A method and system for calculating the hydrothermal intensity index TII in a reservoir

Abstract

The present invention belongs to the field of geological exploration and mineral resource development, and particularly relates to a method and system for calculating the hydrothermal intensity index TII in a reservoir, aiming to solve the problem that the prior art cannot quickly and accurately calculate the hydrothermal activity intensity of the reservoir. The method includes: collecting data related to reservoir characteristics in an oil and gas field as input data; preprocessing the input data to obtain preprocessed data; inputting the preprocessed data into a trained TII prediction model to obtain a predicted value of the comprehensive hydrothermal intensity index TII; the TII prediction model is constructed based on a deep gradient boosting network; the TII prediction model includes a first hidden layer, a second hidden layer, and an output layer. The present invention improves the speed and accuracy of calculating the hydrothermal activity intensity of the reservoir.

Description

Technical Field

[0001] The present invention belongs to the field of geological exploration and mineral resource development, and particularly relates to a method and a system for calculating the hydrothermal intensity index TII in a reservoir. Background Art

[0002] Hydrothermal activity refers to the movement and interaction of hot fluids (such as water, gas, and dissolved minerals) in the earth's crust under specific temperature and pressure conditions. This activity plays a crucial role in the formation and evolution of ore deposits. Especially in the formation of metal mineral resources, hydrothermal fluids are often important ore-forming media.

[0003] However, the existing hydrothermal intensity calculation methods mainly focus on the analysis of single parameters and lack the comprehensive consideration of various geological features. Traditional hydrothermal activity calculations usually rely on empirical rules or qualitative analysis, which makes researchers face the following problems when judging hydrothermal intensity: Diversity and complexity of data: There are various factors affecting hydrothermal activity in the reservoir, such as temperature, pressure, chemical composition, etc. A single index is difficult to comprehensively reflect hydrothermal intensity. Lack of systematic calculation methods: Existing methods often use linear or simple threshold judgments and cannot make full use of advanced statistical and machine learning techniques for in-depth analysis. Furthermore, the main controlling factors of hydrothermal migration cannot be determined, resulting in low efficiency and accuracy in calculating the total geothermal intensity of the entire reservoir. Based on this, the present invention proposes a deep gradient boosting network (DGBN), which improves the prediction accuracy and model generalization ability by constructing a multi-layer structure in the gradient boosting tree (GBT) model. Designing this multi-layer structure includes determining the number of layers, defining the functions of each layer, constructing and training the GBT model of each layer, defining the loss function and optimization algorithm, and evaluating and adjusting the model performance, so as to form a deep integrated model to gradually improve the prediction accuracy and optimize the overall performance. Summary of the Invention

[0004] To solve the above problems in the prior art, that is, to solve the problem that the prior art cannot quickly and accurately calculate the hydrothermal activity intensity of the reservoir, in the first aspect of the present invention, a method for calculating the hydrothermal intensity index TII in the reservoir is proposed. The method includes:

[0005] S100, collecting data related to reservoir characteristics in the oil and gas field as input data; the input data includes parameters for characterizing the hydrothermal source and origin, parameters reflecting the chemical composition of the fluid and its evolution process, parameters affecting the thermodynamic properties of the reservoir, and parameters related to the phase state and physical properties of the reservoir fluid;

[0006] S200, preprocessing the input data to obtain preprocessed data;

[0007] S300, input the preprocessed data into the trained TII prediction model to obtain the predicted value of the comprehensive hydrothermal intensity index TII:

[0008] The TII prediction model is constructed based on a deep gradient boosting network; the TII prediction model includes a first hidden layer, a second hidden layer, and an output layer; the first hidden layer is used to standardize the preprocessed data to obtain standardized data; the second hidden layer is used to perform weighted and non-linear combination on the standardized data to obtain the hydrothermal intensity index TII as the initial TII; the output layer contains multiple GBT models; the output layer is used to calculate the residuals of each GBT model, use the residuals as the target value, and combine the hydrothermal intensity index TII output by the previous layer of GBT model to update the GBT model of the next layer, and then obtain the predicted value of the comprehensive hydrothermal intensity index TII.

[0009] In some preferred embodiments, the preprocessing includes data cleaning, missing value and outlier elimination.

[0010] In some preferred embodiments, the parameters for characterizing the hydrothermal source and origin include the trace element δEu; the parameters for reflecting the chemical composition and evolution process of the fluid include the oxygen isotope ratio δ 18 O; the parameters affecting the thermodynamic properties of the reservoir include the reservoir temperature; the parameters related to the phase state and physical properties of the reservoir fluid include the reservoir pressure.

[0011] In some preferred embodiments, the method for performing weighted and non-linear combination on the standardized data to obtain the hydrothermal intensity index TII is as follows:

[0012] ;

[0013] where, represents the hydrothermal intensity index TII, p represents the non-linear adjustment factor, , , , respectively represent the standardized trace element δEu, oxygen isotope ratio δ 18 O, reservoir temperature, reservoir pressure, k 1 represents the influence strength of δEu on TII, α and β respectively represent the weights of δEu and δ 18 O on TII, γ represents the weight of the reservoir temperature, represents the weight of the reservoir pressure, and c is a constant used to adjust the baseline value.

