Method, device, and equipment for predicting reservoir parameters based on rock physics

Abstract

The present application provides a method, device, and equipment for predicting reservoir parameters based on rock physics. It is applied to the field of reservoir engineering technology. The method includes obtaining core data and logging data of the area to be measured; performing initialization processing on the core data and logging data to obtain the initialized core data and logging data; inputting the initialized logging data into a random forest model to calculate the weights of multiple features of the logging data for reservoir parameters, obtaining the target features of the logging data; constructing a long short-term memory network model based on the initialized core data and the target features of the logging data, constructing a target prediction model according to the long short-term memory network model and rock physics knowledge, and training the target prediction model. The target prediction model predicts the parameters of the reservoir based on rock physics, thereby improving the accuracy of prediction and enhancing work efficiency.

Description

Technical Field

[0001] This application belongs to the technical field of reservoir engineering, and particularly relates to a method, device, and equipment for predicting reservoir parameters of an oil reservoir based on rock physics. Background Art

[0002] Reservoir engineering is a branch field of petroleum engineering that focuses on researching and developing underground oil and gas reservoirs to effectively extract oil and gas resources. During the exploration stage, reservoir engineers use geological exploration techniques such as seismic exploration, geological exploration, and geophysical exploration to search for potential oil and gas reservoirs. Through reserve evaluation, the scale, properties, and recoverable reserves of the oil and gas reservoirs are determined, providing a basis for subsequent development.

[0003] Reservoir parameter prediction is an important basis for fine reservoir evaluation and oil and gas development. Currently, there are mainly two types of methods for determining reservoir parameters: direct measurement methods and indirect interpretation methods. The direct measurement method obtains physical parameters by taking cores and analyzing them through rock physics knowledge; the indirect interpretation method is to predict and estimate relevant reservoir parameters based on well logging data. The main idea is to use the current well logging data to create a prediction model that links reservoir parameters with well logging curves and predict the reservoir parameters.

[0004] However, although the data obtained by the direct measurement method is accurate, due to economic and technical constraints, the limited data collected from rock samples makes it inefficient to accurately estimate reservoir parameters in the entire working area; the indirect interpretation method is prone to problems such as overcomplexity and overfitting, so there are large errors in the accuracy of the prediction results. Summary of the Invention

[0005] This application provides a method, device, and equipment for predicting reservoir parameters of an oil reservoir based on rock physics to solve the defects of poor prediction accuracy and low working efficiency in the prior art.

[0006] In a first aspect, this application provides a method for predicting reservoir parameters of an oil reservoir based on rock physics, and the method includes:

[0007] Obtain core data and well logging data of the area to be measured;

[0008] Perform initialization processing on the core data and well logging data to obtain the initialized core data and well logging data;

[0009] Input the initialized well logging data into a random forest model, calculate the weights of multiple features of the well logging data for reservoir parameters, and obtain the target features of the well logging data;

[0010] Construct a long short - term memory network model according to the initialized core data and the target features of the logging data, where the long short - term memory network model is used to determine the correlation between the core data and the target features of the logging data;

[0011] Construct a target prediction model according to the long short - term memory network model and rock physics knowledge, and train the target prediction model, where the target prediction model predicts the parameters of the reservoir based on rock physics.

[0012] Optionally, the reservoir parameters include porosity, permeability, and water saturation. Inputting the initialized logging data into a random forest model and calculating the re - weights of multiple features of the logging data for the reservoir parameters to obtain the target features of the logging data includes:

[0013] Input the initialized logging data into a random forest model, and calculate the influence weight values of the multiple features on the porosity, permeability, and water saturation respectively to obtain the porosity weight value, permeability weight value, and water saturation weight value;

[0014] Sort the porosity weight value, permeability weight value, and water saturation weight value in descending order of the weight value;

[0015] Take the features corresponding to the first preset number of porosity weight values after sorting as the target features of porosity;

[0016] Take the features corresponding to the first preset number of permeability weight values after sorting as the target features of permeability;

[0017] Take the features corresponding to the first preset number of water saturation weight values after sorting as the target features of water saturation.

[0018] Optionally, before sorting the porosity weight value, permeability weight value, and water saturation weight value in descending order of the weight value, the method further includes:

[0019] Judge whether there are first associated features with a correlation coefficient greater than a threshold among the multiple features;

[0020] If so, compare the weight values of at least two features among the first associated features. The first associated feature with a smaller weight value among the first associated features does not participate in the weight value sorting process, and the first associated feature with a larger weight value among the first associated features participates in the weight value sorting process;

[0021] If not, continue to sort the multiple weight values in descending order of the weight value.

[0022] Optionally, determining whether there is a first associated feature among the multiple features whose correlation coefficient is greater than a threshold includes:

[0023] Calculate the correlation coefficient between the multiple features using the following formula:

[0024]

[0025] where γ is the correlation coefficient, is the mean of the logging curves, x i , y i represent the logging curve values corresponding to the i-th sample.

[0026] Calculate the absolute value of the correlation coefficient γ and determine whether |γ| is greater than the threshold.

[0027] Optionally, constructing a long short-term memory network model based on the initialized core data and the target features of the logging data, where the long short-term memory network model is used to determine the association relationship between the core data and the target features of the logging data, includes:

[0028] Construct a long short-term memory network model based on the initialized core data and the target features of the logging data;

[0029] Perform training processing on the long short-term memory network model according to the initialized core data and the target features of the logging data to obtain a trained long short-term memory network model;

[0030] Use the initialized core data as the input data of the trained long short-term memory network model, and the target features of the logging data as the output data of the trained long short-term memory network model, and control the long short-term memory network model to perform correlation analysis processing on the core data and the target features of the logging data to obtain the association relationship between the core data and the target features of the logging data.

[0031] Optionally, the initialized core data includes multiple features. Constructing a long short-term memory network model according to the initialized core data and the target features of the logging data includes:

[0032] Construct an input layer of the long short-term memory network model with dimensions corresponding to the number of features according to the number of features of the initialized core data;

[0033] Construct an output layer of the long short-term memory network model with dimensions corresponding to the number of features according to the number of features of the target features of the logging data;

[0034] The number of units and the number of layers of the long short-term memory network model are constructed according to the target features of the initialized core data and the logging data.

[0035] Optionally, the training of the target prediction model includes:

[0036] Using the correlation between the core data and the target features of the logging data and petrophysical knowledge as input data of the target prediction model, and controlling the target prediction model to perform iterative processing;

[0037] Obtaining a total loss function of the target prediction model, wherein the total loss function includes a first loss function of a long short-term memory network model and a second loss function of petrophysical knowledge;

[0038] When it is determined that the total loss function converges, the target prediction model is controlled to stop iterative processing, and it is determined that the training of the target prediction model is completed.

