A method for numerical simulation of hydraulic fracture propagation in deep oil and gas reservoirs
By combining image detection and model prediction with data on adamantane series compounds, the problem of large errors in predicting the type and production of deep oil and gas reservoirs has been solved. This has enabled accurate simulation of hydraulic fracture propagation and production prediction in deep oil and gas reservoirs, supporting the optimization of oilfield development.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- DAQING OILFIELD CO LTD
- Filing Date
- 2024-12-17
- Publication Date
- 2026-06-19
AI Technical Summary
Existing numerical simulation methods for hydraulic fracture propagation in deep oil and gas reservoirs cannot accurately determine the reservoir type, resulting in large errors in production prediction and failing to effectively guide the oilfield development process.
By configuring the parameters of the image detection equipment, the type of deep oil and gas reservoir is determined, image features are extracted, hydraulic fracture data is obtained, and a production prediction model is established using long short-term memory networks and Kalman filtering. The model takes into account reservoir heterogeneity and adamantane series compound data in crude oil samples for simulation and prediction.
It has enabled accurate prediction of artificial fractures in deep oil and gas reservoirs, providing a basis and guidance for the optimization design of oilfield development and improving the accuracy of production prediction.
Smart Images
Figure CN122242315A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of numerical simulation technology, and in particular to a method for numerical simulation of hydraulic fracture propagation in deep oil and gas reservoirs. Background Technology
[0002] Hydrodynamic oil and gas reservoirs are formed when hydrodynamic forces impede the migration of oil and gas along the updip direction, causing them to accumulate. The formation and preservation conditions of these reservoirs are complex; once hydrodynamic conditions change, they can transform into structural or stratigraphic reservoirs. These reservoirs tend to form near structural noses and flexural zones where the stratigraphic attitude changes slightly, in zones of heterogeneous lithology and thickness variation in monocline reservoirs, and near stratigraphic unconformities. In these locations, when the hydrodynamic pressure of seeping groundwater is opposite in direction and approximately equal in magnitude to the buoyancy of the oil and gas, it can impede and accumulate oil and gas, forming a hydrodynamic oil and gas reservoir. However, existing numerical simulation methods for hydraulic fracture propagation in deep oil and gas reservoirs cannot accurately determine the type of deep reservoir; furthermore, the prediction error for the production of deep oil and gas reservoirs is large.
[0003] For the reasons mentioned above, the calculation results of the artificial fracture prediction model for deep oil and gas reservoirs currently differ significantly from the actual situation in the field. It cannot accurately predict the propagation law of hydraulic fractures in deep oil and gas reservoirs, cannot effectively guide the adjustment of measures during oilfield development, has poor applicability, and cannot meet the needs of oilfields. Summary of the Invention
[0004] This invention addresses the problems in existing numerical simulation methods for hydraulic fracture propagation in deep oil and gas reservoirs, which fail to accurately determine reservoir type and exhibit large production prediction errors. It provides a new method for numerical simulation of hydraulic fracture propagation in deep oil and gas reservoirs. This method considers more comprehensive factors and can predict artificial fractures in deep oil and gas reservoirs through its model. The results can provide a basis and guidance for oilfield development optimization design.
[0005] The present invention solves its problem through the following technical solution: a method for numerical simulation of hydraulic fracture propagation in deep oil and gas reservoirs, comprising the following steps:
[0006] S101: Configure the parameters of the image detection equipment, detect deep oil and gas reservoir images through the detection equipment; determine the type of deep oil and gas reservoir; perform enhancement processing on the deep oil and gas reservoir images; extract the features of the deep oil and gas reservoir images;
[0007] S102: Obtain hydraulic fracture data of deep oil and gas reservoirs based on the extracted image features; and simulate the hydraulic fracture data of deep oil and gas reservoirs using a simulation program;
[0008] S103: Predict production from deep oil and gas reservoirs.
[0009] Furthermore, the method for determining the type of deep oil and gas reservoir in step S101 is as follows:
[0010] (1) Obtain the types and contents of adamantane series compounds in crude oil samples from a reference deep oil and gas reservoir; analyze and process the reference deep oil and gas reservoir to obtain the type of the reference deep oil and gas reservoir;
[0011] (2) Perform fitting analysis on the types and contents of the adamantane series compounds and the types of the reference deep oil and gas reservoirs to determine the fitting relationship between the types and contents of the adamantane series compounds and the types of the reference deep oil and gas reservoirs.
