Crack prediction method, electronic device and storage medium

Abstract

This application discloses a fracture prediction method, electronic device, and storage medium. The method includes: obtaining a fracture density curve for a first well and multiple seismic attribute curves corresponding to multiple seismic attributes associated with the first well; determining a correlation coefficient and / or similarity between each seismic attribute curve and the fracture density curve; selecting at least one target seismic attribute from the multiple seismic attributes based on the correlation coefficient and / or similarity; and, for a second well to be predicted, obtaining multiple target seismic attribute curves corresponding to the target seismic attribute, and determining a target fracture density curve for the second well based on the target seismic attribute curves using a pre-trained seismic fracture prediction model. This embodiment improves the accuracy of fracture prediction.

Description

Technical Field

[0001] The present application belongs to the technical field of fracture exploration, and specifically relates to a fracture prediction method, electronic equipment, and storage medium. Background Art

[0002] With the continuous advancement of geophysical exploration, fractured oil and gas reservoirs with enormous development potential are emerging as one of the main areas of oil and gas exploration. Fractures play a significant role in the storage and migration of oil and gas, significantly affecting the oil and gas production capacity of reservoirs. Accurately predicting the spatial distribution of fractures is crucial for studying the distribution patterns of fractured oil and gas reservoirs, improving exploration success rates, enhancing development efficiency, and increasing and stabilizing production.

[0003] Seismic attribute analysis is a commonly used technique for fracture prediction. Coherence analysis, curvature analysis, and ant tracking are among the attributes commonly used in fracture interpretation. However, conventional attributes, especially single ones, can only identify large faults on the scale of meters or tens of meters, but lack the ability to identify centimeter-scale fractures that are widespread underground. Summary of the Invention

[0004] The embodiments of the present application provide a crack prediction method, electronic device, and storage medium to solve the problem of low accuracy in predicting small-scale cracks in related technologies.

[0005] In a first aspect, an embodiment of the present application provides a crack prediction method, comprising:

[0006] obtaining a fracture density curve of a first well and a plurality of seismic attribute curves corresponding to a plurality of seismic attributes associated with the first well;

[0007] determining a correlation coefficient and / or similarity between each seismic attribute curve and the fracture density curve;

[0008] selecting at least one target seismic attribute from the plurality of seismic attributes according to the correlation coefficient and / or similarity;

[0009] For the second well to be predicted, multiple target seismic attribute curves corresponding to the target seismic attributes are obtained, and based on the target seismic attribute curves, a target fracture density curve of the second well is determined using a pre-trained seismic fracture prediction model.

[0010] In a second aspect, an embodiment of the present application further provides a crack prediction device, comprising:

[0011] an acquisition module, configured to acquire a fracture density curve of a first well and a plurality of seismic attribute curves corresponding to a plurality of seismic attributes associated with the first well;

[0012] a determination module, configured to determine a correlation coefficient and / or similarity between each seismic attribute curve and the fracture density curve;

[0013] A selection module, configured to select at least one target seismic attribute from the plurality of seismic attributes according to the correlation coefficient and / or similarity;

[0014] The prediction module is used to obtain a plurality of target seismic attribute curves corresponding to the target seismic attributes for the second well to be predicted, and determine the target fracture density curve of the second well based on the target seismic attribute curves using a pre-trained seismic fracture prediction model.

[0015] In a third aspect, an embodiment of the present application provides an electronic device comprising a processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements the steps of the method described in the first aspect.

[0016] In a fourth aspect, an embodiment of the present application provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the method described in the first aspect are implemented.

[0017] The crack prediction method provided in the embodiment of the present application obtains a crack density curve of a first well and a plurality of seismic attribute curves corresponding to a plurality of seismic attributes related to the first well; determines the correlation coefficient and / or similarity between each seismic attribute curve and the crack density curve; selects at least one target seismic attribute from the plurality of seismic attributes based on the correlation coefficient and / or similarity; obtains a plurality of target seismic attribute curves corresponding to the target seismic attribute for a second well to be predicted, and determines the target crack density curve of the second well based on the target seismic attribute curve using a pre-trained seismic crack prediction model; since the formation of cracks has multiple causes, and a single seismic attribute can only predict cracks of a single cause, this embodiment uses the known seismic attribute curves of the first well to predict the crack density curve. The target seismic attribute is determined by measuring the correlation coefficient and / or similarity between the fracture density curve of the first well and the multiple seismic attribute curves of the first well, so that the determined target seismic attribute has a high correlation with the fracture cause of the first well, that is, determining the target seismic attribute also determines the fracture cause of the first well; in addition, by applying the target seismic attribute of the first well to the second well, when the second well and the first well are in the same environment, the target seismic attribute also has a high correlation with the fracture cause of the second well. At this time, according to the target seismic attribute curve, the target fracture density curve of the second well is predicted by using the pre-trained seismic fracture prediction model, thereby avoiding the problem of poor crack recognition ability of a single seismic attribute or an erroneous seismic attribute, which leads to low crack prediction accuracy, and improving the crack prediction accuracy and precision. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 1 is a flow chart of a crack prediction method in an embodiment of the present application;