[0014] In some preferred embodiments, the calculation method of the adaptive dynamic weight is as follows:

[0015] ;

[0016] Among them, represents the weight of the feature, i.e., α, β, γ, , are parameters used to characterize the hydrothermal source and origin, which are δEu, δ 18 O, reservoir temperature or reservoir pressure, , both represent weight coefficients, represents the feature importance evaluation transformation value of the nth layer, is the n weight parameter of the layer.

[0017] In some preferred embodiments, the predicted value of the comprehensive hydrothermal intensity index TII is obtained by the following method:

[0018] ;

[0019] ;

[0020] Among them, represents the predicted value of the comprehensive hydrothermal intensity index TII, represents the weight of each loss term, represents the n predicted value of TII of the nth layer, is the input vector composed of δEu, δ 18 O, reservoir temperature T, and reservoir pressure P, is the output value of the internal non-linear transformation of the residual block, corresponding to the prediction correction amount of the network for the features of this layer, and is the prediction residual of the nth layer, is the standardized inter-layer residual calculated based on geological prior knowledge, reflecting the offset of the actual rock physical properties parameters, and represents the actual residual, represents the factor controlling the influence of the residual on the total loss, represents the n hydrothermal intensity index TII of the nth layer, represents an optimization operation, and the optimal residual function is found through the optimization process,

[0021] In some preferred embodiments, the loss function of the TII prediction model during training is:

[0022] ;

[0023] ;

[0024] ;

[0025] Among them, represents the loss value, that is, the prediction residual, is the true TII value. In the actual application process, the true TII value is the standard value set according to the geological conditions, is the TII value predicted by the model, and 𝑛 is the total number of test samples, the features output by the (n - 1)-th layer, is the convolutional kernel weight matrix, extracting geological features, that is, the features of the parameters used to characterize the hydrothermal source and origin, R n2 is the convolutional kernel weight matrix, expanding the features back to the original dimension, and ReLU() is the activation function, , are both bias parameters, is the measured geological feature at the k-th sampling point, is the standard value set according to the geological conditions, N is the total number of sampling points.

[0026] In some preferred embodiments, after step S300, it further includes obtaining the main controlling factors of hydrothermal migration:

[0027] Using seismic data to extract the connectivity information of faults and fractures, TII data, interpolating them into a grid, and calculating the fault dominance and fracture network dominance. According to the fault dominance and the fracture network dominance, determine the main controlling factors of hydrothermal migration; the main controlling factors include faults and fracture networks;

[0028] The method for obtaining the fault dominance is:

[0029] ;

[0030] Among them, represents the fault dominance, represents the overlapping area between the i-th TII high-value area and the fault area. The TII high-value area is the area with significant hydrothermal activity intensity, that is, the area where TII is greater than the set intensity, and the fault area is the strip-shaped area extending along the fault strike and on both sides, represents the average sliding rate at the overlapping part between the i-th TII high-value area and the fault area, represents the maximum sliding rate in the faults in all overlapping areas, represents the average value of TII in the overlapping area between the i-th TII high-value area and the fault area, represents the maximum TII value in the faults in all overlapping areas, represents the total area of the fault area;

[0031] The method for obtaining the fracture network dominance is:

[0032] ;

[0033] Among them, represents the main control degree of the fracture network, K represents the total number of fracture intersection nodes, and TII 交汇点i represents the TII value of the i-th intersection node, and TII 围岩值 represents the average value of TII in the non-fracture area within the model range.

[0034] In some preferred embodiments, after step S300, it further includes calculating the total geothermal intensity of the entire reservoir:

[0035] ;

[0036] Among them, T represents the total geothermal intensity of the entire reservoir, represents the i , j , k hydrothermal intensity of the grid cell, Δ x , Δ y , Δ z are the intervals in the x, y, and z directions of the dissection cells in the study area respectively, m , n , l are the total numbers of dissection grid cells in the x, y, and z directions respectively.

[0037] In the second aspect of the present invention, a system for calculating the hydrothermal intensity index TII in a reservoir is proposed, and the system includes:

[0038] A data acquisition module configured to collect data related to reservoir characteristics in an oil and gas field as input data; the input data includes parameters for characterizing the source and origin of hydrothermal fluids, parameters reflecting the chemical composition and evolution process of fluids, parameters affecting the thermodynamic properties of the reservoir, and parameters related to the phase state and physical properties of reservoir fluids;

[0039] A preprocessing module configured to preprocess the input data to obtain preprocessed data;

[0040] A model prediction module configured to input the preprocessed data into a trained TII prediction model to obtain a predicted value of the comprehensive hydrothermal intensity index TII:

[0041] The TII prediction model is constructed based on a deep gradient boosting network; the TII prediction model includes a first hidden layer, a second hidden layer, and an output layer; the first hidden layer is used to perform normalization processing on the preprocessed data to obtain normalized data; the second hidden layer is used to perform weighted sum and non-linear combination on the normalized data to obtain the hydrothermal intensity index TII as the initial TII; the output layer contains multiple GBT models; the output layer is used to calculate the residuals of each GBT model, use the residuals as the target value, and combine the hydrothermal intensity index TII output by the previous GBT model to update the GBT model of the next layer, and then obtain the predicted value of the comprehensive hydrothermal intensity index TII;

[0042] Among them, the input of the first GBT model is the initial TII.