[0012] (3) Based on the fitting relationship and the data on the types and contents of adamantane series compounds in the crude oil samples of the deep oil and gas reservoir to be tested, the type of the deep oil and gas reservoir to be tested is determined.
[0013] Furthermore, the deep oil and gas reservoir type mentioned in step S101 includes at least one of the following:
[0014] Dry gas reservoirs, condensate gas reservoirs, volatile oil reservoirs, light oil reservoirs, and ordinary oil reservoirs.
[0015] Furthermore, the method for obtaining the types and contents of adamantane series compounds in crude oil samples from reference deep oil and gas reservoirs as described in step (1) includes:
[0016] Mass spectrometry analysis was performed on crude oil samples from the reference deep oil and gas reservoir to obtain the mass spectrometry information of the crude oil samples from the reference deep oil and gas reservoir.
[0017] Mass spectrometry information of adamantane series compounds was extracted from the crude oil sample of the reference deep oil and gas reservoir.
[0018] Based on the mass spectrometry information of the adamantane series compounds, the structures of the adamantane series compounds are identified to determine the types and contents of the adamantane series compounds in the crude oil samples of the reference deep oil and gas reservoir.
[0019] Furthermore, the adamantane series compounds include at least one of the following:
[0020] Monoadamantane series compounds, bisadamantane series compounds, triadamantane series compounds, tetraadamantane series compounds, and pentaadamantane series compounds.
[0021] Furthermore, step (3), based on the fitting relationship and the obtained data on the types and contents of adamantane series compounds in the crude oil samples of the deep oil and gas reservoir to be tested, determines the type of the deep oil and gas reservoir to be tested, including:
[0022] When the types and contents of adamantane series compounds in the crude oil sample of the deep oil and gas reservoir to be tested are within the first preset value range, the deep oil and gas reservoir to be tested is determined to be an ordinary oil reservoir.
[0023] When the types and contents of adamantane series compounds in the crude oil sample of the deep oil and gas reservoir to be tested are within the second preset value range, the deep oil and gas reservoir to be tested is determined to be a light oil reservoir.
[0024] When the types and contents of adamantane series compounds in the crude oil sample of the deep oil and gas reservoir to be tested are within the third preset value range, the deep oil and gas reservoir to be tested is determined to be a volatile oil reservoir.
[0025] When the types and contents of adamantane series compounds in the crude oil sample of the deep oil and gas reservoir to be tested are within the fourth preset value range, the deep oil and gas reservoir to be tested is determined to be a condensate gas reservoir.
[0026] When the types and contents of adamantane series compounds in the crude oil sample of the deep oil and gas reservoir to be tested are within the fifth preset value range, the deep oil and gas reservoir to be tested is determined to be a dry gas reservoir.
[0027] Furthermore, the method for predicting the production of deep oil and gas reservoirs in step S103 is as follows:
[0028] 1) Establish a deep oil and gas reservoir production prediction model based on long short-term memory network and Kalman filter through modeling program;
[0029] 2) Acquire reservoir data, preprocess the reservoir data, and select characteristic parameters from the preprocessed reservoir data through correlation analysis and stepwise regression methods. Construct a dataset based on the characteristic parameters, wherein the correlation analysis is based on Pearson correlation coefficient or Spearman rank correlation coefficient.
[0030] 3) The dataset is trained and predicted using the deep oil and gas reservoir production prediction model to obtain prediction results.
[0031] Furthermore, the deep oil and gas reservoir production prediction model in step 1) includes a static model and a dynamic adjustment model. The static model predicts oil and gas production through a long short-term memory network, and the dynamic adjustment model dynamically adjusts the predicted production through Kalman filtering.
[0032] Furthermore, the reservoir data in step 2) includes a training set and a prediction set, wherein the training set is used to train the model and the prediction set is used to predict the prediction accuracy of the performance model.
[0033] Furthermore, the preprocessing in step 2) includes one or a combination of the following: missing value processing, outlier processing, and removal of irrelevant variables.