[0019] Figure 2 is a schematic diagram of the crack density curve processing process in the embodiment of the present application;

[0020] Figure 3 Schematic diagram of the correlation coefficient between the predicted crack density and the actual crack density of the earthquake crack prediction model in the embodiment of the present application;

[0021] Figure 4 Schematic diagram comparing well prediction results and imaging logging in the embodiment of the present application;

[0022] Figure 5 This is a schematic structural diagram of a crack prediction device in an embodiment of the present application;

[0023] Figure 6 It is a structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0025] The terms "first," "second," and the like in the specification and claims of this application are used to distinguish similar objects, and are not used to describe a specific order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, so that the embodiments of this application can be implemented in an order other than that illustrated or described herein, and that the objects distinguished by "first," "second," and the like are generally of the same type, and do not limit the number of objects; for example, the first object can be one or more. In addition, the term "and / or" in the specification and claims refers to at least one of the connected objects, and the character " / " generally indicates that the objects connected are in an "or" relationship.

[0026] Specifically, seismic attribute analysis is a commonly used fracture prediction technique. Attributes such as coherence, curvature, and ant-body are often used in fracture interpretation. However, conventional attributes, especially single conventional attributes, can only identify large faults on the scale of meters or tens of meters, lacking the ability to identify widespread centimeter-scale fractures in the subsurface. Fractures have various origins, including tectonic and fault-related fractures. Therefore, fractures respond to various attributes. For example, the presence of effective fractures causes attenuation and dispersion of seismic signals, so frequency attenuation gradient attributes and coherence attributes can indicate the extent of fracture development to a certain extent. Tectonic fractures are related to structure, so tectonic attributes such as curvature can also indicate the extent of fracture development to a certain extent. Fault-derived fractures are positively correlated with fault development, so fault-related attributes such as ant-body can also indicate the extent of fracture development.

[0027] However, under complex geological conditions, cracks have multiple causes. A single attribute can only predict cracks of a single cause, but the identification effect is poor for crack systems with multiple causes.

[0028] In addition, most conventional attribute fusion technologies are based on simple pixel superposition. Although the fusion results increase the probability of larger-scale cracks, different attributes may interfere with each other, resulting in the loss of some detailed information and affecting the accuracy of crack identification. In addition, different attributes have different applicable scopes for cracks of different causes. Conventional fusion technology will cause high-density responses in low-density areas, resulting in large errors.

[0029] To this end, this embodiment obtains a fracture density curve of a first well and multiple seismic attribute curves corresponding to multiple seismic attributes related to the first well; determines the correlation coefficient and / or similarity between each seismic attribute curve and the fracture density curve; selects at least one target seismic attribute from the multiple seismic attributes based on the correlation coefficient and / or similarity; obtains multiple target seismic attribute curves corresponding to the target seismic attribute for a second well to be predicted, and determines the target fracture density curve of the second well based on the target seismic attribute curve using a pre-trained seismic fracture prediction model; since the formation of fractures has multiple causes, and a single seismic attribute can only predict fractures of a single cause, this embodiment uses the correlation between the known fracture density curve of the first well and the multiple seismic attribute curves of the first well. The target seismic attributes are determined by the number and / or similarity, so that the determined target seismic attributes have a high correlation with the cause of the fractures in the first well, that is, determining the target seismic attributes also determines the cause of the fractures in the first well. In addition, by applying the target seismic attributes of the first well to the second well, when the second well is in the same environment as the first well, the target seismic attributes also have a high correlation with the cause of the fractures in the second well. At this time, according to the target seismic attribute curve, the target fracture density curve of the second well is predicted by the pre-trained seismic fracture prediction model, thereby avoiding the problem that a single seismic attribute or an erroneous seismic attribute has poor crack recognition ability and thus leads to low crack prediction accuracy. It also avoids the problem that different attributes may interfere with each other, resulting in the loss of some detailed information and affecting the crack recognition accuracy, thereby improving the crack prediction accuracy and precision.

[0030] The following application describes in detail the crack prediction method provided by the embodiment of the present application through specific embodiments and their application scenarios in conjunction with the accompanying drawings.

[0031] Figure 1 A crack prediction method provided by an embodiment of the present invention is shown. The method can be executed by an electronic device, which may include a server and / or a terminal device. In other words, the method can be executed by software or hardware installed on the electronic device. The method includes the following steps:

[0032] Step 101: Obtain a fracture density curve of a first well and a plurality of seismic attribute curves corresponding to a plurality of seismic attributes related to the first well.

[0033] Specifically, the first well is a well for which fracture development results have been determined, that is, a well for which a fracture density curve has been determined.

[0034] In this step, a fracture density curve of the first well and a plurality of seismic attribute curves corresponding to a plurality of seismic attributes related to the first well may be obtained.

[0035] Specifically, the seismic attributes may include instantaneous frequency, maximum curvature, root mean square amplitude, instantaneous bandwidth and other attributes.

