functional equation representing the surface morphology, the error value from each grid node in the two-dimensional manifold mesh embedded in the three-dimensional space to the corresponding interface matching control point is first calculated based on the smooth and differentiable distance loss function as a position constraint term to ensure the fidelity of the generated surface at the key control points and avoid volumetric mesh collapse.

[0132] Simultaneously, for the geometrically constrained field that is continuously differentiable after distance-weighted smoothing, the directional constraint value of each grid node is extracted, and the sum of squared gradient deviations of the directional change rates between the node and its adjacent grid nodes is calculated. This sum is established as a gradient constraint term to control the smoothness and construction trend of the surface. Subsequently, the weight coefficients are directly retrieved from a preset lookup table to perform a weighted summation operation on the position constraint term and the gradient constraint term, thereby constructing a smooth functional equation model with the spatial position coordinates of the grid nodes as independent variables.

[0133] After obtaining the smooth functional equation, a minimization solution is performed to establish the optimal surface parameters. This solution process calculates the spatial adjustment direction and corresponding adjustment magnitude of the corresponding nodes by obtaining differentiable partial values ​​of the grid node position coordinates. Following this guidance, the position coordinates of all controlled grid nodes are iteratively updated until the calculated values ​​of the constructed smooth functional equation converge to the global minimum.

[0134] Finally, strictly following the stratigraphic relationships defined in the predetermined matching path, the surface units distributed at different depths are stacked spatially, and the grid nodes are geometrically collapsed for missing markers. This completes the construction of a three-dimensional geological model with a three-dimensional topological structure that accurately reflects the true sedimentary history. This method includes: S1051. The position constraint term is the sum of the minimum squared distances from each grid node in the three-dimensional spatial grid to the set of interface coordinate points. The gradient constraint term is the sum of the squared gradient deviations between the direction constraint value of each grid node in the geometric constraint field and the rate of change of direction between adjacent grid nodes. The position constraint term and the gradient constraint term are weighted and summed using preset weight coefficients to construct a smooth functional equation with the position coordinates of the grid nodes in the three-dimensional spatial grid as independent variables.

[0135] The position constraint term refers to the sum of the squares of the minimum Euclidean distances from each grid node in the 3D spatial grid to all spatial coordinates in the interface coordinate point set, and is used to constrain the grid nodes to be as close as possible to the actual formation interface position determined by the interface coordinate point set during the optimization process.

[0136] The gradient constraint term refers to the sum of the squared gradient deviations of the directional constraint value of each grid node in the geometric constraint field and the rate of change of direction between its adjacent grid nodes. It is used to constrain the surface formed by the grid nodes during the optimization process to remain locally smooth and to ensure that the directional change is consistent with the stratigraphic extension trend indicated by the tangent vector.

[0137] The smooth functional equation refers to a global energy functional constructed with the coordinates of grid nodes in a 3D spatial grid as independent variables and the weighted sum of position and gradient constraints as the function value. It is used to integrate interface position constraints and directional gradient constraints into a unified optimization objective function, and to achieve consistent surface reconstruction by minimizing this functional. The preset weight coefficients are shown in Table 4 below. Table 4: Preset Weighting Coefficients Comparison Table

[0138] As shown in Table 4, the weight coefficients of the location constraint terms are set in different levels according to the differences in the complexity of the geological structure of the target area. With gradient constraint term weight coefficients The combination scheme satisfies .

[0139] The complexity of geological structures is classified using objective quantitative indicators: the proportion of interfaces with a structural type component greater than or equal to 2 (i.e., angular unconformity, overlap, truncation, etc.) out of the total number of contact interfaces in the target area. ,when When it is determined to be a simple structural region, It was determined to be a moderately complex structural region when When it is determined to be a complex structural region, It was determined to be a highly complex structural region.