[0034] The missing value handling methods include at least one of the following: deleting completely empty variables, filling with the mean of a sequence, filling with the nearest neighbor mean, filling with the median, and filling with linear interpolation.
[0035] Furthermore, the static model in step 1) includes a sliding window model, which is used to predict data at multiple future time points;
[0036] The static model processing procedure includes an input layer, a hidden layer, and an output layer; the number of neurons in the input layer is determined by the number of input variables.
[0037] Furthermore, step 3) involves training and predicting the dataset using the deep oil and gas reservoir production prediction model as follows:
[0038] The dataset is used as the input to the static model, and the output of the static model is used as the observation set for the Kalman filter. The output is corrected by the dynamic adjustment model. The prediction effect of the model is evaluated by indicators, and the output and prediction results are evaluated and analyzed to obtain the optimal model affecting the daily gas production of oil wells.
[0039] Compared with the above-mentioned background technology, the present invention has the following beneficial effects:
[0040] This invention provides a numerical simulation method for hydraulic fracture propagation in deep oil and gas reservoirs. Compared with existing methods, this method considers more comprehensive factors, including for the first time reservoir heterogeneity, the types and contents of adamantane series compounds in the crude oil sample containing the artificial fractures, and the type of the reference deep oil and gas reservoir. The model can predict artificial fractures in deep oil and gas reservoirs. The results can provide a basis and guidance for the optimization design of oilfield development. Attached Figure Description
[0041] Figure 1 This is a flowchart of a method for numerical simulation of hydraulic fracture propagation in deep oil and gas reservoirs according to the present invention.
[0042] Figure 2 A flowchart of the method for determining deep oil and gas reservoir types provided by the present invention;
[0043] Figure 3 The flowchart of the method for predicting the production of deep oil and gas reservoirs provided by the present invention is shown. Detailed Implementation
[0044] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.
[0045] like Figure 1As shown, this invention discloses a method for numerical simulation of hydraulic fracture propagation in deep oil and gas reservoirs, comprising: S101, configuring image detection equipment parameters, detecting deep oil and gas reservoir images using the detection equipment; determining the type of deep oil and gas reservoir; enhancing the deep oil and gas reservoir images; extracting features from the deep oil and gas reservoir images; S102, obtaining hydraulic fracture data of the deep oil and gas reservoir based on the extracted features; and simulating the hydraulic fracture data of the deep oil and gas reservoir using a simulation program; S103, predicting the production of the deep oil and gas reservoir. Specifically, it includes the following steps:
[0046] S101, Configure the parameters of the image detection equipment, detect deep oil and gas reservoir images through the detection equipment; determine the type of deep oil and gas reservoir; perform enhancement processing on the deep oil and gas reservoir images; extract the features of the deep oil and gas reservoir images;
[0047] like Figure 2 As shown, the method for determining the type of deep oil and gas reservoirs provided by this invention is as follows:
[0048] S201, Obtain the types and contents of adamantane series compounds in crude oil samples from a reference deep oil and gas reservoir; Analyze and process the reference deep oil and gas reservoir to obtain the type of the reference deep oil and gas reservoir;
[0049] The acquisition of data on the types and contents of adamantane series compounds in crude oil samples from reference deep oil and gas reservoirs specifically includes:
[0050] Mass spectrometry analysis was performed on crude oil samples from the reference deep oil and gas reservoir to obtain the mass spectrometry information of the crude oil samples from the reference deep oil and gas reservoir.
[0051] Mass spectrometry information of adamantane series compounds was extracted from the crude oil sample of the reference deep oil and gas reservoir.
[0052] Based on the mass spectrometry information of the adamantane series compounds, the structures of the adamantane series compounds are identified to determine the types and contents of the adamantane series compounds in the crude oil samples of the reference deep oil and gas reservoir.
[0053] The adamantane series compounds include at least one of the following:
[0054] Monoadamantane series compounds, bisadamantane series compounds, triadamantane series compounds, tetraadamantane series compounds, pentaadamantane series compounds.
[0055] The deep oil and gas reservoir types include at least one of the following: dry gas reservoir, condensate gas reservoir, volatile oil reservoir, light oil reservoir, and ordinary oil reservoir.