[0039] In a second aspect, the present application provides a device for predicting oil reservoir parameters based on petrophysics, the device comprising:

[0040] An acquisition module is used to acquire core data and logging data of the area to be tested;

[0041] A processing module, used for initializing the core data and the well logging data to obtain initialized core data and well logging data;

[0042] A calculation module, used for inputting the initialized logging data into a random forest model, calculating the weights of multiple features of the logging data to reservoir parameters, and obtaining target features of the logging data;

[0043] The processing module is further used to construct a long short-term memory network model according to the initialized core data and the target features of the logging data, and the long short-term memory network model is used to determine the correlation relationship between the core data and the target features of the logging data;

[0044] The processing module is also used to construct a target prediction model based on the long short-term memory network model and rock physics knowledge, and train the target prediction model, wherein the target prediction model predicts parameters of oil reservoirs based on rock physics.

[0045] Optionally, the petrophysics-based reservoir parameter prediction device further includes: a determination module;

[0046] The calculation module is further configured to input the initialized well logging data into a random forest model, and calculate the influence weight values of the multiple features on the porosity, permeability, and water saturation respectively, so as to obtain a porosity weight value, a permeability weight value, and a water saturation weight value;

[0047] The processing module is further configured to perform a sorting process on the porosity weight value, the permeability weight value, and the water saturation weight value in descending order of the weight value;

[0048] The determination module is configured to use the features corresponding to the top preset number of porosity weight values after the sorting process as the target features of the porosity;

[0049] The determination module is further configured to use the features corresponding to the top preset number of permeability weight values after the sorting process as the target features of the permeability;

[0050] The determination module is further configured to use the features corresponding to the top preset number of water saturation weight values after the sorting process as the target features of the water saturation.

[0051] Optionally, the reservoir reservoir parameter prediction device based on rock physics further includes: a judgment module;

[0052] The judgment module is configured to judge whether there are first associated features with a correlation coefficient greater than a threshold among the multiple features;

[0053] When there are first associated features with a correlation coefficient greater than a threshold among the multiple features, the processing module is further configured to compare the weight values of at least two features among the first associated features, and the first associated feature with a smaller weight value among the first associated features does not participate in the weight value sorting process, and the first associated feature with a larger weight value among the first associated features participates in the weight value sorting process;

[0054] When there are no first associated features with a correlation coefficient greater than a threshold among the multiple features, the processing module is further configured to continue the sorting process of the multiple weight values in descending order of the weight value.

[0055] Optionally, the calculation module is further configured to calculate the correlation coefficient between the multiple features by using the following formula:

[0056]

[0057] where γ is the correlation coefficient, is the mean value of the well logging curve, and x i , y i represent the well logging curve values corresponding to the i-th sample;

[0058] The calculation module is further configured to calculate the absolute value of the correlation coefficient γ;

[0059] The determination module is further configured to determine whether |γ| is greater than a threshold value.

[0060] Optionally, the processing module is further configured to construct a long short-term memory network model according to the target features of the initialized core data and the logging data;

[0061] The processing module is further configured to perform training processing on the long short-term memory network model according to the target features of the initialized core data and the logging data, so as to obtain a trained long short-term memory network model;

[0062] The processing module is further configured to use the initialized core data as the input data of the trained long short-term memory network model, and the target features of the logging data as the output data of the trained long short-term memory network model, and control the long short-term memory network model to perform correlation analysis processing on the target features of the core data and the logging data, so as to obtain the association relationship between the target features of the core data and the logging data.

[0063] Optionally, the processing module is further configured to construct an input layer of the long short-term memory network model with a dimension corresponding to the number of features according to the number of features of the initialized core data;

[0064] The processing module is further configured to construct an output layer of the long short-term memory network model with a dimension corresponding to the number of features according to the number of features of the target features of the logging data;

[0065] The processing module is further configured to construct the number of units and layers of the long short-term memory network model according to the initialized core data and the target features of the logging data.

[0066] Optionally, the processing module is further configured to use the association relationship between the core data and the target features of the logging data and rock physics knowledge as the input data of the target prediction model, and control the target prediction model to perform iterative processing;

[0067] The acquisition module is further configured to acquire the total loss function of the target prediction model, where the total loss function includes a first loss function of the long short-term memory network model and a second loss function of rock physics knowledge;

[0068] The processing module is further configured to, when determining that the total loss function converges, control the target prediction model to stop iterative processing and determine that the target prediction model is trained.

[0069] In a third aspect, the present application provides an oil reservoir reservoir parameter prediction device based on rock physics, including:

[0070] Memory;

[0071] Processor;

[0072] Wherein, the memory stores computer-executable instructions;

[0073] The processor executes the computer-executable instructions stored in the memory to implement the method for predicting reservoir parameters based on rock physics as described in the first aspect and various possible implementation manners of the first aspect above.

[0074] In a fourth aspect, the present application provides a computer storage medium, on which a computer program is stored, and the computer program is executed by a processor to implement the method for predicting reservoir parameters based on rock physics as described in the first aspect and various possible implementation manners of the first aspect above.

[0075] The present application provides a method, device, and equipment for predicting reservoir parameters based on rock physics. The method includes obtaining core data and logging data of a region to be measured; performing initialization processing on the core data and logging data to obtain the initialized core data and logging data; inputting the initialized logging data into a random forest model, calculating the weights of multiple features of the logging data on reservoir parameters, and obtaining the target features of the logging data; constructing a long short-term memory network model according to the initialized core data and the target features of the logging data, constructing a target prediction model according to the long short-term memory network model and rock physics knowledge, and training the target prediction model. The target prediction model predicts the parameters of the reservoir based on rock physics, thereby improving the prediction accuracy and work efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] The accompanying drawings herein are incorporated into the specification and form a part of the specification, showing embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application.

[0077] Figure 1 Schematic flow chart of the method for predicting reservoir parameters based on rock physics provided by the present application Figure 1 ;

[0078] Figure 2 Schematic flow chart of the method for predicting reservoir parameters based on rock physics provided by the present application Figure 2 ;

[0079] Figure 3 Schematic diagram of the correlation coefficients of different degrees between each logging feature provided by the present application;

[0080] Figure 4 Schematic diagram of the sorting of the influence weight values of multiple features of logging data on porosity provided by the present application;

[0081] Figure 5 Schematic diagram of the sorting of the influence weight values of multiple characteristics of well logging data provided by this application on permeability;

[0082] Figure 6 Schematic diagram of the sorting of the influence weight values of multiple characteristics of well logging data provided by this application on water saturation;

[0083] Figure 7 Reservoir parameter prediction model based on rock physics provided by this application;

[0084] Figure 8 Schematic diagram of the structure of the reservoir parameter prediction device based on rock physics provided by this application;

[0085] Figure 9 Schematic diagram of the structure of the reservoir parameter prediction equipment based on rock physics provided by this application.