[0140] The weighting coefficients are set based on the following: the higher the structural complexity, the stronger the tectonic alteration of the stratigraphic interface, and the more easily simple position interpolation produces cross-layer distortion; therefore, the weights of the gradient constraint terms should be increased. To strengthen the temporal logical constraints of geological evolution and correspondingly reduce the weight of the location constraint terms. The values ​​in Table 4 are calculated according to... Follow Gear increments The rule for determining the value. Specifically, according to the above proportions. Determine the level of structural complexity of the target region, and based on this level, find the corresponding weight coefficient from the preset Table 4.

[0141] The general form of a smooth functional equation is defined as:

[0142] in, The function values ​​of the smooth functional equation are... All in the three-dimensional space mesh The position coordinate vector of each grid node For the set of interface coordinate points, and These are the weight coefficients for the position constraint term and the gradient constraint term, respectively. For the first The orientation constraint values ​​for each grid node. For the first The set of neighboring nodes of a grid node. This represents the number of adjacent nodes.

[0143] Specifically, in the modeling scenario of complex sedimentary systems in sedimentary basins, the three-dimensional spatial grid constructed in S1043 is first used as a basis, and each grid node is then... The interface coordinate point set obtained from S1041 Calculate the Euclidean distance from the node to all spatial coordinates and extract the minimum distance value. Square and for all After repeating this operation on each mesh node and summing the results, the position constraint term is obtained. .

[0144] Next, for each grid node Geometric constraint field constructed from S1043 Extract the corresponding directional constraint value Identify the set of neighboring nodes of the given node in a 3D spatial mesh. Calculate the average value of the directional constraint values ​​of adjacent nodes. The average value is compared with the current node's orientation constraint value. After taking the difference, calculate the square of the Euclidean norm for all... This operation is repeated for each grid node, and the summation is performed to obtain the gradient constraint term. .

[0145] Then, based on the geological structural complexity of the target area, the weighting coefficient combination is determined with reference to Table 4. For example, for a moderately complex tectonic zone, the weighting coefficient combination is set as follows: , Using these two weighting coefficients for the position constraint term With gradient constraint terms We perform weighted summation to construct a smooth functional equation. .

[0146] S1052. Based on the smooth functional equation, the position adjustment direction and adjustment range of each grid node are determined by calculating the derivative value of the position coordinates of each grid node in the three-dimensional space grid. The position coordinates of the corresponding grid node are iteratively updated along the position adjustment direction and according to the adjustment range until the value of the smooth functional equation reaches the minimum value. The target position coordinates of the grid node are then determined as surface parameters.

[0147] The position adjustment direction refers to the negative direction of the gradient vector obtained by taking the partial derivative of the smooth functional equation with respect to the position coordinates of a grid node. It indicates the direction in which the grid node should move during the optimization iteration to reduce the function value of the smooth functional equation. The adjustment magnitude refers to the step size when updating the position of the grid node in the position adjustment direction, and it controls the amount of change in the grid node's position in each iteration.

[0148] The surface parameters refer to the set of final position coordinates of each grid node in the three-dimensional space grid when it converges to a stable state after the smooth functional equation is minimized and iterated. They are used to uniquely determine the optimal surface shape that satisfies both position and gradient constraints.

[0149] Specifically, firstly, based on the smooth functional equations constructed in S1051 For each grid node in the three-dimensional spatial grid Position coordinates Calculate the partial derivatives to obtain the gradient vector at that node. Take its negative direction as the position adjustment direction of the node. .

[0150] Next, set the learning rate λ and calculate the adjustment range. ,in Let λ be the Euclidean norm of the gradient vector. The initial value of the learning rate λ is... An adaptive adjustment is performed using a backtracking search method: in each iteration, the grid node positions are first tentatively updated with the current λ; if the function value of the smooth functional equation does not decrease after the update, then... Then let And retry until the function value decreases. If the function value decreases steadily after multiple iterations, then let... To accelerate convergence, the step size should be increased appropriately, while an upper limit should be set. To prevent iterative divergence. Then, adjust the direction along the position. According to the adjustment range For grid nodes Update the location coordinates:

[0151] in, For the first During the nth iteration The position coordinates of each grid node The updated position coordinates are used. Repeat the above iterative update process, monitoring the changes in the function values ​​of the smooth functional equation, until the convergence condition is met. like Alternatively, iteration can stop when a preset maximum number of iterations, such as 1000, is reached. At this point, the position coordinates of each grid node have converged to the minimum point of the smooth functional equation, and the final position coordinates of all grid nodes are set. The parameters are determined to be surface parameters.