[0036] In addition, each seismic attribute can determine the number of seismic attribute curves corresponding to it according to the number of curve parameters (ie, statistical parameters). For example, if there are N curve parameters, each seismic attribute can correspond to N seismic attribute curves.

[0037] Step 102: Determine the correlation coefficient and / or similarity between each seismic attribute curve and the fracture density curve.

[0038] Specifically, in this step, the similarity between each seismic attribute curve and the fracture density curve can be calculated through Neighborhood Component Analysis (NCA). The specific calculation formula is:

[0039]

[0040] Among them, P ij represents the similarity between the i-th seismic attribute curve and the fracture density curve j, x i represents the i-th earthquake attribute curve, x j Represents the crack density curve.

[0041] In addition, the correlation coefficient between each seismic attribute curve and the fracture density curve represents the correlation between each seismic attribute curve and the fracture density curve.

[0042] By determining the correlation coefficient and / or similarity between each seismic attribute curve and the fracture density curve, the relationship between each seismic attribute and the fracture cause of the first well can be obtained through correlation coefficient and / or similarity analysis. Of course, the higher the correlation coefficient and / or similarity, the greater the correlation between the seismic attribute and the fracture cause of the first well.

[0043] Step 103: Select at least one target seismic attribute from the multiple seismic attributes according to the correlation coefficient and / or similarity.

[0044] Specifically, this step selects at least one target seismic attribute from multiple seismic attributes based on the correlation coefficient, or selects at least one target seismic attribute from multiple seismic attributes by similarity, or selects at least one target seismic attribute from multiple seismic attributes by the correlation coefficient and similarity.

[0045] Since the size of the correlation coefficient and / or similarity represents whether the cause of the fracture in the first well predicted by the seismic attribute is accurate, by selecting at least one target seismic attribute from multiple seismic attributes based on the correlation coefficient and / or similarity, the accuracy of the selected target seismic attribute is improved.

[0046] Step 104: For the second well to be predicted, a plurality of target seismic attribute curves corresponding to the target seismic attributes are obtained, and a target fracture density curve of the second well is determined based on the target seismic attribute curves using a pre-trained seismic fracture prediction model.

[0047] Specifically, the second well to be predicted may be in the same environment as the first well. For example, the second well may be within a preset range of the first well, or it may be determined through geological measurements that the geology of the second well is similar to that of the first well, or the second well and the first well are located in the same fault block. In this case, the second well is considered to have a similar geomechanical environment as the first well.

[0048] This step applies the target seismic attributes determined in the first well to the second well where fractures are to be predicted, obtaining multiple target seismic attribute curves corresponding to the target seismic attributes in the second well. Because the target seismic attributes are selected based on the first well, they can accurately predict the fracture conditions in the second well, avoiding the problem of poor fracture recognition ability due to a single seismic attribute or an erroneous seismic attribute, which can lead to low fracture prediction accuracy.

[0049] In addition, by determining the target fracture density curve of the second well based on the standard seismic attribute curve and the pre-trained seismic fracture prediction model, the problem of interference between different attributes, resulting in the loss of some detailed information and affecting the accuracy of fracture identification, is avoided, thereby improving the prediction precision and accuracy of fractures.

[0050] Specifically, in one embodiment, when obtaining the fracture density curve of the first well, the fracture interpretation results of the imaging logging of the first well can be obtained; the fracture interpretation results are converted into a logging curve according to the seismic resolution; and the logging curve is smoothed to obtain the fracture density curve.

[0051] Specifically, such as Figure 2 As shown, the imaging log for the first well can be obtained, along with the fracture interpretation results from the imaging log. These fracture interpretation results can then be converted into a log curve based on the seismic resolution. Furthermore, because seismic resolution differs significantly from that of logs, when correlating fracture density with seismic data, the seismic response represents a comprehensive response near the target layer. Therefore, the fracture density needs to be smoothed to reflect its variation. In this case, the log curve can be smoothed to obtain a fracture density log curve.

[0052] Specifically, the well logging curve may be smoothed according to the following formula:

[0053]

[0054] in, represents the crack density of the i-th sample point after smoothing; B i+1 represents the crack density of the i+1th sample before smoothing; B i represents the value of the i-th sample before smoothing; B i-1 Represents the value of the i-1th sample before smoothing.

[0055] Furthermore, in one embodiment, selecting at least one target seismic attribute from the plurality of seismic attributes according to the correlation coefficient and / or similarity comprises:

[0056] Select at least one first target correlation coefficient greater than a first preset coefficient value from the correlation coefficients, and / or select at least one first target similarity greater than a first preset similarity value from the similarities; determine the seismic attribute corresponding to the at least one first target correlation coefficient and / or the seismic attribute corresponding to the at least one first target similarity as the target seismic attribute.