[0086] Through the above-mentioned drawings, specific embodiments of this application have been shown, and there will be more detailed descriptions hereinafter. These drawings and textual descriptions are not intended to limit the scope of the concept of this application in any way, but to illustrate the concept of this application to those skilled in the art by referring to specific embodiments. Detailed implementation manners

[0087] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions in this application will be clearly and completely described below in conjunction with the drawings in this application. Obviously, the described embodiments are some, but not all, of the embodiments of this application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in this application without creative efforts shall fall within the protection scope of this application.

[0088] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of this invention and the above-mentioned drawings are used to distinguish similar objects and do not necessarily need to describe a specific order or sequence. It should be understood that such used data can be interchanged under appropriate circumstances so that the embodiments of this invention described here can be implemented in an order different from those illustrated or described here.

[0089] In the embodiments of this application, words such as "exemplary" or "for example" are used to represent examples, illustrations, or explanations. Any embodiment or design solution described as "exemplary" or "for example" in this application should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Rather, the use of words such as "exemplary" or "for example" is intended to present relevant concepts in a specific manner.

[0090] First, the nouns involved in this application are explained.

[0091] Long Short-Term Memory Network Model: (LSTM, Long Short-Term Memory) is a special type of Recurrent Neural Network (RNN) that can learn long-term dependency information. LSTM addresses the problem of vanishing or exploding gradients encountered by traditional RNNs when dealing with long sequence data by introducing three gating mechanisms. These three gates are: the input gate, the forget gate, and the output gate. The input gate determines which information should be added to the cell state; the forget gate determines which information should be forgotten, i.e., removed from the cell state; the output gate determines which information the final output should contain. This structure enables LSTM to perform well in tasks such as time series prediction, language models, and machine translation, capturing long-term dependencies in the data.

[0092] Random Forest Model: Random forest is an ensemble learning method consisting of multiple decision trees, used for classification and regression problems. It improves the accuracy and robustness of the model by constructing multiple decision trees and aggregating their prediction results. Each decision tree uses a randomly selected subset of samples (bootstrap sampling) during training, and when splitting nodes, only a random subset of features is considered. This randomness increases the diversity of the model and reduces the risk of overfitting. The advantages of random forest include easy implementation, fast training speed, and good tolerance to data noise and outliers. In addition, it can provide an assessment of feature importance to help understand the model's decision-making process. However, the random forest model may consume a large amount of memory, and the interpretability of the model is not as good as that of a single decision tree.

[0093] Reservoir engineering is a branch field of petroleum engineering that focuses on the study and development of underground oil and gas reservoirs to efficiently extract oil and gas resources. During the exploration phase, reservoir engineers use various geological exploration techniques, such as seismic exploration, geological exploration, and geophysical exploration, to identify and evaluate potential oil and gas reservoirs. These techniques help engineers understand the physical properties, structures, and fluid distributions of underground rocks. Once an oil and gas reservoir is identified, reservoir engineers conduct a reserve assessment, which includes estimating the volume of the reservoir, the quality of the oil and gas, and the recoverable reserves.

[0094] Reservoir parameter prediction is crucial for the fine evaluation of oil reservoirs and the effective development of oil and gas. It involves the accurate assessment of the physical properties of the reservoir and the fluid properties. The direct measurement method relies on core samples obtained from the reservoir. Through laboratory tests such as core analysis, porosity and permeability tests, the physical properties of the rock can be measured directly and accurately. However, the direct measurement method requires drilling to obtain core samples from the underground reservoir, resulting in high costs; due to the limited number of core samples, it is impossible to comprehensively represent the geological characteristics of the entire reservoir, and the heterogeneity of the reservoir may lead to deviations between the sample data and the actual reservoir conditions; the collection, transportation and analysis of core samples take time, which greatly affects the timeliness of development decisions.

[0095] The indirect interpretation method mainly relies on logging data. Logging is a non-invasive technology that can evaluate the underground reservoir through the wellbore without taking out the core. By using different types of logging tools to collect the response data of underground rocks and fluids, a mathematical model between the logging response and reservoir parameters is established and the established model is applied to predict the physical and fluid properties of the reservoir. The advantage of the indirect interpretation method is that it can provide a continuous distribution map of reservoir parameters without relying on limited core samples. However, the accuracy of this method is limited by the quality of logging data and the reliability of the interpretation model.

[0096] In view of the above problems, the present application proposes a method, device, and equipment for predicting reservoir parameters of an oil reservoir based on rock physics. The method includes obtaining core data and logging data of the area to be measured; initializing the core data and logging data to obtain the initialized core data and logging data; inputting the initialized logging data into a random forest model, calculating the reweights of multiple features of the logging data on the reservoir parameters of the oil reservoir to obtain the target features of the logging data; constructing a long short-term memory network model according to the initialized core data and the target features of the logging data, constructing a target prediction model according to the long short-term memory network model and rock physics knowledge, and training the target prediction model. The target prediction model predicts the parameters of the oil reservoir based on rock physics, thereby improving the prediction accuracy and enhancing the work efficiency.

[0097] The following uses specific embodiments to elaborate in detail on the technical solution of the present application and how the technical solution of the present application solves the above technical problems. These several specific embodiments below can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below with reference to the accompanying drawings.

[0098] Figure 1 Schematic flow of the method for predicting reservoir parameters of an oil reservoir based on rock physics provided for the embodiments of the application Figure 1 . As Figure 1As shown in the figure, the reservoir reservoir parameter prediction method based on rock physics provided by this embodiment includes:

[0099] S101. Obtain core data and logging data of the area to be measured.

[0100] A core is a cylindrical sample of underground rock taken from an oil and gas well. Through a special coring barrel during the drilling process, an actual sample of the underground rock formation can be obtained. To obtain the core data of the area to be measured, specifically obtain the porosity, permeability, and water saturation of the core in the area to be measured.

[0101] Porosity: It represents the percentage of the volume of pore space in the rock to the total volume and is an important parameter for evaluating the reservoir storage capacity. Permeability: It describes the ability of fluid to flow in the pores of the rock, usually measured in Darcy, and is a key parameter determining the flow of oil and gas and the productivity of the reservoir. Water saturation: It is the volume percentage of the water phase in the reservoir and is related to porosity and fluid properties.

[0102] Logging is a method of measuring the physical properties of the rock around the wellbore by lowering special instruments into the well. Logging technology does not require taking rock samples, so it can provide continuous underground information. Specifically obtain the natural gamma, litho-density, acoustic transit time, compensated neutron, deep lateral resistivity, shallow lateral resistivity, spontaneous potential, and caliper logging data of the area to be measured.