[0152] S1053. Connect the adjacent mesh nodes corresponding to the surface parameters to obtain surface elements. Stack the surface elements of different depths according to the layer relationship in the matching path to obtain a three-dimensional geological model.

[0153] A surface element refers to a basic surface patch formed by connecting adjacent grid nodes in a three-dimensional space grid according to the final position coordinates determined by the surface parameters and the grid topology. It can be a triangular or quadrilateral surface patch formed by connecting three or four adjacent grid node vertices and is used to discretize the surface morphology of a certain stratigraphic interface in a local space.

[0154] A three-dimensional geological model is a complete three-dimensional solid model formed by stacking and combining the curved units corresponding to multiple strata interfaces in the target area according to the vertical arrangement order of the strata recorded in the matching path. It is used to comprehensively reflect the spatial distribution, contact relationship and geological evolution logic of multiple underground strata in the target area.

[0155] Specifically, firstly, based on the surface parameters obtained in S1052 Based on the topology of the 3D spatial mesh, the adjacent nodes of each mesh node are identified, and the final position coordinates of each group of adjacent nodes are connected. Next, based on the structured node topology of the 2D manifold mesh, four adjacent mesh nodes belonging to the same stratigraphic interface are linked in row and column order. , , , The connections form a quadrilateral curved surface unit, in which This represents the number of column nodes in the surface mesh.

[0156] By performing this connection operation on all mesh nodes of a specific stratigraphic level, a set of surface elements covering the stratigraphic interface is obtained. ,in This represents the total number of surface elements. Then, the matching path obtained in step S1035... Extract the temporal location and depth information of the strata from the stratigraphic relationship records. Stack the sets of surface units at different depths in spatial dimension according to the vertical arrangement of the strata from deep to shallow.

[0157] For the missing strata markers in the matching path In the corresponding region, the surface elements of the missing region are geometrically overlapped with the underlying effective surface elements in the vertical direction, and the layer thickness of this region is set to zero. This is to intuitively and topologically consistently represent the overburden or truncation pinch-out phenomenon of the strata, ultimately forming a three-dimensional geological model with a clear sequence relationship and a realistic geological evolution logic. .

[0158] This embodiment realizes the collaborative consistency surface reconstruction of multi-layer depositional systems with complex contact relationships such as overlay and truncation, effectively solving the technical problem that discrete borehole data is difficult to achieve topological consistency connection in complex structural regions.

[0159] Figure 4 This is a schematic diagram illustrating a specific implementation of a three-dimensional geological model construction system based on multi-source heterogeneous geological data, as provided in this application embodiment. (Refer to...) Figure 4 The system may include: The acquisition module 410 is used to acquire age data and contact type data of different strata at each borehole point in the target area, as well as the sedimentary facies evolution rate of each borehole point. The generation module 420 is used to numerically map the age data and contact type data to obtain attribute vectors of different strata. By fusing the attribute vectors with the gradient function of the stratum deposition thickness determined according to the sedimentary facies evolution rate on the time axis, a multidimensional time series sequence of each borehole point is obtained. The generation module 420 is also used to calculate the cumulative distance measure between multidimensional time series using dynamic time warping, search for the alignment path with the minimum total cost between different multidimensional time series based on the cumulative distance measure, and determine the missing position of the formation corresponding to each borehole point by extracting the discontinuous interval of the alignment path, thereby obtaining the matching path between the formations of different borehole points. The mapping module 430 is used to extract the spatial coordinates of each borehole point in the corresponding stratum from the matching path to form an interface coordinate point set, convert the gradient function of the corresponding stratum at the alignment time into the tangent vector at the corresponding spatial coordinate in the interface coordinate point set and map it to the three-dimensional spatial grid of the target area to obtain the geometric constraint field. Module 440 is used to construct a smooth functional equation with the interface coordinate point set as the position constraint and the geometric constraint field as the gradient guiding term. The surface parameters are obtained by minimizing the smooth functional equation, and the surface parameters are used to perform surface interpolation fitting on multiple strata in the three-dimensional spatial grid to construct a three-dimensional geological model.