[0057] Specifically, by selecting at least one first target correlation coefficient greater than a first preset coefficient value, and / or selecting at least one first target similarity greater than a first preset similarity value, and determining the seismic attribute corresponding to at least one first target correlation coefficient as the target seismic attribute, or determining the seismic attribute corresponding to at least one first target similarity as the target seismic attribute, or determining the seismic attribute corresponding to at least one first target correlation coefficient and the seismic attribute corresponding to at least one first target similarity as the target seismic attribute, the determined target seismic attribute is highly correlated with the cause of the fractures in the first well, that is, the selected target seismic attribute can better reflect the actual situation of the fractures in the first well, avoiding the problem of poor fracture recognition ability of erroneous seismic attributes and thus low fracture prediction accuracy, and improving the accuracy of predicting fractures in the second well through target seismic attributes.

[0058] Furthermore, in one embodiment, obtaining a plurality of seismic attribute curves corresponding to a plurality of seismic attributes associated with the first well may include the following steps:

[0059] For each of a plurality of pre-set statistical radii, at least one curve parameter corresponding to each of a plurality of seismic attributes is determined; and a seismic attribute curve corresponding to the seismic attribute is determined based on the curve parameters, wherein each of the curve parameters corresponds to a seismic attribute curve.

[0060] Specifically, the pre-set multiple statistical radii can be selected from 25 meters to 150 meters at intervals of 25 meters, for example, the statistical radii include 25 meters, 50 meters, 75 meters, 100 meters, 125 meters, 150 meters, etc. Of course, it should be noted that the pre-set multiple statistical radii can also be other values, which are not specifically limited here.

[0061] In addition, the curve parameters may include at least one of statistical parameters such as variance, mean, percentile, etc. The seismic attributes may include instantaneous frequency, maximum curvature, root mean square amplitude, instantaneous bandwidth, etc.

[0062] This step can calculate the statistical parameters (such as variance, mean, etc.) of different seismic attributes under different statistical radii, determine at least one curve parameter corresponding to each seismic attribute, and determine the seismic attribute curve corresponding to the seismic attribute based on the curve parameter.

[0063] It should be noted that, because different attributes may have different amplitudes and physical meanings, sensitive attributes need to be normalized to a value range of [0, 1]. That is, this embodiment can normalize the curve parameters and obtain the seismic attribute curve corresponding to each curve parameter based on the normalized curve parameters. Specifically, the curve parameters can be normalized using the following formula, thereby limiting the normalized values ​​to between 0 and 1:

[0064]

[0065] Where A represents the parameter value in the curve parameters, and max(A) represents the maximum parameter value in the curve parameters.

[0066] For example, taking a 25m radius as an example, when determining the seismic attribute curve, a sample point can be selected, with the sample point as the center, and the statistical parameters calculated for all points within the statistical radius are determined as the value of the sample point, and so on to obtain a curve (such as the average value curve). Assuming there are m types of seismic attributes, for each statistical radius, each seismic attribute has n types of curve parameters, then for each statistical radius there are m×n seismic attribute curves.

[0067] Furthermore, in one embodiment, selecting at least one target seismic attribute from the multiple seismic attributes according to the correlation coefficient and / or similarity may include the following steps:

[0068] For each of the statistical radii, calculating an average value of a preset number of correlation coefficients with the highest values, and / or calculating an average value of a preset number of similarities with the highest values;

[0069] Determine the statistical radius corresponding to the highest average value among the average values ​​of the correlation coefficients as the target statistical radius, or determine the statistical radius corresponding to the highest average value among the average values ​​of the similarities as the target statistical radius;

[0070] Selecting at least one second target correlation coefficient greater than a second preset coefficient value from the correlation coefficients corresponding to the target statistical radius, and / or selecting at least one second target similarity greater than a second preset similarity value from the similarities corresponding to the target statistical radius;

[0071] The seismic attribute corresponding to the at least one second target correlation coefficient and / or the seismic attribute corresponding to the at least one second target similarity is determined as the target seismic attribute.

[0072] Specifically, after determining the seismic attributes for each statistical radius, the correlation coefficients corresponding to each statistical radius can be sorted from high to low, and the average value of the preset number of correlation coefficients in the first sorting can be calculated, and / or, the similarities corresponding to each statistical radius can be sorted from high to low, and the average value of the preset number of similarities in the first sorting can be calculated, and the target statistical radius corresponding to the highest average value among the average values ​​of the correlation coefficients is selected from all the statistical radii, or the target statistical radius corresponding to the highest average value among the average values ​​of the similarities is selected from all the statistical radii, so that the selected target statistical radius is the statistical radius with the highest correlation with the fracture of the first well.

[0073] In addition, after determining the target statistical radius, the seismic attributes corresponding to at least one second target correlation coefficient greater than the second preset coefficient value and / or the seismic attributes corresponding to at least one second target similarity within the target statistical radius can be determined as target seismic attributes, thereby ensuring the accuracy of the determined target seismic attributes.

[0074] It should be noted that the second preset coefficient value may be the same as or different from the first preset coefficient value, and this is not specifically limited here.