[0103] Natural Gamma: The content of radioactive elements in the rock, usually used to identify rock types such as shale and sandstone. Litho-density: Estimate the density of the rock by measuring the absorption of gamma rays by the rock, which helps to identify rock types and pore fluids. Acoustic transit time: The propagation time of sound waves in the rock, which can be used to estimate the porosity and fluid type of the rock. Compensated neutron: Estimate the porosity by measuring the scattering of neutrons by the rock, which has different responses to the porosity of water and oil. Deep lateral resistivity: Measure the resistivity of the formation far from the wellbore, which helps to identify the oil-water interface and determine the reservoir boundary. Shallow lateral resistivity: Measure the resistivity of the formation close to the wellbore, which can provide information on the lateral variation of the formation resistivity. Spontaneous Potential (SP): The potential difference formed due to the adsorption of fluid on the rock surface, which can be used to identify oil and water layers. Caliper Log: Measure the diameter of the wellbore, which can detect irregularities in the wellbore wall such as hole shrinkage and collapse.

[0104] Obtaining core data and logging data of the area to be measured is a very important step in reservoir engineering. These data are crucial for evaluating the properties and characteristics of underground reservoirs. The comprehensive utilization of core data and logging data can help reservoir engineers better understand the characteristics of underground reservoirs, guide oil and gas exploration and development work, and improve the efficiency and success rate of resource extraction.

[0105] S102. Initialize the core data and logging data to obtain the initialized core data and logging data.

[0106] Among them, for the initialization process of core data, specifically, clean the core data to remove outliers, missing values, and noise data to ensure the accuracy and integrity of the data; standardize the core data so that data with different attributes have the same scale and range, facilitating subsequent data analysis and comparison; extract useful features from the core data, such as porosity, permeability, etc., for subsequent geological modeling and reservoir evaluation.

[0107] For the initialization process of logging data, specifically, clean the logging data to remove outliers, missing values, and noise data to ensure the accuracy and integrity of the data; correct the logging data, considering the response characteristics of logging tools and environmental factors to obtain more accurate formation parameters; interpret the logging curves, compare and match the logging curves with geological characteristics, and extract formation parameters and characteristics.

[0108] Align and match the core data and logging data to ensure the consistency and comparability between the two. By initializing the core data and logging data, data that has been cleaned, standardized, and interpreted can be obtained, providing preparations for subsequent geological modeling, reservoir evaluation, and resource development.

[0109] S103. Input the initialized logging data into the random forest model, calculate the weights of multiple features of the logging data on reservoir parameters, and obtain the target features of the logging data.

[0110] Among them, input the initialized logging data into the random forest model, set the parameters of the random forest model, such as the number of trees, the size of the feature subset considered during splitting, etc., to initialize the model; use the initialized logging data to train the random forest model. Specifically, each decision tree randomly samples from the entire dataset during training and randomly selects features for splitting at each splitting node. The random forest model can evaluate the importance of each feature for the prediction result. Through this evaluation, the logging data features that have the greatest impact on reservoir parameters can be identified, and the logging data features that have the greatest impact on reservoir parameters are used as the target features of the logging data.

[0111] S104. Construct a long short-term memory network model according to the target features of the initialized core data and logging data.

[0112] Among them, according to the target features of the initialized core data and logging data, construct a long short-term memory network model. The long short-term memory network model is used to determine the correlation between the core data and the target features of the logging data.

[0113] Specifically, according to the target features of the initialized core data and logging data, define the architecture of the long short-term memory network, including the number of layers, the number of units in each layer, etc.; select an appropriate activation function, such as tanh or ReLU; determine the loss function and optimizer for the training process. Input the organized core data and logging data into the long short-term memory model for model training. During the training process, the model will learn the correlation between the features of the core data and the logging data. The trained long short-term memory network model can show the correlation between the core data and the target features of the logging data.

[0114] S105. Construct a target prediction model according to the long short-term memory network model and rock physics knowledge, and train the target prediction model. The target prediction model predicts the parameters of the reservoir based on rock physics.

[0115] Among them, before constructing the target prediction model, first integrate rock physics knowledge, including but not limited to understanding the relationship between different rock properties and reservoir parameters; determine which rock properties have an important impact on the prediction of reservoir parameters and incorporate this knowledge into the model design. Combine the long short-term memory network model and rock physics knowledge to construct a comprehensive target prediction model for predicting reservoir parameters.

[0116] Input the data into the comprehensive model for training the target prediction model. During the training process, the model will learn the complex relationship between rock properties and reservoir parameters and optimize the parameters. Optimize and adjust the model according to the evaluation results to improve the accuracy and generalization ability of the model. Obtain the trained target prediction model, and the target prediction model predicts the parameters of the reservoir based on rock physics.

[0117] This embodiment proposes a method for predicting reservoir parameters based on petrophysics. The method includes obtaining core data and logging data of the area to be measured; performing initialization processing on the core data and logging data to obtain the initialized core data and logging data; inputting the initialized logging data into a random forest model, calculating the reweights of multiple features of the logging data on the reservoir parameters, and obtaining the target features of the logging data; constructing a long short-term memory network model based on the initialized core data and the target features of the logging data, constructing a target prediction model according to the long short-term memory network model and petrophysics knowledge, and training the target prediction model. The target prediction model predicts the parameters of the reservoir based on petrophysics, thereby improving the prediction accuracy and work efficiency.

[0118] Figure 2 It is a schematic flow of the method for predicting reservoir parameters based on petrophysics provided by the embodiments of the present application. Figure 2 This embodiment is based on Figure 1 the embodiment and elaborates in detail on the method for predicting reservoir parameters based on petrophysics. As Figure 2 shown, the method for predicting reservoir parameters based on petrophysics provided by this embodiment includes:

[0119] S201. Obtain core data and logging data of the area to be measured.

[0120] Step S201 is the same as the above step S101 and will not be elaborated here.

[0121] S202. Perform initialization processing on the core data and logging data to obtain the initialized core data and logging data.

[0122] Step S202 is the same as the above step S102 and will not be elaborated here.

[0123] The reservoir parameters include porosity, permeability, and water saturation. Input the initialized logging data into a random forest model, and calculate the influence weight values of multiple features on porosity, permeability, and water saturation respectively to obtain the porosity weight value, permeability weight value, and water saturation weight value.

[0124] Among them, a random forest model is constructed for each reservoir parameter: porosity, permeability, and water saturation. Input the logging data into the random forest model for training, and adjust the parameters of the random forest, such as the number of trees, the randomness of feature selection, etc., to obtain the best performance. The random forest model can provide the influence weight values of each feature of the logging data on the reservoir parameters porosity, permeability, and water saturation, and obtain the porosity weight value, permeability weight value, and water saturation weight value.