[0043] Advantages of the present invention:

[0044] The present invention improves the speed and accuracy of calculating the hydrothermal activity intensity of the reservoir.

[0045] 1) This method first collects data related to reservoir characteristics, including characteristic variables such as δEu, δ 18 O, temperature, pressure, etc., and strictly ensures the data quality, eliminates outliers through preprocessing to improve the reliability of the analysis; then, uses the improved algorithm DGBN to calculate the hydrothermal intensity index (TII). This algorithm analyzes the main parameters affecting the TII value and their specific contributions by constructing multiple levels of gradient boosting tree (GBT) models and combining various statistical and machine learning techniques; this process not only improves the calculation accuracy of TII, but also lays a foundation for subsequent in-depth analysis, such as obtaining the main controlling factors of hydrothermal migration and calculating the total geothermal intensity of the entire reservoir;

[0046] 2) The present invention also classifies the calculated TII values and formulates reservoir effectiveness evaluation criteria. These classification results will provide guidance for oil and gas exploration, helping oilfield workers optimize resource allocation and development strategies. By applying this method, the hydrothermal intensity of different reservoirs can be effectively evaluated, so as to identify areas with higher development potential. The present invention not only provides a novel method for calculating the hydrothermal intensity index, but also has important practical value in reservoir evaluation, can promote the efficient development of oil and gas resources, and has broad application prospects and economic benefits.

[0047] 3) The present invention constructs a quantitative model to quickly evaluate the hydrothermal activity intensity of the reservoir, provides a scientific basis for the development, management and optimization of the reservoir, and helps to achieve the efficient utilization of underground resources. Description of the Drawings

[0048] Other features, objects, and advantages of the present application will become more apparent from the following detailed description of non - limiting embodiments read in conjunction with the accompanying drawings.

[0049] Figure 1 is a schematic flow chart of a method for calculating the hydrothermal intensity index TII in a reservoir according to an embodiment of the present invention;

[0050] Figure 2 is a detailed schematic flow chart of a method for calculating the hydrothermal intensity index TII in a reservoir according to an embodiment of the present invention;

[0051] Figure 3 is a schematic structural diagram of a TII prediction model according to an embodiment of the present invention;

[0052] Figure 4 is a schematic comparison diagram of the true value and the predicted value of a method for quantitatively characterizing the hydrothermal intensity index according to an embodiment of the present invention;

[0053] Figure 5 is a schematic plan view of a reservoir affected by hydrothermal fluids calculated according to the TII index according to an embodiment of the present invention;

[0054] Figure 6 is a schematic diagram of the correlation factor ranking of the influencing parameters of the hydrothermal intensity index according to an embodiment of the present invention. Detailed Embodiments

[0055] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0056] The following further elaborates the present application with reference to the accompanying drawings and embodiments. It can be understood that the specific embodiments described herein are only for explaining the relevant invention and not for limiting the invention. Additionally, it should be noted that for the sake of description, only parts related to the relevant invention are shown in the drawings.

[0057] It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other.

[0058] The method for calculating the hydrothermal intensity index TII of the present invention, as Figure 1 shown, includes the following steps:

[0059] S100, collect data related to reservoir characteristics in the oil and gas field as input data; the input data includes parameters for characterizing the hydrothermal source and origin, parameters reflecting the chemical composition of the fluid and its evolution process, parameters affecting the thermodynamic properties of the reservoir, and parameters related to the phase state and physical properties of the reservoir fluid;

[0060] S200, preprocess the input data to obtain preprocessed data;

[0061] S300, input the preprocessed data into the trained TII prediction model to obtain the predicted value of the comprehensive hydrothermal intensity index TII:

[0062] The TII prediction model is constructed based on a deep gradient boosting network; the TII prediction model includes a first hidden layer, a second hidden layer, and an output layer; the first hidden layer is used to standardize the preprocessed data to obtain standardized data; the second hidden layer is used to perform weighted and non-linear combination on the standardized data to obtain the hydrothermal intensity index TII as the initial TII; the output layer contains multiple GBT models; the output layer is used to calculate the residuals of each GBT model, use the residuals as the target value, and combine the hydrothermal intensity index TII output by the previous GBT model to update the GBT model of the next layer, and then obtain the predicted value of the comprehensive hydrothermal intensity index TII;

[0063] Among them, the input of the first GBT model is the initial TII.

[0064] To more clearly illustrate a method for calculating the hydrothermal intensity index TII in a reservoir of the present invention, the following combines the attached Figure 2 , and details each step in an embodiment of the method of the present invention.