[0160] The three-dimensional geological model construction system based on multi-source heterogeneous geological data in this application embodiment is used to implement the aforementioned three-dimensional geological model construction method based on multi-source heterogeneous geological data. Therefore, the specific implementation of the three-dimensional geological model construction system based on multi-source heterogeneous geological data can be found in the embodiment section of the three-dimensional geological model construction method based on multi-source heterogeneous geological data above. The specific implementation can be referred to the description of the corresponding embodiment, which will not be repeated here.

[0161] Figure 5 A schematic diagram of the hardware structure of an electronic device provided in one embodiment of this application is shown.

[0162] The electronic device may include a processor 510 and a memory 520 storing computer program instructions.

[0163] Specifically, the processor 510 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0164] Memory 520 may include mass storage for data or instructions. For example, and not limitingly, memory 520 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 520 may include removable or non-removable (or fixed) media. Where appropriate, memory 520 may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, memory 520 is non-volatile solid-state memory.

[0165] Memory may include read-only memory (ROM), random access memory (RAM), disk storage media devices, optical storage media devices, flash memory devices, and electrical, optical, or other physical / tangible memory storage devices. Therefore, typically, memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to the first aspect of this disclosure.

[0166] The processor 510 reads and executes computer program instructions stored in the memory 520 to implement any of the three-dimensional geological model construction methods based on multi-source heterogeneous geological data in the above embodiments.

[0167] In one example, the electronic device may also include a communication interface 530 and a bus 540. Wherein, such as Figure 5 As shown, the processor 510, memory 520, and communication interface 530 are connected through bus 540 and complete communication with each other.

[0168] The communication interface 530 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0169] Bus 540 includes hardware, software, or both, that couples components of an online data traffic metering device together. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 540 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, any suitable bus or interconnect is contemplated herein.

[0170] The electronic device can execute the three-dimensional geological model construction method based on multi-source heterogeneous geological data in the embodiments of this application, thereby realizing the three-dimensional geological model construction method based on multi-source heterogeneous geological data described in conjunction with the accompanying drawings.

[0171] Furthermore, in conjunction with the three-dimensional geological model construction method based on multi-source heterogeneous geological data in the above embodiments, this application embodiment can provide a computer-readable storage medium for implementation. This computer-readable storage medium stores computer program instructions; when executed by a processor, these computer program instructions implement any of the three-dimensional geological model construction methods based on multi-source heterogeneous geological data in the above embodiments.

[0172] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0173] The functional blocks shown in the above-described structural diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0174] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0175] The aspects of this application have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by dedicated hardware performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0176] The foregoing has provided a detailed description of a method and system for constructing a three-dimensional geological model based on multi-source heterogeneous geological data, as provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and its core ideas. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from its principles, and these improvements and modifications also fall within the protection scope of this application.

Claims

1. A method for constructing a three-dimensional geological model based on multi-source heterogeneous geological data, characterized in that, include: Acquire age and contact type data of different strata at each borehole point in the target area, as well as the sedimentary facies evolution rate at each borehole point; The age data and the contact type data are numerically mapped to obtain attribute vectors for different strata. By fusing the attribute vectors with the gradient function of the stratum deposition thickness determined according to the sedimentary facies evolution rate on the time axis, a multidimensional time series sequence for each borehole point is obtained. The cumulative distance measure between the multidimensional time series is calculated using dynamic time warping. Based on the cumulative distance measure, the alignment path with the minimum total cost between different multidimensional time series is searched. By extracting the discontinuous intervals of the alignment path, the missing position of the formation corresponding to each borehole point is determined, and the matching path between the formations of different borehole points is obtained. The spatial coordinates of each borehole point in the corresponding stratum are extracted from the matching path to form an interface coordinate point set. The gradient function of the corresponding stratum at the alignment time is converted into a tangent vector at the corresponding spatial coordinate in the interface coordinate point set and mapped to the three-dimensional spatial grid of the target area to obtain the geometric constraint field. A smooth functional equation is constructed using the interface coordinate point set as the position constraint and the geometric constraint field as the gradient guiding term. The surface parameters are obtained by minimizing the smooth functional equation, and the surface parameters are used to perform surface interpolation fitting on multiple strata in the three-dimensional spatial grid to construct a three-dimensional geological model.

