Perforation effect evaluation method and system based on artificial intelligence

Abstract

The invention belongs to the technical field of unconventional oil and gas exploitation, and particularly relates to a perforation effect evaluation method and system based on artificial intelligence. The method comprises the steps that influence factors influencing the perforation effect and the yield are determined; obtaining a target prediction model based on the influence factors; and judging the perforation effect by using the target prediction model. According to the method, perforation effect evaluation is carried out by means of fracturing data, geological data and field monitoring data, so that the method is more persuasive, and a reasonable range, beneficial to transformation and yield increase, of each influence factor can be found; according to the method, a random forest algorithm in machine learning is used, a model composed of perforation parameters, geological data and monitoring data is used for yield prediction, all data are from historical data, and manual intervention is reduced; according to the method, important reservoir transformation and yield influence factors can be identified by adopting the two-stage dimension reduction strategy and the random forest algorithm, and yield prediction can be well carried out.

Description

Technical Field

[0001] The present invention belongs to the technical field of unconventional oil and gas exploitation, and in particular relates to a perforation effect evaluation method and system based on artificial intelligence. Background Art

[0002] With the large-scale investment in fracturing development of unconventional reservoirs in my country and the gratifying results achieved, unconventional resources as oil and gas resources have gradually become the main force to solve my country's "gas shortage" problem. Due to the heterogeneity of unconventional oil and gas reservoirs, the production differences between different sections of horizontal wells are large. Statistics from several major shale gas blocks in the United States show that about 1 / 3 of the perforations of fracturing gas wells have no production, and 60% of the total gas production of a single well comes from 40% of the fracturing sections. Conventional analysis methods show that the dependence of various construction parameters (perforation parameters, fracturing parameters) and geological parameters on the contribution of single-section gas production is not obvious. With the development of the shale gas development industry from extensive to intensive and refined, it has become a key process for the efficient exploitation of unconventional reservoirs to accurately evaluate the effect of perforation on reservoir transformation with the help of multiple factors and realize the perforation parameter optimization plan of "one method for one section, one policy for one well".

[0003] There are two meanings of perforation effect evaluation:

[0004] 1) The purpose of the evaluation conducted during the implementation of the overall plan is not only to evaluate the impact of perforation effect on reservoir transformation, but also to verify the degree of conformity between the design and the actual situation, especially the rationality of the basic data involved in the design, so as to improve the design and guide subsequent work;

[0005] 2) Overall evaluation after implementation, the purpose of which is to evaluate the perforation scheme (number of perforation sections, number of perforation clusters, perforation location, perforation density, perforation diameter, and perforation depth) and improve the design scheme from the top. A large number of studies have been conducted on reservoir transformation effect evaluation methods at home and abroad, but there are not many studies on the post-evaluation of unconventional reservoir perforation effects.

[0006] The inventors found that the existing conventional reservoirs mainly evaluate the pros and cons of perforation completion schemes by productivity; the evaluation of perforation effects in unconventional reservoirs mainly includes two categories:

[0007] 1) On-site monitoring method: mainly through indirect evaluation through the initiation rate of microseismic interpretation and the test results of the fluid production profile, and the visual logging system - VideoLog system, through video images, intuitively understand the casing condition, perforation morphology, and qualitatively analyze the perforation output; quantitatively analyze the perforation erosion and proppant distribution.

[0008] 2) Theoretical model method: By establishing a fracture propagation model, the effects of different perforation parameters on fracture initiation rate and SRV volume are simulated and analyzed to evaluate the effect. Although numerical simulation methods are becoming more and more sophisticated, and the accuracy of production prediction can be pursued by establishing increasingly refined geological models and describing complex dynamic models, they are mainly used for evaluation in the early stage of perforation scheme optimization design, and the calculation is complex and the workload is large. Summary of the invention

[0009] In view of the above problems, the present invention provides a perforation effect evaluation method based on artificial intelligence, the method comprising:

[0010] Determine the factors that affect perforation performance and production;

[0011] Obtain a target prediction model based on influencing factors;

[0012] The target prediction model is used to judge the perforation effect.