[0125] The random forest algorithm provides an effective mechanism to evaluate the importance of features. It measures the importance of features by calculating the average contribution of each feature across all decision trees in the random forest. Random forest is an ensemble learning method that constructs multiple decision trees and uses bootstrap sampling during the construction of each decision tree to generate individual training subsets. Due to the use of sampling with replacement, some data may be omitted in a single sampling, and the data sets not selected for constructing a specific decision tree are called out-of-bag (OOB) data. Random forest uses this OOB data to evaluate the performance of the model, thus providing a built-in evaluation mechanism without cross-validation.

[0126] Specifically, the steps to evaluate feature importance in a random forest are usually as follows:

[0127] For each tree, calculate the importance of the feature by comparing the performance difference of the model with and without the feature when making predictions using OOB data; for each feature, this performance difference is accumulated and averaged across all trees to obtain the average importance score of the feature; by sorting the average importance scores of all features, the most important features and their contribution to the model's predictive ability can be determined. This method of evaluating feature importance based on random forest can not only provide guidance for feature selection, but also, due to its inclusion of the randomness of the model and OOB estimation, is generally considered a feature importance estimation method with relatively low bias.

[0128] In this study, random forest was used for feature selection, and the total importance of feature variables was normalized to 1. The results were arranged in descending order of the importance of feature variables.

[0129] S204. Determine whether there is a first associated feature among multiple features whose correlation coefficient is greater than the threshold. If so, execute step S205; if not, execute step S206.

[0130] Among them, use a formula to calculate whether there is a first associated feature among multiple features of logging data whose correlation coefficient is greater than the threshold. If there is a first associated feature whose correlation coefficient is greater than the threshold, execute step S205; if there is no first associated feature whose correlation coefficient is greater than the threshold, execute step S206.

[0131] Specifically, use the following formula to calculate the correlation coefficient between the multiple features:

[0132]

[0133] Among them, γ is the correlation coefficient, is the mean value of the logging curve, x i 、yi represents the logging curve value corresponding to the i-th sample.

[0134] Calculate the absolute value of the correlation coefficient γ, and determine whether |γ| is greater than the threshold.

[0135] The value of the correlation coefficient ranges between [-1, 1], and different values indicate different degrees of correlation.

[0136] Through correlation analysis, the correlation coefficients of different degrees between each logging feature are obtained. The correlation coefficients are visually displayed on a heat map, as Figure 3 shown.

[0137] Among them, according to Figure 3 the visual representation of the correlation coefficient on the heat map, determine whether there is a first associated feature with a correlation coefficient greater than the threshold among multiple features.

[0138] Specifically, Figure 3 the data presented in shows that the correlation coefficient between the deep and shallow resistivity curves is 0.934, indicating a strong correlation between them. This shows that there is a significant covariance between the two curves, making them not suitable for being used as sensitive curves simultaneously. The correlation coefficients between the spontaneous potential and the deep and shallow resistivities are 0.336 and 0.311 respectively; the correlation coefficients between the natural gamma and the deep and shallow lateral resistivities are 0.405 and 0.35 respectively; the correlation coefficient between the density logging and the deep lateral resistivity reaches 0.309, and the correlation coefficient with the acoustic travel time reaches 0.387.

[0139] Therefore, determine the deep and shallow resistivity curves as the first associated features with a correlation coefficient greater than the threshold.

[0140] S205. Compare the weight values of at least two features in the first associated features. The first associated feature with a smaller weight value in the first associated features does not participate in the weight value sorting process, and the first associated feature with a larger weight value in the first associated features participates in the weight value sorting process.

[0141] Among them, when there is a first associated feature with a correlation coefficient greater than the threshold, compare the weight values of at least two features in the first associated features. The first associated feature with a smaller weight value in the first associated features does not participate in the weight value sorting process, and the first associated feature with a larger weight value in the first associated features participates in the weight value sorting process.

[0142] Specifically, from Figure 3 it is determined that the deep and shallow resistivity curves are the first associated features with a correlation coefficient greater than the threshold. And referring to Figure 4 、 Figure 5 、 Figure 6It can be seen that the contribution value of deep resistivity to porosity, permeability, and water saturation is lower than that of shallow resistivity. Because in the subsequent sorting process of weight values, deep resistivity does not participate in the sorting, while shallow resistivity participates in the sorting.

[0143] S206. Perform a sorting process on the porosity weight value, permeability weight value, and water saturation weight value in descending order of weight value.

[0144] The sorting of the influence weight values of multiple characteristics of well logging data on porosity is shown in the appendix Figure 4 。

[0145] The sorting of the influence weight values of multiple characteristics of well logging data on permeability is shown in the appendix Figure 5 。

[0146] The sorting of the influence weight values of multiple characteristics of well logging data on water saturation is shown in the appendix Figure 6 。

[0147] S207. Take the characteristics corresponding to the top preset number of porosity weight values after the sorting process as the target characteristics of porosity.

[0148] Among them, take the characteristics corresponding to the top preset number of porosity weight values after the sorting process as the target characteristics of porosity. Specifically, according to the appendix Figure 4 , select these 6 characteristics, namely "litho-density", "acoustic transit time", "natural gamma", "spontaneous potential", "shallow lateral resistivity", and "depth", as the target characteristics of porosity.

[0149] S208. Take the characteristics corresponding to the top preset number of permeability weight values after the sorting process as the target characteristics of permeability.

[0150] Among them, take the characteristics corresponding to the top preset number of permeability weight values after the sorting process as the target characteristics of permeability. Specifically, according to the appendix Figure 5 , select these 6 characteristics, namely "porosity", "spontaneous potential", "depth", "acoustic transit time", "density", and "natural gamma", as the target characteristics of permeability.

[0151] S209. Take the characteristics corresponding to the top preset number of water saturation weight values after the sorting process as the target characteristics of water saturation.

[0152] Among them, take the characteristics corresponding to the top preset number of water saturation weight values after the sorting process as the target characteristics of water saturation. Specifically, according to the appendix Figure 6 , select these 7 characteristics, namely "permeability", "spontaneous potential", "shallow resistivity", "porosity", "acoustic transit time", "natural gamma", and "density", as the target characteristics of water saturation.

[0153] S210. Construct an input layer of a long short-term memory network model with dimensions corresponding to the number of features based on the characteristics of the initialized core data.

[0154] Among them, count the number of features in the initialized core data, which will determine the dimension of the input layer of the long short-term memory network model. Use deep learning libraries in Python (such as Keras or PyTorch) to construct the long short-term memory network model. In the input layer, set parameters with dimensions equal to the number of features to adapt to core data with different numbers of features.