[0075] Of course, it should be noted that when selecting at least one second target correlation coefficient greater than the second preset coefficient value, a specified number of second target correlation coefficients greater than the second preset coefficient value can be selected, or all second target correlation coefficients greater than the second preset coefficient value can be selected, and this is not specifically limited here. Of course, when selecting at least one second target similarity greater than the second preset similarity value, a specific number of second target correlation coefficients greater than the second preset similarity value can also be selected, or all second target correlation coefficients greater than the second preset similarity value can also be selected. For example, as an example, for each statistical radius, the correlation coefficients and / or similarities are sorted from high to low, and the average values ​​of the top 10 correlation coefficients or the top 10 similarities of each statistical radius are compared. The statistical radius with the highest average correlation coefficient or the highest average similarity is selected as the target statistical radius. Then, within this target statistical radius, the top 20 seismic attributes with the top 15 seismic attributes with the top 20 similarities are selected, and these seismic attributes are determined as target seismic attributes. This ensures that the determined target seismic attributes are strongly correlated with the fracture genesis of the first well, ensuring the accuracy of predicting fracture conditions through the target seismic attributes.

[0076] Furthermore, in one embodiment, before determining the target fracture density curve of the second well based on the target seismic attribute curve using a pre-trained seismic fracture prediction model, the following steps are further included:

[0077] The preset neural network model is trained using the sample set to obtain the earthquake crack prediction model;

[0078] The sample set includes multiple seismic attribute curves corresponding to the target seismic attribute in the first well and a fracture density curve of the first well, wherein the fracture density curve of the first well is label data corresponding to the multiple seismic attribute curves.

[0079] Specifically, this embodiment can train the preset neural network model through the sample set to obtain the earthquake crack prediction model. Among them, this embodiment can randomly divide the sample set into a training set, a verification set and a test set, train the neural network model through the training set, verify the neural network model through the verification set, and test the verified model through the test set to obtain the final earthquake crack prediction model, realizing the establishment of a deep learning algorithm model based on the constraints of natural cracks and effectiveness control factors (target earthquake attributes), and completing the prediction of effective small-scale earthquake cracks. Among them, Figure 3 As shown, in Figure 3In the , the correlation coefficient between the actual crack density (label) corresponding to the training set and the predicted crack density of the earthquake crack prediction model is 0.92, the correlation coefficient between the actual crack density (label) corresponding to the validation set and the predicted crack density of the earthquake crack prediction model is 0.90, the correlation coefficient between the actual crack density (label) corresponding to the test set and the predicted crack density of the earthquake crack prediction model is 0.83, and the correlation coefficient between the actual crack density (label) corresponding to all data in the sample set and the predicted crack density of the earthquake crack prediction model is 0.89. In this way, the accuracy of the earthquake crack prediction model is guaranteed by iteratively training the neural network model with reference to the above correlation coefficients. In addition, Figure 4 As shown, by comparing the well prediction results with the imaging logging of the first well, it can be seen that the fracture density trend on the imaging logging is consistent with the prediction results. Through the above verification process, it can be seen that this embodiment can more accurately predict the fracture development at the decimeter level or even the centimeter level.

[0080] Furthermore, in one embodiment, before acquiring a plurality of target seismic attribute curves corresponding to the target seismic attributes and determining the target fracture density curve of the second well based on the target seismic attribute curves using a pre-trained seismic fracture prediction model, the following steps may also be included:

[0081] Obtaining an attribute weight factor corresponding to each target seismic attribute according to the seismic fracture prediction model, wherein the attribute weight factor is a ratio between a change value of a fracture density curve of the first well predicted by the seismic fracture prediction model and a change value of a seismic attribute curve of the first well input into the seismic fracture prediction model;

[0082] In the event that there is an attribute weight factor greater than a preset value, the first preset coefficient value and / or the first preset similarity value is adjusted to redetermine the target seismic attribute; or, in the event that there is an attribute weight factor greater than a preset value, the second preset coefficient value and / or the second preset similarity value is adjusted to redetermine the target seismic attribute.

[0083] Specifically, in the process of training the earthquake crack prediction model, since changes in the model input will lead to changes in the model output, the earthquake crack prediction model can obtain the attribute weight factor corresponding to each target earthquake attribute. The attribute weight factor is the ratio between the change value of the crack density curve output by the earthquake crack prediction model and the change value of the seismic attribute curve, that is, the attribute weight factor is used to characterize the impact of changes in input parameters on changes in crack density.

[0084] This embodiment considers the attribute weight factor as an important parameter for model optimization. The higher the attribute weight factor, the greater the impact of the genesis of the target seismic attribute on fracture formation. If necessary, the rationality of the result can be analyzed in conjunction with the geomechanical context. If not, it is necessary to re-enter the target seismic attribute determination step to improve the accuracy of the determined target seismic attribute.

[0085] Specifically, when the target seismic attribute is determined by a first preset coefficient value and / or a first preset similarity value, that is, when at least one first target correlation coefficient greater than the first preset coefficient value is selected from the correlation coefficients, and / or at least one first target similarity greater than the first preset similarity value is selected from the similarities, then in the presence of an attribute weight factor greater than the preset value, the first preset coefficient value and / or the first preset similarity value can be adjusted to re-determine the target seismic attribute.