[0013] Preferably, the determining of factors affecting the perforation effect and production includes:

[0014] Acquire an original sample set, wherein the original sample set includes multiple influencing factors;

[0015] Screen the influencing factors in the original sample set to obtain an intermediate sample set;

[0016] The intermediate sample set is classified and data dimension reduction is performed to obtain a final sample set, and the elements in the final sample set are the influencing factors affecting the perforation effect and production.

[0017] Preferably, obtaining the target prediction model based on the influencing factors includes:

[0018] Divide the elements in the final sample set into training set and test set;

[0019] Construct an initial prediction model and determine the model parameters of the prediction model;

[0020] The initial prediction model is trained using the training set, and the model parameters are adjusted to obtain the target training model.

[0021] Preferably, the judging of the perforation effect by using the target prediction model includes:

[0022] According to the target prediction model, the partial dependence relationship between influencing factors and output is constructed;

[0023] The gain of production by influencing factors is analyzed based on partial dependence, and the perforation effect is judged based on the analysis results.

[0024] Preferably, the method further comprises: predicting the output using a target prediction model.

[0025] Preferably, the method of predicting the output by using the target prediction model comprises:

[0026] The target prediction model is used to obtain the predicted production corresponding to each fracturing stage in the test set;

[0027] The root mean square error is used to determine the error degree of predicted output.

[0028] The present invention also proposes a perforation effect evaluation system based on artificial intelligence, the system comprising:

[0029] Determination module, used to determine the factors affecting perforation effect and production;

[0030] A building module is used to obtain a target prediction model based on influencing factors;

[0031] The judgment module is used to judge the perforation effect by using the target prediction model.

[0032] Preferably, the determination module is used to determine the factors affecting the perforation effect and production, including:

[0033] The determination module is used to obtain an original sample set, wherein the original sample set includes multiple influencing factors;

[0034] Screening the influencing factors in the original sample set to obtain an intermediate sample set;

[0035] The intermediate sample set is classified and data dimension reduction is performed to obtain a final sample set, and the elements in the final sample set are the influencing factors affecting the perforation effect and production.

[0036] Preferably, the construction module is used to obtain a target prediction model based on influencing factors, including:

[0037] The building module is used to divide the elements in the final sample set into training set and test set;

[0038] Construct an initial prediction model and determine the model parameters of the prediction model;

[0039] The initial prediction model is trained using the training set, and the model parameters are adjusted to obtain the target training model.

[0040] Preferably, the judgment module is used to judge the perforation effect using the target prediction model, including:

[0041] The judgment module is used to construct the partial dependency relationship between influencing factors and output according to the target prediction model;

[0042] The gain of production by influencing factors is analyzed based on partial dependence, and the perforation effect is judged based on the analysis results.

[0043] Preferably, the system further comprises:

[0044] The prediction module is used to predict the output using the target prediction model.

[0045] Preferably, the prediction module is used to predict the output using the target prediction model, including:

[0046] The prediction module is used to obtain the predicted production corresponding to each fracturing stage in the test set using the target prediction model;

[0047] The root mean square error is used to determine the error degree of predicted output.

[0048] The present invention also provides an electronic device, comprising:

[0049] Processor and memory;

[0050] The processor calls the computer program stored in the memory to execute any one of the above-mentioned perforation effect evaluation methods based on artificial intelligence.

[0051] The present invention also provides a computer-readable storage medium.

[0052] The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor is enabled to execute any one of the above-mentioned artificial intelligence-based perforation effect evaluation methods.

[0053] The present invention has the following beneficial effects:

[0054] (1) The present invention uses fracturing data, geological data and field monitoring data to evaluate the perforation effect, which is more convincing and can find the reasonable range of each influencing factor that is beneficial to transformation and production increase;

[0055] (2) The present invention uses the random forest algorithm in machine learning and a model composed of perforation parameters, geological data, and monitoring data to predict production. All data are from historical data, which reduces human intervention.

[0056] (3) The present invention adopts a two-level dimensionality reduction strategy and a random forest algorithm to identify important reservoir transformation and production influencing factors, and can also perform good production prediction.