[0155] S211. Construct an output layer of a long short-term memory network model with dimensions corresponding to the number of features of the target features in the logging data.

[0156] Among them, determine the number of target features in the logging data, which will determine the dimension of the output layer of the long short-term memory network model. Use deep learning libraries in Python (such as Keras or PyTorch) to construct the long short-term memory network model. In the output layer, set parameters with dimensions equal to the number of target features to adapt to logging data with different numbers of target features.

[0157] S212. Construct the number of units and layers of the long short-term memory network model based on the initialized core data and the target features of the logging data.

[0158] Among them, construct the number of units and layers of the long short-term memory network model based on the initialized core data and the target features of the logging data. The number of units refers to the number of LSTM units in the LSTM layer. More units can increase the representational ability of the model, but may also lead to overfitting. Usually, you can start with a smaller number of units and then gradually increase it to improve the model performance. The number of layers refers to the number of LSTM layers in the LSTM network. Increasing the number of layers can increase the complexity and representational ability of the model, but may also increase the training time and the risk of overfitting. For simpler tasks, usually start with a single layer. For more complex tasks, consider increasing the number of layers. Based on the above principles, the appropriate number of units and layers can be selected according to the initialized core data and the target features of the logging data to construct a long short-term memory network model with the specified number of units and layers.

[0159] S213. Perform training processing on the long short-term memory network model based on the initialized core data and the target features of the logging data to obtain a trained long short-term memory network model.

[0160] Among them, the initialized core data and the target features of well logging data are used for model training, and parameters such as the batch size, number of iterations, and validation set of the training are specified. The performance of the trained long short-term memory network model is evaluated using the test set data. After multiple iterative trainings, the trained long short-term memory network model is obtained, as Figure 7 shown.

[0161] S214. Take the initialized core data as the input data of the trained long short-term memory network model, and the target features of well logging data as the output data of the trained long short-term memory network model. Control the long short-term memory network model to perform correlation analysis processing on the target features of core data and well logging data, and obtain the correlation relationship between the target features of core data and well logging data.

[0162] Among them, take the initialized core data as the input data of the trained long short-term memory network model, and the target features of well logging data as the output data of the trained long short-term memory network model. Control the long short-term memory network model to perform correlation analysis processing on the target features of core data and well logging data, and compare and analyze the prediction results with the target features of actual well logging data to evaluate the correlation relationship between the target features of core data and well logging data.

[0163] S215. Construct a target prediction model based on the long short-term memory network model and rock physics knowledge. Take the correlation relationship between the target features of core data and well logging data and rock physics knowledge as the input data of the target prediction model, and control the target prediction model to perform iterative processing.

[0164] Among them, a target prediction model is constructed based on the long short-term memory network model and rock physics knowledge. The target prediction model combines a pure data-driven long short-term memory neural network and a long short-term memory neural network with rock physics knowledge constraints. The reservoir parameter prediction model based on rock physics is as Figure 7 shown.

[0165] Specifically, the two networks are consistent in the optimizer, batch size, and number of training epochs, and the Adam optimizer, batch size of 64, and 200 training epochs are adopted respectively. The pure data-driven long short-term memory network model uses the root mean square error (RMSE) as the loss function, while the physically constrained long short-term memory network model uses a comprehensive loss function that combines rock physics knowledge constraints and data constraints. In order to analyze the prediction effects of the pure data-driven neural network and the physically constrained neural network, the same network hyperparameters are used for the same prediction target. For different prediction targets, systematic hyperparameter tuning is carried out.

[0166] S216. Obtain the total loss function of the target prediction model.

[0167] Among them, the total loss function includes a first loss function of the long short-term memory network model and a second loss function of rock physics knowledge.

[0168] The total loss function includes: a first loss function RMSE of the prediction error of rock physical property parameters driven by data data and a second loss function RMSE of the prediction error of the neural network for domain knowledge p .

[0169] Specifically, the total loss function is defined as:

[0170] LOSS = α1RMSE data + α2RMSE p

[0171] Where:

[0172]

[0173] LOSE p = |PK(x i , y i ) - MK(x i , y i )| - ξ

[0174]

[0175] In the formula, RMSE data represents the prediction error generated by predicting reservoir parameters using data-driven rock parameters, and RMSE p represents the prediction error of the neural network for reservoir parameters based on rock physics domain knowledge. represents the true value, PK(x i , y i ) represents the reservoir physical property parameters calculated by the neural network based on rock physics domain knowledge, and MK(x i , y i ) represents the reservoir physical property parameters calculated by using data-driven rock parameters. ξ is the allowable range of model error, and α1, α2 are weight coefficients.

[0176] S217. When it is determined that the total loss function converges, control the target prediction model to stop iterative processing and determine that the training of the target prediction model is completed.

[0177] Among them, before training the model, a minimum loss threshold can be set. When the loss function is lower than this threshold, stop training the target prediction model and determine that the training of the target prediction model is completed. Setting the loss function threshold can ensure that the model stops training after reaching a certain performance level and avoid overfitting.

[0178] This embodiment proposes a method for predicting reservoir parameters based on rock physics. The method inputs the initialized well logging data into a random forest model, calculates the influence weight values of multiple features in the well logging data on porosity, permeability, and water saturation respectively, and sorts them in descending order of the weight values; takes the features corresponding to the top preset number of porosity weight values, permeability weight values, and water saturation weight values after sorting as the porosity target feature, permeability target feature, and water saturation target feature respectively; constructs a long short-term memory network model based on the initialized core data and well logging data; uses the initialized core data as the input data of the trained long short-term memory network model, and the well logging data target feature as the output data of the trained long short-term memory network model, controls the long short-term memory network model to perform correlation analysis processing on the core data and the well logging data target feature, and obtains the correlation relationship between the core data and the well logging data target feature; constructs a target prediction model according to the long short-term memory network model and rock physics knowledge, takes the correlation relationship between the core data and the well logging data target feature and rock physics knowledge as the input data of the target prediction model, and controls the target prediction model to perform iterative processing; obtains the total loss function of the target prediction Moses, and when the total loss function converges, controls the target prediction model to stop iterative processing and determines that the target prediction model is trained. This can avoid overfitting, improve the accuracy of prediction, and enhance work efficiency.