[0056] S202, perform fitting analysis on the types and contents of the adamantane series compounds and the type of the reference deep oil and gas reservoir to determine the fitting relationship between the types and contents of the adamantane series compounds and the type of the reference deep oil and gas reservoir.
[0057] S203, based on the fitting relationship and the obtained data on the types and contents of adamantane series compounds in crude oil samples from the deep oil and gas reservoir to be tested, the type of the deep oil and gas reservoir to be tested is determined; specifically, the process of determining the type of the deep oil and gas reservoir to be tested includes:
[0058] When the types and contents of adamantane series compounds in the crude oil sample of the deep oil and gas reservoir to be tested are within the first preset value range, the deep oil and gas reservoir to be tested is determined to be an ordinary oil reservoir.
[0059] When the types and contents of adamantane series compounds in the crude oil sample of the deep oil and gas reservoir to be tested are within the second preset value range, the deep oil and gas reservoir to be tested is determined to be a light oil reservoir.
[0060] When the types and contents of adamantane series compounds in the crude oil sample of the deep oil and gas reservoir to be tested are within the third preset value range, the deep oil and gas reservoir to be tested is determined to be a volatile oil reservoir.
[0061] When the types and contents of adamantane series compounds in the crude oil sample of the deep oil and gas reservoir to be tested are within the fourth preset value range, the deep oil and gas reservoir to be tested is determined to be a condensate gas reservoir.
[0062] When the types and contents of adamantane series compounds in the crude oil sample of the deep oil and gas reservoir to be tested are within the fifth preset value range, the deep oil and gas reservoir to be tested is determined to be a dry gas reservoir.
[0063] S102, Obtain hydraulic fracture data of deep oil and gas reservoirs based on the extracted image features; and simulate the hydraulic fracture data of deep oil and gas reservoirs through a simulation program;
[0064] S103, predicting the production of deep oil and gas reservoirs;
[0065] like Figure 3 As shown, the method for predicting the production of deep oil and gas reservoirs, as described in S103, includes the following steps:
[0066] S301, A deep oil and gas reservoir production prediction model based on long short-term memory network and Kalman filter is established through a modeling program. The deep oil and gas reservoir production prediction model includes a static model and a dynamic adjustment model. The static model predicts the production of oil and gas through long short-term memory network, and the dynamic adjustment model dynamically adjusts the predicted production through Kalman filter.
[0067] The static model provided by this invention includes a sliding window model, which is used to predict data at multiple future time points;
[0068] The static model processing method includes an input layer, a hidden layer, and an output layer. The number of neurons in the input layer is determined by the number of input variables.
[0069] S302, acquire reservoir data, preprocess the reservoir data, the preprocessing includes one or a combination of the following: missing value processing, outlier processing and removal of irrelevant variables; for the preprocessed reservoir data, select feature parameters through correlation analysis and stepwise regression methods, and construct a dataset based on the feature parameters, wherein the correlation analysis is based on Pearson correlation coefficient or Spearman rank correlation coefficient;
[0070] The reservoir data includes a training set and a prediction set. The training set is used to train the model, and the prediction set is used to predict the prediction accuracy of the performance model.
[0071] The missing value handling includes one or a combination of the following: deleting completely empty variables, filling with the mean of the sequence, filling with the nearest neighbor mean, filling with the median, and filling with linear interpolation.
[0072] S303, The dataset is trained and predicted using the deep oil and gas reservoir production prediction model to obtain prediction results. The dataset is used as the input of the static model, and the output of the static model is used as the observation set for the Kalman filter. The output results are corrected by the dynamic adjustment model. The prediction effect of the model is evaluated by indicators, and the output results and the prediction results are evaluated and analyzed to obtain the optimal model affecting the daily gas production of oil wells.
[0073] Application Examples
[0074] To demonstrate the inventiveness and technical value of the technical solution of this invention, this section provides specific product or related technology application examples of the technical solution claimed.