[0057] Other features and advantages of the present invention will be described in the following description, and partly become obvious from the description, or be understood by implementing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures pointed out in the description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0059] Figure 1 A diagram showing a method for evaluating perforation effects based on artificial intelligence in an embodiment of the present invention is shown;

[0060] Figure 2 A dependency curve diagram showing the influence of the number of layers on the perforation-fracturing effect through a dependency analysis based on random forests in an embodiment of the present invention is shown;

[0061] Figure 3 A dependency curve diagram showing the influence of the number of clusters on the perforation-fracturing effect through a dependency analysis based on random forests in an embodiment of the present invention is shown;

[0062] Figure 4 A dependency curve diagram showing the influence of cluster spacing on perforation-fracturing effect through random forest-based dependency analysis in an embodiment of the present invention is shown;

[0063] Figure 5 A dependency curve diagram showing the influence of hole depth on perforation-fracturing effect through random forest-based dependency analysis in an embodiment of the present invention is shown;

[0064] Figure 6 A dependency curve diagram showing the influence of aperture on perforation-fracturing effect through random forest-based dependency analysis in an embodiment of the present invention is shown;

[0065] Figure 7 A dependency curve diagram showing the influence of pore density on perforation-fracturing effect through dependency analysis based on random forest in an embodiment of the present invention is shown;

[0066] Figure 8 A dot graph showing a detailed comparison of the relationship between the measured yield value and the predicted yield value through a dependency analysis based on random forests in an embodiment of the present invention;

[0067] Fig. 9 A diagram of a perforation effect evaluation system based on artificial intelligence in an embodiment of the present invention is shown;

[0068] Fig.10 A diagram of an electronic device in an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0069] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as being limited to the examples set forth herein; on the contrary, these embodiments are provided so that the present disclosure will be more comprehensive and complete, and the concepts of the example embodiments are fully conveyed to those skilled in the art. The described features, structures, or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced while omitting one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known technical solutions are not shown or described in detail to avoid obscuring various aspects of the present disclosure.

[0070] In addition, the accompanying drawings are only schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and thus their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or implemented in one or more hardware units or integrated circuits, or implemented in different networks and / or processor devices and / or microcontroller devices.

[0071] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the steps. For example, some steps may be decomposed, while some steps may be combined or partially combined, so the actual execution order may change according to the actual situation.

[0072] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the terms used in this way can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein, for example.

[0073] In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or sub-modules is not necessarily limited to those steps or sub-modules explicitly listed, but may include other steps or sub-modules not explicitly listed or inherent to these processes, methods, products, or apparatuses.

[0074] like Figure 1 As shown, the present invention proposes a perforation effect evaluation method based on artificial intelligence, and the method comprises the following steps:

[0075] S1 Determine the factors affecting perforation effect and production;

[0076] S2 obtains the target prediction model based on the influencing factors;

[0077] S3 uses the target prediction model to judge the perforation effect.

[0078] Specifically, S1 includes the following steps:

[0079] Acquire an original sample set, wherein the original sample set includes multiple influencing factors;

[0080] Screen the influencing factors in the original sample set to obtain an intermediate sample set;

[0081] The intermediate sample set is classified and data dimension reduction is performed to obtain a final sample set, and the elements in the final sample set are the influencing factors affecting the perforation effect and production.

[0082] Specifically, S2 includes the following steps:

[0083] Divide the elements in the final sample set into training set and test set;

[0084] Construct an initial prediction model and determine the model parameters of the prediction model;

[0085] The initial prediction model is trained using the training set, and the model parameters are adjusted to obtain the target training model.

[0086] Specifically, S3 includes the following steps:

[0087] According to the target prediction model, the partial dependence relationship between influencing factors and output is constructed;

[0088] The gain of production by influencing factors is analyzed based on partial dependence, and the perforation effect is judged based on the analysis results.

[0089] The method also includes: S4 predicting the output using the target prediction model.

[0090] Specifically, S4 includes the following steps:

[0091] The target prediction model is used to obtain the predicted production corresponding to each fracturing stage in the test set;

[0092] The root mean square error is used to determine the error degree of predicted output.