[0065] The present invention aims to provide a rapid calculation method for reservoir effectiveness based on the hydrothermal intensity index (TII), and mainly achieves the following objectives: By constructing a quantification model, quickly calculate the hydrothermal activity intensity of the reservoir, thus significantly shortening the time and resources required by traditional calculation (i.e., evaluation) methods. Provide a scientific basis for the development, management, and optimization of the reservoir, and help achieve the efficient utilization of underground resources. Provide accurate and reliable reservoir effectiveness evaluation for engineers and decision-makers, support the formulation of more effective mining strategies, reduce risks, and improve economic benefits. Provide a new research perspective on the interaction between reservoir characteristics and hydrothermal activities, and promote the theoretical development and practical application in related fields such as geology and mineral resource development. Adopt advanced machine learning technology, combined with cross-validation and feature importance analysis, to ensure the accuracy and generalization ability of the established model, thereby enhancing its reliability in practical applications. By achieving these objectives, the present invention will provide strong technical support for the sustainable development and management of underground resources. Specifically as follows:

[0066] S100, collect data related to reservoir characteristics in an oil and gas field as input data; the input data includes parameters for characterizing the source and origin of hydrothermal fluids, parameters reflecting the chemical composition and evolution process of fluids, parameters affecting the thermodynamic properties of the reservoir, and parameters related to the phase state and physical properties of reservoir fluids.

[0067] In this embodiment, data related to reservoir characteristics are collected from an actual oil and gas field, including but not limited to characteristic variables such as trace element δEu, oxygen isotope ratio δ 18 O, reservoir temperature, and pressure. These parameters are considered the main factors affecting the intensity of hydrothermal activity. Trace element δEu: δEu refers to the isotope ratio of europium in the reservoir and is used as an indicative parameter for the intensity of hydrothermal activity. Research shows that when δEu > 1, it indicates that there may be strong hydrothermal activity in the reservoir because the distribution of europium is significantly affected by hydrothermal fluids. δ 18 O value in calcite: δ 18 O refers to the ratio of oxygen isotopes and is usually used to indicate the source and temperature of fluids. When δ 18 O < -10‰, it indicates a possible low-temperature hydrothermal environment, suggesting that the fluid has experienced a mineralization process at a lower temperature. Temperature (T): The temperature of hydrothermal activity is an important factor affecting mineral formation and the reaction rate of chemical reactions. Temperature is usually expressed in degrees Celsius (°C) and is directly related to the physical and chemical properties of the fluid. Pressure (P): The pressure in the reservoir is a key parameter affecting hydrothermal flow and the mineralization process. Pressure is usually measured in megapascals (MPa) and directly affects the phase state of the fluid and chemical reactions.

[0068] S200, preprocess the input data to obtain preprocessed data;

[0069] In this embodiment, after data collection, data cleaning is performed to ensure the quality of the data input into the model, and the accuracy and reliability of the data are improved by eliminating missing values, outliers, and unreasonable data.

[0070] S300, input the preprocessed data into a trained TII prediction model to obtain a predicted value of the comprehensive hydrothermal intensity index TII.

[0071] In this embodiment, the TII prediction model is constructed based on the deep gradient boosting network DGBN; it includes a first hidden layer, a second hidden layer, and an output layer, as Figure 3 shown;

[0072] The first hidden layer (i.e., the first GBT layer) is used to standardize the preprocessed data to obtain standardized data;

[0073] In the present invention, the Z-score normalization method is preferably used to normalize geological parameters with different units and dimensions to ensure effective comparison of each parameter on the same scale. Specifically, the formula for Z-score normalization is: The formula for Z-score normalization is:

[0074] ;

[0075] where X is the original data value, μ is the sample mean, and σ is the sample standard deviation. Through this formula, the obtained normalized value Z will follow the standard normal distribution (mean of 0 and standard deviation of 1), making each parameter comparable when comparing.

[0076] The second hidden layer (the second GBT layer); used to perform weighted and non-linear combination on the normalized data to obtain the hydrothermal intensity index TII as the initial TII; specifically as follows:

[0077] ;

[0078] where, represents the hydrothermal intensity index TII, p represents the non-linear adjustment factor, controlling the influence form of δ 18 O on TII, which can be determined through data analysis, 、 、 、 respectively represent the normalized trace element δEu, oxygen isotope ratio δ 18 O, reservoir temperature, and reservoir pressure, k 1 represents the influence strength of δEu on TII, which is a positive value and is used to amplify or reduce the influence degree of δEu on TII, α represents the weight of δEu on TII, reflecting its relative importance in the hydrothermal intensity evaluation. Through the training data, its optimal value can be determined, β represents the weight of δ 18 O on TII, similar to α, obtained through data training, γ represents the reservoir temperature weight, represents the weight of the reservoir pressure, and c is a constant used to adjust the baseline value to ensure that the model output conforms to the actual situation.

[0079] The present invention is based on the Tanh and Sigmoid activation functions, combined with a hierarchical adaptive adjustment mechanism, and dynamically adjusts the weights according to the feature importance of different levels. Each level performs non-linear transformation on the features to varying degrees, enabling the output of each layer to better adapt to the different structures of the data. For the input feature set , calculate the importance of each feature to the model prediction, and assign a dynamic weight to each feature based on the self-attention mechanism. This weight is calculated by the following formula:

[0080] ;

[0081] Among them, represents the weight of the feature, namely α, β, γ, , are parameters used to characterize the hydrothermal source and origin, which are δEu, δ 18 O, reservoir temperature or reservoir pressure, , both represent the weight coefficient, represents the evaluation transformation value of the feature importance of the nth layer, is the n layer's weight parameter.