[0179] Figure 8 FIG. is a schematic structural diagram of a reservoir parameter prediction device based on rock physics provided by the present application, as Figure 8 shown, the reservoir parameter prediction device 300 based on rock physics provided in this embodiment includes:

[0180] An acquisition module 301, configured to acquire core data and well logging data of a to-be-measured area;

[0181] A processing module 302, configured to perform initialization processing on the core data and the well logging data to obtain the initialized core data and well logging data;

[0182] A calculation module 303, configured to input the initialized well logging data into a random forest model, calculate the weights of multiple features of the well logging data on reservoir parameters, and obtain the target features of the well logging data;

[0183] The processing module 302 is further configured to construct a long short-term memory network model according to the initialized core data and the target features of the well logging data, and the long short-term memory network model is used to determine the correlation relationship between the core data and the target features of the well logging data;

[0184] The processing module 302 is further configured to construct a target prediction model according to the long short-term memory network model and petrophysical knowledge, and train the target prediction model, where the target prediction model predicts the parameters of the reservoir based on petrophysics.

[0185] Optionally, the apparatus for predicting reservoir parameters based on petrophysics further includes: a determination module 304;

[0186] The calculation module 303 is further configured to input the initialized well logging data into a random forest model, and calculate the influence weight values of the multiple features on the porosity, permeability, and water saturation respectively, to obtain a porosity weight value, a permeability weight value, and a water saturation weight value;

[0187] The processing module 302 is further configured to perform a sorting process on the porosity weight value, the permeability weight value, and the water saturation weight value in descending order of the weight value;

[0188] The determination module 304 is configured to use the features corresponding to the first preset number of porosity weight values after the sorting process as the target features of the porosity;

[0189] The determination module 304 is further configured to use the features corresponding to the first preset number of permeability weight values after the sorting process as the target features of the permeability;

[0190] The determination module 304 is further configured to use the features corresponding to the first preset number of water saturation weight values after the sorting process as the target features of the water saturation.

[0191] Optionally, the apparatus for predicting reservoir parameters based on petrophysics further includes: a judgment module 305;

[0192] The judgment module 305 is configured to judge whether there are first associated features with a correlation coefficient greater than a threshold among the multiple features;

[0193] When there are first associated features with a correlation coefficient greater than a threshold among the multiple features, the processing module 302 is further configured to compare the weight values of at least two features among the first associated features, and the first associated feature with a smaller weight value among the first associated features does not participate in the weight value sorting process, and the first associated feature with a larger weight value among the first associated features participates in the weight value sorting process;

[0194] When there are no first associated features with a correlation coefficient greater than a threshold among the multiple features, the processing module 302 is further configured to continue the sorting process of the multiple weight values in descending order of the weight value.

[0195] Optionally, the calculation module 303 is further configured to calculate the correlation coefficient between the multiple features using the following formula:

[0196]

[0197] where γ is the correlation coefficient, is the mean value of the logging curve, and x i , y i represent the logging curve values corresponding to the i-th sample;

[0198] The calculation module 303 is further configured to calculate the absolute value of the correlation coefficient γ;

[0199] The judgment module 305 is further configured to judge whether |γ| is greater than the threshold.

[0200] Optionally, the processing module 302 is further configured to construct a long short-term memory network model according to the initialized core data and the target features of the logging data;

[0201] The processing module 302 is further configured to perform training processing on the long short-term memory network model according to the initialized core data and the target features of the logging data to obtain a trained long short-term memory network model;

[0202] The processing module 302 is further configured to use the initialized core data as the input data of the trained long short-term memory network model, and the target features of the logging data as the output data of the trained long short-term memory network model, and control the long short-term memory network model to perform correlation analysis processing on the target features of the core data and the logging data to obtain the association relationship between the target features of the core data and the logging data.

[0203] Optionally, the processing module 302 is further configured to construct an input layer of the long short-term memory network model with a dimension corresponding to the number of features according to the number of features of the initialized core data;

[0204] The processing module 302 is further configured to construct an output layer of the long short-term memory network model with a dimension corresponding to the number of features according to the number of features of the target features of the logging data;

[0205] The processing module 302 is further configured to construct the number of units and layers of the long short-term memory network model according to the initialized core data and the target features of the logging data.

[0206] Optionally, the processing module 302 is further configured to use the association relationship between the core data and the target features of the logging data and the rock physics knowledge as the input data of the target prediction model, and control the target prediction model to perform iterative processing;

[0207] The obtaining module 301 is further configured to obtain the total loss function of the target prediction model, where the total loss function includes a first loss function of a long short-term memory network model and a second loss function of rock physics knowledge.

[0208] The processing module 302 is further configured to, when determining that the total loss function converges, control the target prediction model to stop iterative processing and determine that the training of the target prediction model is completed.

[0209] Figure 9 This is a schematic structural diagram of a reservoir parameter prediction device based on rock physics provided by the present application. As Figure 9 shown, the present application provides a reservoir parameter prediction device based on rock physics. The reservoir parameter prediction device 400 based on rock physics includes: a receiver 401, a transmitter 402, a processor 403, and a memory 404.

[0210] The receiver 401 is configured to receive instructions and data.

[0211] The transmitter 402 is configured to transmit instructions and data.

[0212] The memory 404 is configured to store computer execution instructions.

[0213] The processor 403 is configured to execute the computer execution instructions stored in the memory 404 to implement each step performed by the reservoir parameter prediction method based on rock physics in the above embodiments. For specific reference, see the relevant descriptions in the foregoing embodiments of the reservoir parameter prediction method based on rock physics.

[0214] Optionally, the above memory 404 can be either independent or integrated with the processor 403.

[0215] When the memory 404 is independently provided, the electronic device further includes a bus for connecting the memory 404 and the processor 403.

[0216] The present application further provides a computer storage medium. The computer storage medium stores computer execution instructions. When the processor executes the computer execution instructions, the reservoir parameter prediction method based on rock physics performed by the reservoir parameter prediction device based on rock physics as described above is implemented.

[0217] Those of ordinary skill in the art will understand that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, and their appropriate combinations. In the hardware implementation, the division of the functional modules / units mentioned above does not necessarily correspond to the division of physical components; for example, one physical component can have multiple functions, or one function or step can be executed by several physical components in cooperation. Some or all of the physical components can be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or can be implemented as hardware, or can be implemented as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include a computer storage medium (or non-transitory medium) and a communication medium (or transitory medium). As is well known to those of ordinary skill in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable instructions, data structures, program modules, or other data. Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory, or other memory technologies, CD-ROM, digital versatile disks (DVDs), or other optical disk storage, magnetic cartridges, tapes, magnetic disk storage, or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, as is well known to those of ordinary skill in the art, a communication medium typically includes computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transmission mechanism, and can include any information delivery medium.

[0218] After considering the specification and practicing the invention disclosed herein, those skilled in the art will readily conceive of other embodiments of the present application. The present application is intended to cover any variations, uses, or adaptations of the present application, which follow the general principles of the present application and include known common knowledge or conventional technical means in the technical field not disclosed in the present application. The specification and examples are only regarded as exemplary, and the true scope and spirit of the present application are pointed out by the following claims.