[0093] Specifically, the present invention comprises the following steps:

[0094] 1) Determine the main factors affecting perforation effect and production:

[0095] a. Level 1 dimensionality reduction - Pearson correlation coefficient:

[0096] (1) First, assume that the original sample set is represented by a matrix A of size m×(d+1), expressed as:

[0097]

[0098] Each row corresponds to a fracturing stage, and the last element is the corresponding fracturing stage production, the average fracturing pressure of the stage, the fracturing rate of the stage, and the SRV of the stage transformation (one of them, the stage production is taken as an example below). The first d elements of each row correspond to d influencing factors of the fracturing stage. The d influencing factors are represented by a set as D = {a1, a2, ..., a d};

[0099] (2) Then use the Pearson correlation coefficient to screen the d influencing factors in the original sample set table, remove the influencing factors with a correlation greater than 0.9, and obtain e influencing factors. The total sample set is obtained and expressed by matrix B as:

[0100]

[0101] Each row corresponds to a fracturing stage, the last element is the production of the corresponding fracturing stage, the first e elements of each row correspond to the e influencing factors of the fracturing stage, and the e influencing factors are represented by a set as D*={a1,a2,…,a e};

[0102] b. Second level dimensionality reduction - recursive feature elimination based on support vector machine:

[0103] (1) First, the yield in the total sample is classified: according to different standards, the yield is divided into two categories: high yield and low yield. The classification is as follows:

[0104]

[0105] Among them, Y n is the category of segment yield; 1 represents high yield, i.e., high yield is greater than 1.3 times the average yield, and -1 represents low yield, i.e., low yield is less than 1.3 times the average yield; y represents the set of segment yields y=(y1,y2,…,y m ); avg(y) represents the average value of segment output y;

[0106] (2) Recursive feature elimination based on support vector machine is used to reduce data dimension, and the influencing factor with the smallest ranking criterion is removed from the e influencing factors. After multiple iterations, f main influencing factors affecting production are obtained, thereby obtaining the final sample;

[0107] Among them, the f main influencing factors are represented by a set as D**={a1,a2,…,a f}, the final sample set is represented by matrix C as;

[0108]

[0109] c. Create training and test sets

[0110] The final sample set is divided into a training set Te containing m1 samples m1×f and the test set Te containing m2 samples m2×f Among them, m1+m2=m;

[0111] 2) Build a random forest model

[0112] According to the training set Te m1×f The data in is used to construct a single decision regression tree; the results of multiple decision regression trees are averaged to obtain the corresponding random forest algorithm results; thus, a random forest model is constructed;

[0113] 3) Perforation effect evaluation and production prediction based on random forest algorithm

[0114] According to the training set Te obtained in step 1) m1×f The random forest model constructed in step 2) is used to complete the perforation effect evaluation and production prediction;

[0115] a. Perforation effect evaluation

[0116] Based on the above random forest model, the partial dependence relationship between the above f influencing factors and yield is constructed:

[0117] ①The linear relationship between the influencing factors and output in the random forest algorithm model is as follows;

[0118]

[0119] Among them, x im2 It represents the value corresponding to the i-th sample and the m2-th influencing factor of the data set. is the corresponding prediction;

[0120] ② The partial dependence is obtained by calculating the average value of the following formula and plotting it within the valid range of x:

[0121]

[0122] Among them, med(y) is the median value of output y, Display: After averaging other factors, the influencing factor k has an impact on the model prediction value The influence of x im2 Indicates the value corresponding to the i-th sample and the m2-th influencing factor of the data set;

[0123] ③ Analyze the gain of the influencing factors on the output based on the above partial dependence relationship;

[0124] When the partial dependence corresponding to the influencing factor is greater than 0, the value of the influencing factor in this interval is conducive to high production; and the greater the partial dependence, the more conducive to high production, and the better the perforation effect (in the construction plan, the interval with the partial dependence greater than 0 is used to optimize the construction);

[0125] When the partial dependence corresponding to the influencing factor is less than or equal to 0, the value of the influencing factor in this interval is not conducive to high production; and the smaller the partial dependence, the more unfavorable it is for high production, and the worse the perforation effect (in the construction plan, the construction in the interval where the partial dependence is less than 0 should be avoided);

[0126] b. Establishing yield prediction based on random forest

[0127] According to the data of the test set Tem2×f obtained in step c of step 1) and the random forest model in step 2), the predicted production corresponding to each fracturing stage in the test set is obtained, and the prediction effect is judged according to the root mean square error:

[0128]

[0129] Where: y i ,y pre,i are the actual output and predicted output corresponding to sample i respectively.