[0082] Among them, the Permutation feature importance correlation factors (i.e., the evaluation transformation values of feature importance) of different influencing factors are shown in Table 1 and Figure 6 as follows:

[0083] Table 1

[0084]

[0085] The output layer contains multiple GBT models; the output layer is used to calculate the residuals of each GBT model, use the residuals as the target value, and combine the hydrothermal intensity index TII output by the previous layer of GBT model to update the GBT model of the next layer, and then obtain the predicted value of the comprehensive hydrothermal intensity index TII. Specifically as follows:

[0086] The output layer receives the output of the second hidden layer and generates the final predicted value of TII. This layer can contain one or more GBT models, and its output is the final estimate of the comprehensive hydrothermal intensity index.

[0087] ;

[0088] Mathematical expression of the hierarchical structure: The overall prediction function F(x) of the DGBN model is represented in a nested form as:

[0089] ;

[0090] Among them, represents the predicted value of the comprehensive hydrothermal intensity index TII, is the weight of each loss term, indicating the contribution degree of different loss terms to the total loss, is the predicted value of TII of the n −1 layer, is a factor that controls the influence of the residual on the total loss, and can affect the sensitivity of the model to the residual by adjustment, represents the hydrothermal intensity index TII of the n layer, is the output value of the internal non - linear transformation of the residual block, corresponding to the predicted correction amount of the network for the features of this layer, and is the predicted residual of the nth layer. is the standardized inter - layer residual calculated based on geological prior knowledge, reflecting the offset of the actual rock physical parameters, representing the actual residual, that is, the difference between layer n and layer n−1. This allows the network to learn the error of the current layer and the information passed from the previous layer. represents an optimization operation. The output of each layer is used as the input of the next layer, which can retain the historical information of each layer and perform complex prediction tasks through layer - by - layer optimization. The optimal residual function is found through the optimization process. , minimizing the overall loss function, where M represents the total number of iterations of the GBT model in each layer.

[0091] In addition, for the DGBN model, its loss function during training is:

[0092] ;

[0093] ;

[0094] ;

[0095] Among them, represents the loss value, that is, the predicted residual. is the true TII value. In the actual application process, the true TII value is the standard value set according to the geological situation. is the TII value predicted by the model, and 𝑛 is the total number of test samples. The features output by the (n - 1)th layer. is the convolutional kernel weight matrix, extracting geological features, that is, the features of the parameters used to characterize the hydrothermal source and origin, R n2 is the convolutional kernel weight matrix, expanding the features back to the original dimension, and ReLU() is the activation function. 、 are both bias parameters. is the measured geological feature at the kth sampling point. is the standard value set according to the geological situation. N is the total number of sampling points.

[0096] In this way, the loss function not only depends on the prediction error but also takes into account the correction of the prediction error of each layer, as shown in Figure 4 .

[0097] That is, during the training process, the DGBN is first trained using the training set, and then the validation set is used to evaluate the performance of the model. If the error on the validation set is lower than the preset threshold, or the accuracy on the validation set is greater than 90%, the training is stopped and the model is used for prediction. Otherwise, the learning step size is adjusted and the training continues.

[0098] After S300, it also includes: numerically simulating the hydrothermal migration path. The hydrothermal migration channels are mainly divided into faults and fracture networks. The connectivity information of faults and fractures and TII data are extracted using seismic data and interpolated into the grid. Among them, the fault dominance is an index used to quantify the control ability of faults on the hydrothermal migration path. By analyzing the correlation between the hydrothermal activity intensity (TII) and the spatial distribution of faults, it is judged whether the fault is the main controlling factor for hydrothermal migration:

[0099] ;

[0100] Among them, represents the fault dominance. The high-value area of TII is the area with significant hydrothermal activity intensity, that is, the area where TII is greater than the set intensity (preferably TII>5). The fault area is a banded area extending along the fault strike and on both sides. represents the overlapping area between the i-th high-value area of TII and the fault area. represents the sliding rate at the overlap between the i-th high-value area of TII and the fault area. represents the maximum sliding rate in the faults in all overlapping areas. represents the average value of TII in the overlapping area between the i-th high-value area of TII and the fault area. represents the maximum TII value in the faults in all overlapping areas. represents the total area of the fault area; in the present invention, it is preferred that FDI is greater than or equal to 50%, and the fault is the main controlling factor for hydrothermal migration.

[0101] Assume that there are a total of 12 overlapping areas between the high-value areas of TII and the fault area, with a total area of 27.39 KM 2 , and the total area of the fault area is 43.88 KM 2 , is 0.28, is 0.74, and the fault dominance =58.31%, then the fault is the main controlling factor for hydrothermal migration.