[0086] Of course, when the target seismic attribute is determined by the second preset coefficient value and / or the second preset similarity value, the second preset coefficient value and / or the second preset similarity value can be adjusted to redetermine the target seismic attribute if there is an attribute weight factor greater than the preset value.

[0087] Of course, it should also be noted that the weight factor of a certain attribute is too high. For example, the weight factors of other attributes are all in the zero tenths, while the weight factor of a certain attribute is several hundred. This abnormal order of magnitude of a single attribute is considered to be caused by overfitting. It can also be considered that the parameters of the model need to be modified. At this time, the parameters of the trained earthquake crack prediction model can be modified according to the actual situation.

[0088] In addition, in one embodiment, this embodiment can also characterize the fracture corresponding to the second well to obtain a fracture result;

[0089] After determining the target fracture density curve of the second well according to the target seismic attribute curve using a pre-trained seismic fracture prediction model, the method further includes: determining the fracture development result of the second well according to the fracture result and the target fracture density curve.

[0090] Specifically, when characterizing the fault corresponding to the second well, it can be characterized by using ant bodies, coherence bodies, etc.

[0091] In addition, when determining the fracture development result of the second well based on the fracture result and the target fracture density curve, the fracture result and the target fracture density curve may be combined to obtain the final fracture development result.

[0092] Figure 5 A crack prediction device provided by one embodiment of the present invention includes:

[0093] An acquisition module 501 is configured to acquire a fracture density curve of a first well and a plurality of seismic attribute curves corresponding to a plurality of seismic attributes associated with the first well;

[0094] A determination module 502 is configured to determine a correlation coefficient and / or similarity between each seismic attribute curve and the fracture density curve;

[0095] A selection module 503 is configured to select at least one target seismic attribute from the multiple seismic attributes based on the correlation coefficient and / or similarity;

[0096] The prediction module 504 is used to obtain multiple target seismic attribute curves corresponding to the target seismic attributes for the second well to be predicted, and determine the target fracture density curve of the second well based on the target seismic attribute curves using a pre-trained seismic fracture prediction model.

[0097] The crack prediction device provided in the embodiment of the present application can achieve Figure 1 To avoid repetition, the various processes implemented in the method embodiment are not described here.

[0098] It should be noted that the embodiments of the crack prediction device in this specification and the embodiments of the crack prediction method in this specification are based on the same inventive concept. Therefore, the specific implementation of the embodiments of the crack prediction device can refer to the implementation of the corresponding embodiments of the crack prediction method, and the repeated parts will not be repeated.

[0099] The crack prediction device in the embodiments of the present application can be a device, or a component, integrated circuit, or chip in a terminal. The device can be a mobile electronic device or a non-mobile electronic device. For example, the mobile electronic device can be a mobile phone, a tablet computer, a laptop computer, a PDA, an in-vehicle electronic device, a wearable device, an ultra-mobile personal computer (UMPC), a netbook, or a personal digital assistant (PDA), etc. The non-mobile electronic device can be a server, a network attached storage (NAS), a personal computer (PC), a television (TV), a teller machine, or a self-service machine, etc., and the embodiments of the present application do not specifically limit this.

[0100] The crack prediction device in the embodiment of the present application may be a device having an operating system. The operating system may be an Android operating system, an iOS operating system, or other possible operating systems, which are not specifically limited in the embodiment of the present application.

[0101] Based on the same technical concept, an embodiment of the present application further provides an electronic device for executing the above-mentioned crack prediction method. Figure 6 The following is a schematic diagram of the structure of an electronic device for implementing various embodiments of the present application. Electronic devices may vary significantly due to different configurations or performances, and may include a processor 610, a communications interface 620, a memory 630, and a communication bus 640. The processor 610, the communications interface 620, and the memory 630 communicate with each other via the communication bus 640. The processor 610 may call a computer program stored in the memory 630 and executable on the processor 610 to perform the following steps:

[0102] obtaining a fracture density curve of a first well and a plurality of seismic attribute curves corresponding to a plurality of seismic attributes associated with the first well;

[0103] determining a correlation coefficient and / or similarity between each seismic attribute curve and the fracture density curve;

[0104] selecting at least one target seismic attribute from the plurality of seismic attributes according to the correlation coefficient and / or similarity;

[0105] For the second well to be predicted, multiple target seismic attribute curves corresponding to the target seismic attributes are obtained, and based on the target seismic attribute curves, a target fracture density curve of the second well is determined using a pre-trained seismic fracture prediction model.

[0106] In one embodiment, obtaining the fracture density curve of the first well includes: obtaining fracture interpretation results of imaging logging of the first well; converting the fracture interpretation results into a logging curve according to seismic resolution; and smoothing the logging curve to obtain the fracture density curve.

[0107] In one embodiment, obtaining multiple seismic attribute curves corresponding to multiple seismic attributes related to the first well includes: determining at least one curve parameter corresponding to each seismic attribute among the multiple seismic attributes for each statistical radius in a plurality of pre-set statistical radii; and determining a seismic attribute curve corresponding to the seismic attribute based on the curve parameter, wherein each curve parameter corresponds to a seismic attribute curve.