[0130] Furthermore, in step 1) in sub-step a, the Pearson correlation coefficient is:

[0131]

[0132] Among them, Corr ij Influencing factor a i and influencing factors j The correlation coefficient between ni Influencing factor a i The corresponding sample value, Indicates attribute value a i The corresponding sample mean, Indicates attribute value a i The standard deviation of the corresponding sample, m is the total number of samples.

[0133] Furthermore, in step 2), the steps of constructing a single decision regression tree are as follows:

[0134] Step 1: From the training set Te m1×f Randomly extract data set T from

[0135] Step 2: Create node N;

[0136] Step 3: If all nodes N belong to the same category, the value of label N is the average value of the output in T#. End the process;

[0137] Step 4: Select the influencing factors from the influencing factor set D** as candidate split attributes

[0138] Step 5: For Calculate the squared error for each possible partition of each influencing factor in and determine the binary partition;

[0139] Step 6: Based on the binary partition determined in step 5, divide T into two parts and

[0140] Step 7: Label the value as T′ j The mean of the output in the set If the number of samples is less than 10, the process ends.

[0141] Step 8: Replace the average fracture initiation pressure of the next segment, the fracture initiation rate of the segment, and the SRV of the segment transformation, and repeat the above steps.

[0142] Example

[0143] The original sample set A is constructed based on the perforation construction data of 168 fracturing stages in a shale gas field in eastern Sichuan, as well as microseismic and production fluid profile interpretation data. The 11 fracturing construction factors used are shown in Table 1:

[0144] Table 1

[0145]

[0146]

[0147] 1) Determine the main factors affecting perforation effect and production

[0148] a. Level 1 dimensionality reduction - Pearson correlation coefficient

[0149] The Pearson correlation coefficients between the 14 influencing factors are calculated. Since the correlation coefficients between these 14 influencing factors are low, all below 0.7, these 14 influencing factors enter the subsequent second-level dimensionality reduction. At this time, the total sample set B is the same as the original sample set A.

[0150] b. Second level dimensionality reduction: Recursive feature elimination based on support vector machine

[0151] The results of the recursive feature elimination method based on the support vector machine: the cross-validation score of the 10 influencing factors is the highest. Therefore, the 10 influencing factors are selected as the main influencing factors, and together with the segment production, they constitute the final sample set C for the subsequent construction of the random forest model and the evaluation of the perforation effect. The main influencing factors are shown in Table 2.

[0152] Table 2

[0153]

[0154]

[0155] c. Create training and test sets

[0156] Based on the above final sample set C, the final sample set C is randomly divided into training sets Te m1×f (137 fracturing stages) and the test set Te m2×f (57 fracturing stages).

[0157] 2) Build a random forest model

[0158] Training set Te m1×f Some of the data in are shown in Table 3 below, from which a random forest model can be constructed.

[0159] Table 3

[0160]

[0161]

[0162] 3) Perforation effect evaluation and production prediction based on random forest algorithm.

[0163] a. Perforation effect evaluation

[0164] Through random forest-based dependency analysis, Figure 2-Figure 7 As shown in the figure, the partial dependence of the six main perforation influencing factors on production leads to the following conclusions:

[0165] (1) The impact on the production of the fracturing stage is small. Figure 2 As shown in the figure, when the fracturing layers are 1, 2, and 3, the response degree of the layers to the production of the fracturing stage is low, and when the fracturing layers are 4, 5, and 6, it indicates a slight gain effect on the production of the fracturing stage. Therefore, the subsequent fracturing construction should be selected in layers 4, 5, and 6 as much as possible.