[0102] Among them, the fracture network control degree is an index used to quantify the control ability of the fracture network on the hydrothermal migration path. By analyzing the correlation between the hydrothermal activity intensity (TII) and the spatial distribution of the fracture network, it is judged whether the fracture network is the main controlling factor for hydrothermal migration. Its core is to evaluate the contribution ratio of the fault area to the formation of hydrothermal channels. The connectivity of the fractures is generated by ant body tracking, curvature attributes or discrete fracture network (DFN) models to form a three-dimensional fracture network, including fracture surface geometric parameters (strike, dip, length) and topological structure. If the distance between two fracture surfaces in three-dimensional space ≤ the grid resolution (such as 1 m), it is marked as an intersection node, and the coordinates of the intersection node are mapped to the TII three-dimensional grid data to extract the TII value at the corresponding position. Through the fracture network control degree (Cctrl) calculation formula:

[0103] ;

[0104] Among them, represents the fracture network control degree (in the present invention, preferably, when the fracture network control degree is greater than 0.3, the fracture network is the main controlling factor for hydrothermal migration), K represents the total number of fracture intersection nodes, TII intersection point i represents the TII value of the i-th intersection node (that is, the TII value is calculated by the above formula for calculating the hydrothermal intensity index TII), and TII 围岩值 represents the average TII value of the non-fracture area within the model range (calculation method: excluding the grid TII mean value outside the fracture buffer zone).

[0105] Assume that TII 围岩值 is 0.0234 and K is 851, then the fracture control degree

[0106] , then the fracture network is not the main controlling factor for hydrothermal migration.

[0107] After S300, it also includes obtaining the total geothermal intensity of the entire reservoir (the total hydrothermal distribution and geothermal fluid in the entire reservoir) calculated according to the following formula:

[0108] ;

[0109] Among them, represents the hydrothermal intensity (or the influence of the fluid) of the i, j, k grid unit. It is a predicted value and can be dynamically changed. is a dynamic weighting factor that can be adjusted according to geological features (such as temperature, pressure, etc.) and is used to adjust the influence of the TII of each grid cell on the total fluid volume. Δx, Δy, and Δz are the intervals in the x, y, and z directions of the subdivision cells in the study area, respectively. m, n, and l are the total numbers of subdivision grid cells in the x, y, and z directions, respectively. k, j, and i are the subdivision grid cells in the x, y, and z directions, respectively, which helps to quickly identify potential hydrothermal activity areas, provides accurate and reliable reservoir effectiveness evaluation for engineers and decision-makers, supports the formulation of more effective exploitation strategies, reduces risks, and improves economic benefits.

[0110] After S300, it also includes: Result interpretation and visualization: Generate detailed hydrothermal intensity for each sample: Comprehensive hydrothermal intensity index (TII): The calculated TII value reflects the hydrothermal activity intensity of the sample. The relationship diagram between the comprehensive hydrothermal intensity index (TII) and depth shows the trend of the TII value changing with depth, as Figure 5 shown, clearly reflecting the intensity of hydrothermal activity in the sample. It reveals the influence of hydrothermal activity on reservoir characteristics under different depth conditions, which can help researchers understand the dynamic process of deep hydrothermal fluids. By analyzing the depth data, the high and low changes of TII can be identified, thereby evaluating the reservoir effectiveness in different depth intervals.

[0111] Grade classification: According to the TII value, the samples are divided into different hydrothermal activity grades (such as extremely strong, strong, medium, weak, and no hydrothermal activity), which helps to quickly identify potential hydrothermal activity areas. Parameter analysis: List in detail the main parameters affecting the TII value, including trace elements δEu, δ 18 O value, temperature, pressure, etc., as well as the specific values of these parameters in the current sample and their contributions to the TII.

[0112] The specific steps are as follows:

[0113] 1) Set the classification criteria, and reasonable division can be carried out by methods such as the quantile method;

[0114] 2) Analyze the reservoir characteristics corresponding to the TII value of each category, and combine the geological background and actual data to formulate the reservoir effectiveness evaluation criteria;

[0115] 3) Finally, form a set of reservoir evaluation system based on the TII value, providing a scientific basis for subsequent reservoir development, management, and optimization, as shown in Table 2 specifically:

[0116] Table 2

[0117]

[0118] A system for calculating the hydrothermal intensity index TII in a reservoir according to the second embodiment of the present invention includes:

[0119] A data acquisition module configured to collect data related to reservoir characteristics in an oil and gas field as input data; the input data includes parameters for characterizing the source and origin of hydrothermal fluids, parameters reflecting the chemical composition of fluids and their evolution process, parameters affecting the thermodynamic properties of the reservoir, and parameters related to the phase state and physical properties of reservoir fluids.

[0120] A preprocessing module configured to preprocess the input data to obtain preprocessed data.

[0121] A model prediction module configured to input the preprocessed data into a trained TII prediction model to obtain a predicted value of the comprehensive hydrothermal intensity index TII:

[0122] The TII prediction model is constructed based on a deep gradient boosting network; the TII prediction model includes a first hidden layer, a second hidden layer, and an output layer; the first hidden layer is used to perform standardization processing on the preprocessed data to obtain standardized data; the second hidden layer is used to perform weighted and non-linear combination on the standardized data to obtain the hydrothermal intensity index TII as the initial TII; the output layer includes multiple GBT models; the output layer is used to calculate the residuals of each GBT model, use the residuals as the target value, and combine the hydrothermal intensity index TII output by the previous layer of GBT model to update the GBT model of the next layer, thereby obtaining a predicted value of the comprehensive hydrothermal intensity index TII.