[0075] This invention, through a method for determining the type of deep oil and gas reservoir, obtains data on the types and contents of adamantane series compounds in crude oil samples of a reference deep oil and gas reservoir, as well as the type of the reference deep oil and gas reservoir. By performing a fitting analysis on the data on the types and contents of adamantane series compounds and the type of the reference deep oil and gas reservoir, a fitting relationship can be obtained. Then, based on this fitting relationship and the obtained data on the types and contents of adamantane series compounds in crude oil samples of the deep oil and gas reservoir to be tested, the type of the deep oil and gas reservoir to be tested can be quickly and accurately determined. Simultaneously, a method for predicting the production of deep oil and gas reservoirs can accurately predict their production.
[0076] It should be noted that embodiments of the present invention can be implemented in hardware, software, or a combination of both. The hardware portion can be implemented using dedicated logic; the software portion can be stored in memory and executed by a suitable instruction execution system, such as a microprocessor or dedicated-design hardware. Those skilled in the art will understand that the above-described devices and methods can be implemented using computer-executable instructions and / or included in processor control code, for example, such code provided on a carrier medium such as a disk, CD, or DVD-ROM, a programmable memory such as read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The devices and modules of the present invention can be implemented by hardware circuitry such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field-programmable gate arrays, programmable logic devices, etc., or by software executed by various types of processors, or by a combination of the above-described hardware circuitry and software, such as firmware.
[0077] Those skilled in the art will recognize that the embodiments described herein are intended to help the reader understand the implementation methods of the present invention, and should be understood that the scope of protection of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific modifications and combinations based on the technical teachings disclosed in this invention without departing from the spirit of the invention, and these modifications and combinations are still within the scope of protection of the present invention.
Claims
1. A method for numerical simulation of hydraulic fracture propagation in deep oil and gas reservoirs, characterized in that: Includes the following steps: S101: Configure the parameters of the image detection equipment, detect deep oil and gas reservoir images through the detection equipment; determine the type of deep oil and gas reservoir; perform enhancement processing on the deep oil and gas reservoir images; extract the features of the deep oil and gas reservoir images; S102: Obtain hydraulic fracture data of deep oil and gas reservoirs based on the extracted image features; and simulate the hydraulic fracture data of deep oil and gas reservoirs using a simulation program; S103: Predict production from deep oil and gas reservoirs.
2. The method for numerical simulation of hydraulic fracture propagation in deep oil and gas reservoirs according to claim 1, characterized in that: The method for determining the type of deep oil and gas reservoir in step S101 is as follows: (1) Obtain the types and contents of adamantane series compounds in crude oil samples from a reference deep oil and gas reservoir; analyze and process the reference deep oil and gas reservoir to obtain the type of the reference deep oil and gas reservoir; (2) Perform fitting analysis on the types and contents of the adamantane series compounds and the types of the reference deep oil and gas reservoirs to determine the fitting relationship between the types and contents of the adamantane series compounds and the types of the reference deep oil and gas reservoirs. (3) Based on the fitting relationship and the data on the types and contents of adamantane series compounds in the crude oil samples of the deep oil and gas reservoir to be tested, the type of the deep oil and gas reservoir to be tested is determined.
3. The method for numerical simulation of hydraulic fracture propagation in deep oil and gas reservoirs according to claim 1, characterized in that: The deep oil and gas reservoir type mentioned in step S101 includes at least one of the following: Dry gas reservoirs, condensate gas reservoirs, volatile oil reservoirs, light oil reservoirs, and ordinary oil reservoirs.
4. The method for numerical simulation of hydraulic fracture propagation in deep oil and gas reservoirs according to claim 2, characterized in that: The method for obtaining the types and contents of adamantane series compounds in crude oil samples from reference deep oil and gas reservoirs as described in step (1) includes: Mass spectrometry analysis was performed on crude oil samples from the reference deep oil and gas reservoir to obtain the mass spectrometry information of the crude oil samples from the reference deep oil and gas reservoir. Mass spectrometry information of adamantane series compounds was extracted from the crude oil sample of the reference deep oil and gas reservoir. Based on the mass spectrometry information of the adamantane series compounds, the structures of the adamantane series compounds are identified to determine the types and contents of the adamantane series compounds in the crude oil samples of the reference deep oil and gas reservoir.