[0108] In one embodiment, selecting at least one target seismic attribute from the multiple seismic attributes based on the correlation coefficient and / or similarity includes: selecting at least one first target correlation coefficient greater than a first preset coefficient value from the correlation coefficients, and / or selecting at least one first target similarity greater than a first preset similarity value from the similarities; and determining the seismic attribute corresponding to the at least one first target correlation coefficient and / or the seismic attribute corresponding to the at least one first target similarity as the target seismic attribute.

[0109] In one embodiment, selecting at least one target seismic attribute from the multiple seismic attributes according to the correlation coefficient and / or similarity includes:

[0110] For each of the statistical radiuses, the average value of the preset number of correlation coefficients with the highest values ​​is calculated, and / or the average value of the preset number of similarities with the highest values ​​is calculated; the statistical radius corresponding to the highest average value among the average values ​​of the correlation coefficients is determined as the target statistical radius, or the statistical radius corresponding to the highest average value among the average values ​​of the similarities is determined as the target statistical radius; at least one second target correlation coefficient greater than a second preset coefficient value is selected from the correlation coefficients corresponding to the target statistical radius, and / or at least one second target similarity greater than a second preset similarity value is selected from the similarities corresponding to the target statistical radius; the seismic attribute corresponding to the at least one second target correlation coefficient and / or the seismic attribute corresponding to the at least one second target similarity is determined as the target seismic attribute.

[0111] In one embodiment, before determining the target fracture density curve of the second well based on the target seismic attribute curve through a pre-trained seismic fracture prediction model, the method further includes: training a preset neural network model through a sample set to obtain the seismic fracture prediction model; wherein the sample set includes multiple seismic attribute curves corresponding to the target seismic attribute in the first well and a fracture density curve of the first well, wherein the fracture density curve of the first well is label data corresponding to the multiple seismic attribute curves.

[0112] In one embodiment, before obtaining multiple target seismic attribute curves corresponding to the target seismic attribute and determining the target fracture density curve of the second well according to the target seismic attribute curve through a pre-trained seismic fracture prediction model, the method further includes: obtaining an attribute weight factor corresponding to each of the target seismic attributes according to the seismic fracture prediction model, wherein the attribute weight factor is a ratio between a change value of the fracture density curve of the first well predicted by the seismic fracture prediction model and a change value of the seismic attribute curve of the first well input into the seismic fracture prediction model; in the case where there is an attribute weight factor greater than a preset value, adjusting a first preset coefficient value and / or a first preset similarity value to redetermine the target seismic attribute; or, in the case where there is an attribute weight factor greater than a preset value, adjusting a second preset coefficient value and / or a second preset similarity value to redetermine the target seismic attribute.

[0113] In one embodiment, the method further includes: characterizing the fracture corresponding to the second well to obtain a fracture result;

[0114] After determining the target fracture density curve of the second well according to the target seismic attribute curve using a pre-trained seismic fracture prediction model, the method further includes: determining the fracture development result of the second well according to the fracture result and the target fracture density curve.

[0115] The specific execution steps can refer to the various steps of the above crack prediction method embodiment, and can achieve the same technical effect. To avoid repetition, they will not be described here.

[0116] It should be noted that the electronic devices in the embodiments of the present application include: servers, terminals, or other devices other than terminals.

[0117] The above electronic device structure does not constitute a limitation of the electronic device. The electronic device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently. For example, the input unit may include a graphics processing unit (GPU) and a microphone, and the display unit may be configured in the form of a liquid crystal display, an organic light emitting diode, etc. to configure the display panel. The user input unit includes a touch panel and at least one of other input devices. The touch panel is also called a touch screen. Other input devices may include but are not limited to a physical keyboard, function keys (such as volume control buttons, switch buttons, etc.), a trackball, a mouse, and a joystick, which will not be repeated here.

[0118] The memory can be used to store software programs and various data. The memory may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first storage area may store an operating system, applications or instructions required for at least one function (such as a sound playback function, an image playback function, etc.), etc. In addition, the memory may include a volatile memory or a non-volatile memory, or the memory may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. Volatile memory can be random access memory (RAM), static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM) and direct rambus random access memory (DRRAM).

[0119] The processor may include one or more processing units; optionally, the processor may integrate an application processor and a modem processor, wherein the application processor primarily handles operations related to the operating system, user interface, and application programs, and the modem processor primarily processes wireless communication signals, such as a baseband processor. It is understood that the modem processor may not be integrated into the processor.

[0120] An embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the various processes of the above-mentioned crack prediction method embodiment are implemented and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0121] The processor is the processor in the electronic device described in the above embodiment. The readable storage medium includes a computer-readable storage medium, such as a computer read-only memory (ROM), random access memory (RAM), a magnetic disk, or an optical disk.