2. The method according to claim 1, characterized in that, The step of numerically mapping the age data and the contact type data to obtain attribute vectors for different strata includes: Extract the time tag of the stratum corresponding to each of the said ages, and based on the time tag, obtain the time component of each stratum according to the preset correspondence between the time tag and the time component; Extract the interface property information of each contact type data, and based on the interface property information, obtain the structural type component of each stratum according to the preset correspondence between interface property information and structural type components. Based on the arrangement of strata in vertical space, the attribute vector of each stratum is constructed according to the time component and the structure type component.

3. The method according to claim 1, characterized in that, The process involves fusing the attribute vector with a gradient function of the formation depositional thickness determined based on the sedimentary facies evolution rate on the time axis to obtain a multidimensional time series sequence for each borehole point, including: Calculate the slope of the sedimentary facies evolution rate at each time node corresponding to each borehole point, and based on the slope, determine the thickness evolution component at each time node according to the preset mapping relationship between the slope and the thickness evolution component. The gradient function is obtained by arranging and curve fitting the thickness evolution components of each time node according to the chronological order of the time axis. The time axis is determined based on the continuous distribution span of the time labels of the era data in the time dimension. Based on the time label of the stratum corresponding to the attribute vector, the mapping interval of the attribute vector on the time axis is determined, the target thickness evolution component corresponding to the mapping interval is extracted from the gradient function, and a multidimensional feature vector is constructed based on the target thickness evolution component and the corresponding attribute vector. The multidimensional feature vectors are arranged in chronological order according to the time axis to construct the multidimensional time sequence of each drilling point.

4. The method according to claim 1, characterized in that, The calculation of the cumulative distance measure between the multidimensional time series using dynamic time warping includes: Calculate the sum of squares of the differences between each component of the multidimensional feature vector in the multidimensional time sequence of any two borehole points to obtain the point-to-point distance value. Arrange all the point-to-point distance values ​​according to the time position of each multidimensional feature vector on the time axis to obtain the distance matrix. The cumulative distance measure for each time position is obtained by accumulating the point-to-point distance value with the smallest value at the previous time position in the distance matrix along the time axis and the point-to-point distance value at the current time position.

5. The method according to claim 1, characterized in that, The step of searching for the alignment path with the minimum total cost among different multidimensional time series based on the cumulative distance metric, and obtaining the matching path between different borehole points by extracting the discontinuous intervals of the alignment path to determine the missing positions of the formation corresponding to each borehole point, includes: The preceding time position corresponding to the cumulative distance measure with the smallest value is extracted from each time position along the reverse direction of the time axis as a path node. All the path nodes are arranged in chronological order according to the time axis to obtain the aligned path. Calculate the time interval between adjacent path nodes in the alignment path on the time axis, and determine the non-contiguous interval between adjacent path nodes whose time interval is greater than a preset interval threshold as the missing position; Based on the multidimensional feature vector corresponding to each path node in the alignment path, a stratigraphic relationship is established between the two borehole points. The time position corresponding to the missing position is marked as a stratigraphic missing identifier in the stratigraphic relationship, and the stratigraphic relationship is arranged according to the order of the path nodes in the alignment path to obtain the matching path.