[0123] Among them, the input of the first GBT model is the initial TII.

[0124] Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process and related descriptions of the above-described system can refer to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0125] It should be noted that the system for calculating the hydrothermal intensity index TII in the reservoir provided in the above embodiments is only illustrated by the division of the above functional modules. In practical applications, the above functions can be allocated to different functional modules as needed, that is, the modules or steps in the embodiments of the present invention can be further decomposed or combined. For example, the modules in the above embodiments can be combined into one module, or further split into multiple sub-modules to complete all or part of the functions described above. The names of the modules and steps involved in the embodiments of the present invention are only for distinguishing each module or step, and are not regarded as an improper limitation of the present invention.

[0126] An apparatus for calculating the hydrothermal intensity index TII in a reservoir according to the third embodiment of the present invention includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the processor, and the instructions are used to be executed by the processor to implement the above-mentioned method for calculating the hydrothermal intensity index TII in a reservoir.

[0127] A computer-readable storage medium according to the fourth embodiment of the present invention stores computer instructions, and the computer instructions are used to be executed by the computer to implement the above-mentioned method for calculating the hydrothermal intensity index TII in a reservoir.

[0128] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes and related descriptions of the above-described apparatus for calculating the hydrothermal intensity index TII in a reservoir and the computer-readable storage medium can refer to the corresponding processes in the foregoing method examples, and will not be repeated here.

[0129] Those skilled in the art should be able to realize that the modules and method steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. The programs corresponding to the software modules and method steps can be placed in a random access memory (RAM), memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium well-known in the art. To clearly illustrate the interchangeability of electronic hardware and software, the components and steps of each example have been generally described according to their functions in the above description. Whether these functions are executed in the form of electronic hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0130] Terms such as "first", "second", "third", etc. are used to distinguish similar objects, rather than to describe or represent a specific order or sequence.

[0131] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. A method for calculating the hydrothermal intensity index TII in a reservoir, characterized in that: The method includes: S100, collecting data related to reservoir characteristics in the oil and gas field as input data; the input data includes parameters for characterizing the source and genesis of hydrothermal fluids, parameters reflecting the chemical composition of fluids and their evolution process, parameters affecting the thermodynamic properties of the reservoir, and parameters related to the phase state and physical properties of reservoir fluids; S200, preprocessing the input data to obtain preprocessed data; S300, inputting the preprocessed data into the trained TII prediction model to obtain the predicted value of the comprehensive hydrothermal intensity index TII: The TII prediction model is constructed based on a deep gradient boosting network; the TII prediction model includes a first hidden layer, a second hidden layer, and an output layer; the first hidden layer is used to standardize the preprocessed data to obtain standardized data; the second hidden layer is used to weight and nonlinearly combine the standardized data to obtain the hydrothermal intensity index TII as the initial TII; the output layer includes multiple GBT models; the output layer is used to calculate the residual of each GBT model, use the residual as the target value, and combine the hydrothermal intensity index TII output by the GBT model of the previous layer to update the GBT model of the next layer, thereby obtaining the predicted value of the comprehensive hydrothermal intensity index TII; The input of the first GBT model is the initial TII; the standardized data is weighted and nonlinearly combined to obtain the hydrothermal intensity index TII, and the method is: ； in, represents the hydrothermal intensity index TII, p represents the nonlinear adjustment factor, , , , represent the standardized trace element δEu, oxygen isotope ratio δ 18 O, reservoir temperature, reservoir pressure, k 1 represents the influence of δEu on TII, α and β represent δEu and δ 18 O represents the weight of TII, γ represents the weight of reservoir temperature, represents the weight of reservoir pressure, and c is a constant used to adjust the baseline value.

2. A method for calculating the hydrothermal intensity index TII in a reservoir according to claim 1, characterized in that: The preprocessing includes data cleaning, missing value and outlier elimination.

3. A method for calculating the hydrothermal intensity index TII in a reservoir according to claim 1, characterized in that: The parameters used to characterize the source and genesis of the hydrothermal fluid include trace element δEu; the parameters reflecting the chemical composition of the fluid and its evolution process include oxygen isotope ratio δ 18 O; the parameters affecting the thermodynamic properties of the reservoir include reservoir temperature; the parameters related to the phase state and physical properties of the reservoir fluid include reservoir pressure.

4. A method for calculating the hydrothermal intensity index TII in a reservoir according to claim 3, characterized in that: In the TII prediction model, a dynamic weight is assigned to each feature based on the self-attention mechanism, and its calculation method is: ； in, Represents the weight of the feature, i.e., α, β, γ, , are parameters used to characterize the source and origin of hydrothermal fluids, which are δEu, δ 18 O, reservoir temperature or reservoir pressure, , Both represent weight coefficients, Represents the feature importance evaluation transformation value of the nth layer, It is n The weight parameters of the layer.