5. The method for numerical simulation of hydraulic fracture propagation in deep oil and gas reservoirs according to claim 4, characterized in that: The adamantane series compounds include at least one of the following: Monoadamantane series compounds, bisadamantane series compounds, triadamantane series compounds, tetraadamantane series compounds, and pentaadamantane series compounds.
6. The method for numerical simulation of hydraulic fracture propagation in deep oil and gas reservoirs according to claim 4, characterized in that: Step (3), based on the fitting relationship and the obtained data on the types and contents of adamantane series compounds in the crude oil samples of the deep oil and gas reservoir to be tested, determines the type of the deep oil and gas reservoir to be tested, including: When the types and contents of adamantane series compounds in the crude oil sample of the deep oil and gas reservoir to be tested are within the first preset value range, the deep oil and gas reservoir to be tested is determined to be an ordinary oil reservoir. When the types and contents of adamantane series compounds in the crude oil sample of the deep oil and gas reservoir to be tested are within the second preset value range, the deep oil and gas reservoir to be tested is determined to be a light oil reservoir. When the types and contents of adamantane series compounds in the crude oil sample of the deep oil and gas reservoir to be tested are within the third preset value range, the deep oil and gas reservoir to be tested is determined to be a volatile oil reservoir. When the types and contents of adamantane series compounds in the crude oil sample of the deep oil and gas reservoir to be tested are within the fourth preset value range, the deep oil and gas reservoir to be tested is determined to be a condensate gas reservoir. When the types and contents of adamantane series compounds in the crude oil sample of the deep oil and gas reservoir to be tested are within the fifth preset value range, the deep oil and gas reservoir to be tested is determined to be a dry gas reservoir.
7. The method for numerical simulation of hydraulic fracture propagation in deep oil and gas reservoirs according to claim 6, characterized in that: The method for predicting the production of deep oil and gas reservoirs in step S103 is as follows: 1) Establish a deep oil and gas reservoir production prediction model based on long short-term memory network and Kalman filter through modeling program; 2) Acquire reservoir data, preprocess the reservoir data, and select characteristic parameters from the preprocessed reservoir data through correlation analysis and stepwise regression methods. Construct a dataset based on the characteristic parameters, wherein the correlation analysis is based on Pearson correlation coefficient or Spearman rank correlation coefficient. 3) The dataset is trained and predicted using the deep oil and gas reservoir production prediction model to obtain prediction results.
8. The method for numerical simulation of hydraulic fracture propagation in deep oil and gas reservoirs according to claim 7, characterized in that: The deep oil and gas reservoir production prediction model in step 1) includes a static model and a dynamic adjustment model. The static model predicts oil and gas production through a long short-term memory network, and the dynamic adjustment model dynamically adjusts the predicted production through Kalman filtering.
9. The method for numerical simulation of hydraulic fracture propagation in deep oil and gas reservoirs according to claim 8, characterized in that: The reservoir data in step 2) includes a training set and a prediction set. The training set is used to train the model, and the prediction set is used to predict the prediction accuracy of the performance model.
10. The method for numerical simulation of hydraulic fracture propagation in deep oil and gas reservoirs according to claim 9, characterized in that: The preprocessing in step 2) includes one or a combination of the following: missing value processing, outlier processing, and removal of irrelevant variables; The missing value handling methods include at least one of the following: deleting completely empty variables, filling with the mean of a sequence, filling with the nearest neighbor mean, filling with the median, and filling with linear interpolation.
11. The method for numerical simulation of hydraulic fracture propagation in deep oil and gas reservoirs according to claim 10, characterized in that: The static model in step 1) includes a sliding window model, which is used to predict data at multiple future time points; The static model processing procedure includes an input layer, a hidden layer, and an output layer; the number of neurons in the input layer is determined by the number of input variables.
12. The method for numerical simulation of hydraulic fracture propagation in deep oil and gas reservoirs according to claim 11, characterized in that: The method for training and predicting the dataset using the deep oil and gas reservoir production prediction model in step 3) is as follows: The dataset is used as the input to the static model, and the output of the static model is used as the observation set for the Kalman filter. The output is corrected by the dynamic adjustment model. The prediction effect of the model is evaluated by indicators, and the output and prediction results are evaluated and analyzed to obtain the optimal model affecting the daily gas production of oil wells.