6. The method according to claim 1, characterized in that, The spatial coordinates of each borehole point in the corresponding stratum are extracted from the matching path to form an interface coordinate point set. The gradient function of the corresponding stratum at the alignment time is converted into a tangent vector at the corresponding spatial coordinate in the interface coordinate point set and mapped to the three-dimensional spatial mesh of the target region to obtain the geometric constraint field, including: Extract the spatial coordinates of the two borehole points associated with each layer relationship in the matching path in the corresponding stratum, and arrange all the spatial coordinates according to the order of the layer relationships in the matching path to obtain the interface coordinate point set; Based on the time position of the path node corresponding to each layer relationship in the matching path on the time axis, the thickness evolution component corresponding to the time position is extracted from the gradient function, the direction vector between adjacent spatial coordinates in the interface coordinate point set is calculated, and the tangent vector of each spatial coordinate is constructed with the thickness evolution component as the amplitude and the direction vector as the direction. Extract the spatial coordinates that are closest to each grid node in the three-dimensional spatial grid from the set of interface coordinate points, and construct a geometric constraint field using the tangent vector corresponding to the spatial coordinates as the directional constraint value of the grid node.

7. The method according to claim 1, characterized in that, A smooth functional equation is constructed using the interface coordinate point set as position constraints and the geometric constraint field as gradient guiding terms. Surface parameters are obtained by minimizing the smooth functional equation, and these surface parameters are then used to perform surface interpolation fitting on multiple strata in the three-dimensional spatial grid to construct a three-dimensional geological model, including: The position constraint term is the sum of the minimum squared distances from each grid node in the three-dimensional spatial grid to the set of interface coordinate points. The gradient constraint term is the sum of the squared gradient deviations between the direction constraint value of each grid node in the geometric constraint field and the rate of change of direction between adjacent grid nodes. The position constraint term and the gradient constraint term are weighted and summed using preset weight coefficients to construct a smooth functional equation with the position coordinates of the grid nodes in the three-dimensional spatial grid as independent variables. Based on the smooth functional equation, by calculating the derivative value of the position coordinates of each grid node in the three-dimensional space grid, the position adjustment direction and adjustment range of each grid node are determined. The position coordinates of the corresponding grid node are iteratively updated along the position adjustment direction and according to the adjustment range until the value of the smooth functional equation reaches the minimum value. The target position coordinates of the grid node are then determined as the surface parameters. Connect the adjacent mesh nodes corresponding to the surface parameters to obtain surface units. Stack the surface units of different depths according to the layer relationship in the matching path to obtain the three-dimensional geological model.

8. A three-dimensional geological model construction system based on multi-source heterogeneous geological data, characterized in that, include: The acquisition module is used to acquire age data and contact type data of different strata at each borehole point in the target area, as well as the sedimentary facies evolution rate of each borehole point; The generation module is used to numerically map the age data and the contact type data to obtain attribute vectors of different strata. By fusing the attribute vectors with the gradient function of the stratum deposition thickness determined according to the sedimentary facies evolution rate on the time axis, a multidimensional time series sequence of each borehole point is obtained. The generation module is also used to calculate the cumulative distance measure between the multidimensional time series using dynamic time warping, search for the alignment path with the minimum total cost between different multidimensional time series based on the cumulative distance measure, and determine the missing position of the formation corresponding to each borehole point by extracting the non-continuous interval of the alignment path, thereby obtaining the matching path between the formations of different borehole points. The mapping module is used to extract the spatial coordinates of each borehole point in the corresponding stratum from the matching path to form an interface coordinate point set, convert the gradient function of the corresponding stratum at the alignment time into a tangent vector at the corresponding spatial coordinate in the interface coordinate point set and map it to the three-dimensional spatial grid of the target area to obtain a geometric constraint field. The construction module is used to construct a smooth functional equation with the interface coordinate point set as the position constraint and the geometric constraint field as the gradient guiding term. The surface parameters are obtained by minimizing the smooth functional equation, and the surface parameters are used to perform surface interpolation fitting on multiple strata in the three-dimensional spatial grid to construct a three-dimensional geological model.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to execute the computer program to implement the steps of the method for constructing a three-dimensional geological model based on multi-source heterogeneous geological data as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, enables the method for constructing a three-dimensional geological model based on multi-source heterogeneous geological data as described in any one of claims 1 to 7.

Images