[0122] An embodiment of the present application further provides a chip, which includes a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the various processes of the above-mentioned method embodiment and achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0123] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0124] It should be noted that, in this article, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the statement "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device comprising the element. In addition, it should be noted that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in the opposite order according to the functions involved. For example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0125] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0126] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of this application, ordinary technicians in this field can also make many forms without departing from the purpose of this application and the scope of protection of the claims, all of which are within the protection of this application.

Claims

1. A crack prediction method, characterized in that: include: obtaining a fracture density curve of a first well and a plurality of seismic attribute curves corresponding to a plurality of seismic attributes associated with the first well; determining a correlation coefficient and / or similarity between each seismic attribute curve and the fracture density curve; selecting at least one target seismic attribute from the plurality of seismic attributes according to the correlation coefficient and / or similarity; For a second well to be predicted, a plurality of target seismic attribute curves corresponding to the target seismic attributes are obtained, and a target fracture density curve of the second well is determined based on the target seismic attribute curves using a pre-trained seismic fracture prediction model; The selecting at least one target seismic attribute from the multiple seismic attributes according to the correlation coefficient and / or similarity includes: Selecting at least one first target correlation coefficient greater than a first preset coefficient value from the correlation coefficients, and / or selecting at least one first target similarity greater than a first preset similarity value from the similarities; The seismic attribute corresponding to the at least one first target correlation coefficient and / or the seismic attribute corresponding to the at least one first target similarity is determined as the target seismic attribute.

2. The crack prediction method according to claim 1, characterized in that: The obtaining of the fracture density curve of the first well includes: Obtaining fracture interpretation results from imaging logging of the first well; converting the fracture interpretation results into well logging curves according to seismic resolution; The logging curve is smoothed to obtain the fracture density curve.

3. The crack prediction method according to claim 1, characterized in that: Acquiring a plurality of seismic attribute curves corresponding to a plurality of seismic attributes associated with the first well, including: For each of the plurality of pre-set statistical radii, determining at least one curve parameter corresponding to each of the plurality of seismic attributes; A seismic attribute curve corresponding to the seismic attribute is determined according to the curve parameters, wherein each of the curve parameters corresponds to a seismic attribute curve.

4. The crack prediction method according to claim 3, characterized in that: The selecting at least one target seismic attribute from the multiple seismic attributes according to the correlation coefficient and / or similarity includes: For each of the statistical radii, calculating an average value of a preset number of correlation coefficients with the highest values, and / or calculating an average value of a preset number of similarities with the highest values; Determine the statistical radius corresponding to the highest average value among the average values ​​of the correlation coefficients as the target statistical radius, or determine the statistical radius corresponding to the highest average value among the average values ​​of the similarities as the target statistical radius; Selecting at least one second target correlation coefficient greater than a second preset coefficient value from the correlation coefficients corresponding to the target statistical radius, and / or selecting at least one second target similarity greater than a second preset similarity value from the similarities corresponding to the target statistical radius; The seismic attribute corresponding to the at least one second target correlation coefficient and / or the seismic attribute corresponding to the at least one second target similarity is determined as the target seismic attribute.

5. The crack prediction method according to claim 1, characterized in that: Before determining the target fracture density curve of the second well according to the target seismic attribute curve using a pre-trained seismic fracture prediction model, the method further includes: The preset neural network model is trained using the sample set to obtain the earthquake crack prediction model; The sample set includes multiple seismic attribute curves corresponding to the target seismic attribute in the first well and a fracture density curve of the first well, wherein the fracture density curve of the first well is label data corresponding to the multiple seismic attribute curves.

6. The crack prediction method according to claim 1 or 4, characterized in that: Before acquiring a plurality of target seismic attribute curves corresponding to the target seismic attributes and determining the target fracture density curve of the second well based on the target seismic attribute curves using a pre-trained seismic fracture prediction model, the method further includes: Obtaining an attribute weight factor corresponding to each target seismic attribute according to the seismic fracture prediction model, wherein the attribute weight factor is a ratio between a change value of a fracture density curve of the first well predicted by the seismic fracture prediction model and a change value of a seismic attribute curve of the first well input into the seismic fracture prediction model; In the case where there is an attribute weight factor greater than a preset value, adjusting the first preset coefficient value and / or the first preset similarity value to redetermine the target earthquake attribute; or In the case that there is an attribute weight factor greater than a preset value, the second preset coefficient value and / or the second preset similarity value is adjusted to re-determine the target seismic attribute.

7. The crack prediction method according to claim 1, characterized in that: Also includes: Characterizing the fracture corresponding to the second well to obtain a fracture result; After determining the target fracture density curve of the second well according to the target seismic attribute curve using a pre-trained seismic fracture prediction model, the method further includes: The fracture development result of the second well is determined according to the fracture result and the target fracture density curve.

8. An electronic device, characterized in that: The method comprises a processor, a memory and a program or instruction stored in the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements the steps of the crack prediction method according to any one of claims 1 to 7.

9. A readable storage medium, characterized in that The readable storage medium stores a program or instruction, and when the program or instruction is executed by a processor, the steps of the crack prediction method according to any one of claims 1 to 7 are implemented.

Images