[0166] (2) The effect of cluster number on the production of the fracturing stage is similar to that of the layer position, e.g. Figure 3 As shown, when the number of clusters is greater than 2, there is a slight gain effect on the production of the fracturing stage, which is conducive to high production.

[0167] (3) Cluster spacing is more sensitive to the production of the fracturing stage. Figure 4 As shown in the figure, with the increase of cluster spacing, the unfavorable situation for production gradually improves and then turns into a favorable situation for high production. When the cluster spacing is 30m, the partial dependence reaches the highest, and it can be considered that the cluster spacing at this time has the best perforation effect. When the cluster spacing is greater than 30m, the contribution to the production of the fracturing stage is no longer obvious.

[0168] (4) The perforation density is more sensitive to the section production, such as Figure 7 As shown in the figure, thereafter, as the number of holes / m increases, it shows a positive contribution to the production of the fracturing stage, but the increase is not large. When it is lower than 18 holes / m, the production of the stage increases negatively. Explanation: When the number of holes / m is 18, the perforation effect is the best.

[0169] (5) The sensitivity of perforation depth to segment production is general, such as Figure 5 As shown, when the perforation depth increases from 600 mm, the positive contribution to the fracturing stage production changes slowly, indicating that the perforation depth has little effect on the stage production.

[0170] (6) The effect of perforation diameter on the production of the fracturing stage is relatively sensitive, such as Figure 6 As shown in the figure, the response curve of the fracturing stage production increases with the increase of the aperture. This shows that there is room for further optimization of the perforation aperture, and the perforation aperture should be increased when optimizing the perforation parameters.

[0171] b. Production forecast

[0172] According to the random forest model in 2), the test set Te m2×f By substituting the main influencing factor data (as shown in Table 4) into the model, the corresponding predicted output can be obtained.

[0173] Table 4

[0174]

[0175] like Figure 8 As shown in the figure, the comparison between the predicted yield and the measured value shows that in the logarithmic coordinate axis, the predicted yield and the measured value show good correspondence. The root mean square error obtained by the predicted yield and the measured value is 0.306, which also shows that the prediction effect is good.

[0176] like Fig. 9 As shown, the present invention also proposes a perforation effect evaluation system based on artificial intelligence, the system comprising:

[0177] A determination module 10 is used to determine factors affecting perforation effect and production;

[0178] A construction module 20 is used to obtain a target prediction model based on influencing factors;

[0179] The judgment module 30 is used to judge the perforation effect by using the target prediction model.

[0180] The system comprises: a prediction module 40, which is used to predict the output using a target prediction model.

[0181] like Fig.10 As shown, corresponding to the above-mentioned artificial intelligence-based perforation effect evaluation method, the present invention also provides an electronic device. Since the embodiment of the device is similar to the above-mentioned method embodiment, the description is relatively simple. For relevant parts, please refer to the description of the above-mentioned method embodiment part. The device described below is only schematic. The device may include: a processor (processor) 1, a memory (memory) 2 and a communication bus (i.e., the above-mentioned device bus) and a search engine, wherein the processor 1 and the memory 2 communicate with each other through the communication bus and communicate with the outside through the communication interface. The processor 1 can call the logic instructions in the memory 2 to execute the perforation effect evaluation method based on artificial intelligence.

[0182] In addition, the logic instructions in the above-mentioned memory 2 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: a storage chip, a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a disk or an optical disk, and other media that can store program codes.

[0183] On the other hand, an embodiment of the present invention further provides a processor-readable storage medium, on which a computer program 3 is stored. When the computer program 3 is executed by the processor 1, the artificial intelligence-based perforation effect evaluation method provided in the above embodiments is implemented.

[0184] The processor-readable storage medium can be any available medium or data storage device that can be accessed by the processor 1, including but not limited to magnetic storage (such as floppy disks, hard disks, magnetic tapes, magneto-optical disks (MO), etc.), optical storage (such as CD, DVD, BD, HVD, etc.), and semiconductor storage (such as ROM, EPROM, EEPROM, non-volatile memory (NANDFLASH), solid-state drive (SSD)), etc.