[0219] It should be understood that the present application is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present application is only limited by the appended claims.

Claims

1. A method for predicting reservoir parameters based on petrophysics, characterized in that, The method includes: Obtaining core data and logging data of the area to be measured; Performing initialization processing on the core data and logging data to obtain the initialized core data and logging data; inputting the initialized logging data into a random forest model, calculating the weights of multiple features of the logging data on reservoir parameters, and obtaining the target features of the logging data; Constructing a long short-term memory network model according to the initialized core data and the target features of the logging data, where the long short-term memory network model is used to determine the correlation between the core data and the target features of the logging data; Constructing a target prediction model according to the long short-term memory network model and petrophysical knowledge, using the correlation between the core data and the target features of the logging data and petrophysical knowledge as input data of the target prediction model, controlling the target prediction model to perform iterative processing, obtaining the total loss function of the target prediction model, and when it is determined that the total loss function converges, controlling the target prediction model to stop iterative processing and determining that the target prediction model is trained; where the total loss function of the target prediction model includes a first loss function of the long short-term memory network model and a second loss function of petrophysical knowledge; the target prediction model predicts reservoir parameters based on petrophysical knowledge; the petrophysical knowledge includes the relationships between different rock properties and reservoir parameters; The reservoir parameters include: porosity, permeability, and water saturation. The step of inputting the initialized logging data into a random forest model, calculating the weights of multiple features of the logging data on reservoir parameters, and obtaining the target features of the logging data includes: Inputting the initialized logging data into a random forest model, respectively calculating the influence weight values of the multiple features on the porosity, permeability, and water saturation to obtain porosity weight values, permeability weight values, and water saturation weight values; Performing sorting processing on the porosity weight values, permeability weight values, and water saturation weight values in descending order of weight values; Taking the features corresponding to the top preset number of porosity weight values after sorting as the target features of porosity; Taking the features corresponding to the top preset number of permeability weight values after sorting as the target features of permeability; Taking the features corresponding to the top preset number of water saturation weight values after sorting as the target features of water saturation.

2. The method according to claim 1, wherein Before performing the sorting processing on the porosity weight values, permeability weight values, and water saturation weight values in descending order of weight values, the method further includes: Judging whether there are first associated features with a correlation coefficient greater than a threshold among the multiple features; If so, comparing the weight values of at least two features in the first associated features, and the first associated feature with a smaller weight value in the first associated features does not participate in the weight value sorting process, while the first associated feature with a larger weight value in the first associated features participates in the weight value sorting process; If not, continue to perform the sorting process on the multiple weight values in descending order of weight values.

3. The method according to claim 2, wherein The determining whether there is a first associated feature with a correlation coefficient greater than a threshold value among the multiple features includes: The correlation coefficient between the multiple features is calculated using the following formula: Among them, is the correlation coefficient, and are the average values of logging curves, and represent the logging curve values corresponding to the th sample; Calculate the absolute value of the correlation coefficient and determine whether it is greater than the threshold value.

4. The method according to claim 1, characterized in that, The constructing of a long short-term memory network model according to the initialized core data and the target features of the logging data, wherein the long short-term memory network model is used to determine the correlation between the core data and the target features of the logging data, comprises: Constructing a long short-term memory network model according to the initialized core data and the target features of the logging data; According to the target features of the initialized core data and the logging data, a long short-term memory network model is trained to obtain a trained long short-term memory network model; The initialized core data is used as input data of the trained long short-term memory network model, and the target features of the well logging data are used as output data of the trained long short-term memory network model. The long short-term memory network model is controlled to perform correlation analysis on the target features of the core data and the well logging data to obtain the correlation relationship between the core data and the target features of the well logging data.

5. The method according to claim 4, wherein The initialized core data includes multiple features, and the long short-term memory network model is constructed according to the initialized core data and the target features of the logging data, including: According to the number of features of the initialized core data, a long short-term memory network model input layer with a dimension corresponding to the number of features is constructed; According to the number of features of the target features of the well logging data, a long short-term memory network model output layer with a dimension corresponding to the number of features is constructed; The number of units and the number of layers of the long short-term memory network model are constructed according to the target features of the initialized core data and the logging data.

6. A reservoir parameter prediction device based on petrophysics, characterized in that, The device comprises: An acquisition module is used to acquire core data and logging data of the area to be tested; A processing module, used for initializing the core data and the well logging data to obtain initialized core data and well logging data; A calculation module, used for inputting the initialized logging data into a random forest model, calculating the weights of multiple features of the logging data to oil reservoir parameters, and obtaining target features of the logging data; The processing module is further used to construct a long short-term memory network model according to the initialized core data and the target features of the logging data, and the long short-term memory network model is used to determine the correlation relationship between the core data and the target features of the logging data; The processing module is further configured to construct a target prediction model according to the long short-term memory network model and petrophysical knowledge, use the correlation relationship between the core data and the target features of the logging data and petrophysical knowledge as the input data of the target prediction model, control the target prediction model to perform iterative processing, obtain the total loss function of the target prediction model, and when it is determined that the total loss function converges, control the target prediction model to stop iterative processing and determine that the training of the target prediction model is completed; wherein, the total loss function of the target prediction model includes a first loss function of the long short-term memory network model and a second loss function of petrophysical knowledge; the target prediction model predicts reservoir parameters of an oil reservoir based on petrophysical knowledge, and the petrophysical knowledge includes the relationship between different rock properties and reservoir parameters of the oil reservoir; the reservoir parameters of the oil reservoir include: porosity, permeability, and water saturation. The calculation module is further configured to input the initialized logging data into a random forest model, and respectively calculate the influence weight values of multiple features in the logging data on porosity, permeability, and water saturation, to obtain a porosity weight value, a permeability weight value, and a water saturation weight value. The processing module is further configured to perform a sorting process on the porosity weight value, the permeability weight value, and the water saturation weight value in descending order of the weight values. The determination module is configured to use the features corresponding to the top preset number of porosity weight values after the sorting process as the target features of porosity. The determination module is further configured to use the features corresponding to the top preset number of permeability weight values after the sorting process as the target features of permeability. The determination module is further configured to use the features corresponding to the top preset number of water saturation weight values after the sorting process as the target features of water saturation.

7. A reservoir parameter prediction device based on petrophysics, characterized in that, including: a memory; a processor; wherein, the memory stores computer execution instructions; The processor executes the computer execution instructions stored in the memory to implement the petrophysics-based oil reservoir parameter prediction method according to any one of claims 1-5.

8. A computer storage medium, characterized in that, Computer execution instructions are stored in the computer storage medium, and when the computer execution instructions are executed by a processor, they are used to implement the petrophysics-based oil reservoir parameter prediction method according to any one of claims 1-5.

Images