5. A method for calculating the hydrothermal intensity index TII in a reservoir according to claim 3, characterized in that: The predicted value of the comprehensive hydrothermal intensity index TII is obtained by: ； ； in, represents the predicted value of the comprehensive hydrothermal intensity index TII, represents the weight of each loss term, Indicates n The TII prediction value of the layer, is δEu, δ 18 The input vector consists of O, reservoir temperature T, and reservoir pressure P. is the output value of the nonlinear transformation inside the residual block, corresponding to the prediction correction of the network for the features of this layer, and is the prediction residual of the nth layer. It is a standardized interlayer residual calculated based on geological prior knowledge, reflecting the offset of the actual rock formation physical property parameters and representing the actual residual. represents the factor that controls the effect of the residual on the total loss, Indicates n The hydrothermal intensity index TII of the layer, Represents an optimization operation, which finds the optimal residual function through the optimization process , so that the overall loss function is minimized, and M represents the total number of iterations of the GBT model in each layer.

6. A method for calculating the hydrothermal intensity index TII in a reservoir according to claim 5, characterized in that: The loss function of the TII prediction model during training is: ； ； ； in, Represents the loss value, that is, the prediction residual, It is the real TII value. In actual application, the real TII value is the standard value set according to geological conditions. is the TII value predicted by the model, 𝑛 is the total number of test samples, The features of the output of the n-1th layer, is the convolution kernel weight matrix, which extracts geological features, i.e., the features of parameters used to characterize the source and genesis of hydrothermal fluids, R n2 is the convolution kernel weight matrix, which expands the features back to the original dimension, and ReLU() is the activation function. , are bias parameters, is the measured geological characteristics of the kth sampling point, is a standard value set according to geological conditions. N is the total number of sampling points.

7. A method for calculating the hydrothermal intensity index TII in a reservoir according to claim 3, characterized in that: After step S300, the method further includes obtaining the main controlling factors of the hydrothermal migration: Extracting connectivity information of faults and fractures and TII data from seismic data, interpolating them into a grid, and calculating the master control degree of faults and the master control degree of fracture networks, and determining the master control factors of hydrothermal migration based on the master control degrees of faults and fracture networks; the master control factors include faults and fracture networks; The method for obtaining the fault master control degree is as follows: ； in, Indicates the fault master degree, represents the overlapping area between the ith TII high value area and the fault area. The TII high value area is the area with significant hydrothermal activity, that is, the area where the TII is greater than the set intensity. The fault area is the belt-shaped area extending along the fault direction and on both sides. represents the average slip rate at the overlap between the ith TII high value area and the fault area, represents the maximum slip rate in all overlapping faults, represents the average value of TII in the overlapping area between the ith TII high value area and the fault area, represents the maximum TII value among all overlapping fault layers, represents the total area of ​​the fault zone; The method for obtaining the master control degree of the fracture network is as follows: ； in, represents the master control degree of the crack network, K represents the total number of crack intersection nodes, TII 交汇点i Represents the TII value of the i-th intersection node, TII 围岩值 It represents the average TII value in the non-crack area within the model range.

8. A method for calculating the hydrothermal intensity index TII in a reservoir according to claim 3, characterized in that: After step S300, the total amount of geothermal intensity of the entire reservoir is calculated: ； in, T represents the total geothermal intensity of the entire reservoir, Indicates i , j , k Hydrothermal intensity of the grid cell, Δ x , Δ y , Δ z are the x, y, and z direction intervals of the subdivision cells of the study area, m , n , l are the total number of mesh elements in the x, y, and z directions, respectively.

9. A system for calculating the hydrothermal intensity index TII in a reservoir, characterized in that: The system includes: A data acquisition module is configured to collect data related to reservoir characteristics in the oil and gas field as input data; the input data includes parameters for characterizing the source and genesis of hydrothermal fluids, parameters reflecting the chemical composition of fluids and their evolution process, parameters affecting the thermodynamic properties of the reservoir, and parameters related to the phase state and physical properties of reservoir fluids; A preprocessing module, configured to preprocess the input data to obtain preprocessed data; The model prediction module is configured to input the preprocessed data into the trained TII prediction model to obtain the predicted value of the comprehensive hydrothermal intensity index TII: The TII prediction model is constructed based on a deep gradient boosting network; the TII prediction model includes a first hidden layer, a second hidden layer, and an output layer; the first hidden layer is used to standardize the preprocessed data to obtain standardized data; the second hidden layer is used to weight and nonlinearly combine the standardized data to obtain the hydrothermal intensity index TII as the initial TII; the output layer includes multiple GBT models; the output layer is used to calculate the residual of each GBT model, use the residual as the target value, and combine the hydrothermal intensity index TII output by the GBT model of the previous layer to update the GBT model of the next layer, thereby obtaining the predicted value of the comprehensive hydrothermal intensity index TII; The input of the first GBT model is the initial TII; the standardized data is weighted and nonlinearly combined to obtain the hydrothermal intensity index TII, and the method is: ； in, represents the hydrothermal intensity index TII, p represents the nonlinear adjustment factor, , , , represent the standardized trace element δEu, oxygen isotope ratio δ 18 O, reservoir temperature, reservoir pressure, k 1 represents the influence of δEu on TII, α and β represent δEu and δ 18 O represents the weight of TII, γ represents the weight of reservoir temperature, represents the weight of reservoir pressure, and c is a constant used to adjust the baseline value.

Images