[0185] Those skilled in the art should understand that although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible to modify the technical solutions described in the aforementioned embodiments, or to make equivalent replacements for some of the technical features therein; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A perforation effect evaluation method based on artificial intelligence, characterized in that: The method comprises: Determine the factors that affect perforation performance and production; Obtain a target prediction model based on influencing factors; The target prediction model is used to judge the perforation effect.

2. The perforation effect evaluation method based on artificial intelligence according to claim 1 is characterized in that: The determination of factors affecting the perforation effect and production includes: Acquire an original sample set, wherein the original sample set includes multiple influencing factors; Screen the influencing factors in the original sample set to obtain an intermediate sample set; The intermediate sample set is classified and data dimension reduction is performed to obtain a final sample set, and the elements in the final sample set are the influencing factors affecting the perforation effect and production.

3. The perforation effect evaluation method based on artificial intelligence according to claim 1 is characterized in that: The target prediction model is obtained based on the influencing factors, including: Divide the elements in the final sample set into training set and test set; Construct an initial prediction model and determine the model parameters of the prediction model; The initial prediction model is trained using the training set, and the model parameters are adjusted to obtain the target training model.

4. The perforation effect evaluation method based on artificial intelligence according to claim 1 is characterized in that: The method of using the target prediction model to judge the perforation effect includes: According to the target prediction model, the partial dependence relationship between influencing factors and output is constructed; The gain of production by influencing factors is analyzed based on partial dependence, and the perforation effect is judged based on the analysis results.

5. The perforation effect evaluation method based on artificial intelligence according to claim 1 is characterized in that: The method also includes: predicting the output using the target prediction model.

6. The perforation effect evaluation method based on artificial intelligence according to claim 1 is characterized in that: The method of using the target prediction model to predict the output includes: The target prediction model is used to obtain the predicted production corresponding to each fracturing stage in the test set; The root mean square error is used to determine the error degree of predicted output.

7. A perforation effect evaluation system based on artificial intelligence, characterized in that: The system comprises: Determination module, used to determine the factors affecting perforation effect and production; A building module is used to obtain a target prediction model based on influencing factors; The judgment module is used to judge the perforation effect by using the target prediction model.

8. The artificial intelligence-based perforation effect evaluation system according to claim 7, characterized in that: The determination module is used to determine the factors affecting the perforation effect and production, including: The determination module is used to obtain an original sample set, wherein the original sample set includes multiple influencing factors; Screening the influencing factors in the original sample set to obtain an intermediate sample set; The intermediate sample set is classified and data dimension reduction is performed to obtain a final sample set, and the elements in the final sample set are the influencing factors affecting the perforation effect and production.

9. The artificial intelligence-based perforation effect evaluation system according to claim 7, characterized in that: The building module is used to obtain a target prediction model based on influencing factors, including: The building module is used to divide the elements in the final sample set into training set and test set; Construct an initial prediction model and determine the model parameters of the prediction model; The initial prediction model is trained using the training set, and the model parameters are adjusted to obtain the target training model.

10. The artificial intelligence-based perforation effect evaluation system according to claim 7, characterized in that: The judgment module is used to judge the perforation effect by using the target prediction model, including: The judgment module is used to construct the partial dependency relationship between influencing factors and output according to the target prediction model; The gain of production by influencing factors is analyzed based on partial dependence, and the perforation effect is judged based on the analysis results.

11. The artificial intelligence-based perforation effect evaluation system according to claim 7, characterized in that: The system further comprises: The prediction module is used to predict the output using the target prediction model.

12. The artificial intelligence-based perforation effect evaluation system according to claim 11, characterized in that: The prediction module is used to predict the output using the target prediction model, including: The prediction module is used to obtain the predicted production corresponding to each fracturing stage in the test set using the target prediction model; The root mean square error is used to determine the error degree of predicted output.

13. An electronic device, characterized in that: include: Processor and memory; The processor calls the computer program stored in the memory to execute the artificial intelligence-based perforation effect evaluation method described in any one of claims 1 to 6.

14. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor is enabled to execute the artificial intelligence-based perforation effect evaluation method according to any one of claims 1 to 6.

Images