Shale gas well pressure channeling risk assessment and prediction model and its establishment method and application

Abstract

The present invention discloses a shale gas well pressure channeling risk evaluation and prediction model and its establishment method and application, including the following steps: S1: obtaining evaluation indicators before pressure channeling and evaluation indicators after pressure channeling, establishing data set one and data set two; S2: performing preprocessing to obtain data set three and data set four; S3: performing clustering processing to obtain the number of clusters, which is the number of shale gas well pressure channeling risk classifications; S4: sorting the classifications obtained after clustering, obtaining shale gas well pressure channeling risk classifications, completing the establishment of shale gas well pressure channeling risk evaluation models, and obtaining risk evaluation models before pressure channeling and risk evaluation models after pressure channeling; S5: using the shale gas well pressure channeling risk classifications as labels and each indicator as a feature to divide a training set and a test set; S6: using a classification method to establish a model to obtain a risk prediction model before pressure channeling and a risk prediction model after pressure channeling. The present invention can perform risk evaluation and prediction before and after pressure channeling occurs.

Description

Technical Field

[0001] The present invention relates to the technical field of shale gas exploitation, and in particular to a shale gas well pressure channeling risk assessment and prediction model, and an establishment method and application thereof. Background Art

[0002] With the surge in natural gas consumption, conventional natural gas resources can no longer meet the needs of rapid economic development. Unconventional gas reservoirs have gradually attracted people's attention due to their wide distribution and rich recoverable reserves. Among them, shale gas reservoirs have great development potential. The efficient development of shale gas reservoirs is a necessary and favorable condition for building a low-carbon, clean and safe energy system. Shale gas reservoirs have extremely low permeability. In order to achieve efficient exploitation of shale gas reservoirs, the original shale gas reservoirs need to be transformed. Hydraulic fracturing technology combined with horizontal wells has been recognized as an effective means of increasing production to achieve the above goals. By forming a complex fracture network composed of natural fractures and hydraulic fractures, a high-permeability flow channel is provided for shale gas. Although the complex fracture network generated by large-scale hydraulic fracturing has realized the economical exploitation of shale gas reservoirs, the complex fracture network and small well spacing often cause fracture connectivity or stress interference between adjacent wells, resulting in a sharp decline in shale gas well production.

[0003] The impact of shale gas well pressure channeling on the average production of gas wells exceeds 90%. Shale gas well pressure channeling has become a serious problem in the development of shale gas reservoirs, seriously affecting the effective exploitation of shale gas. After shale gas well pressure channeling occurs, it is often difficult for the gas well to recover the production level before pressure channeling, which seriously affects the efficient development of shale gas reservoirs. Existing research mainly focuses on shale gas well pressure channeling phenomenon, production dynamic characteristics, analytical models or numerical models. The above models have poor generalization and the assumptions are far from the actual shale gas wells, making it difficult to accurately predict the risk of shale gas well pressure channeling. In addition, the main controlling factors for evaluating shale gas well pressure channeling are still unclear, and there is a lack of evaluation and prediction methods for shale gas well pressure channeling risks at the scale of the entire mine. It is difficult to carry out targeted production increase measures for gas wells with strong pressure channeling, or to carry out effective interference prevention measures for gas wells that may be strongly affected by pressure channeling before shale gas well pressure channeling occurs. Summary of the invention

[0004] In view of the above problems, the present invention aims to provide a shale gas well pressure channeling risk assessment and prediction model and its establishment method and application.

[0005] The technical solution of the present invention is as follows:

[0006] On the one hand, a method for establishing a shale gas well pressure channeling risk assessment and prediction model is provided, comprising the following steps:

[0007] S1: obtaining a pre-pressure channeling evaluation index and a post-pressure channeling evaluation index, and establishing a data set 1 and a data set 2 according to the pre-pressure channeling evaluation index and the post-pressure channeling evaluation index respectively;

[0008] S2: preprocessing the data set 1 and the data set 2 respectively to obtain a preprocessed data set 3 and a preprocessed data set 4;

[0009] S3: performing clustering processing on the data set 3 and the data set 4 respectively to obtain the number of clusters, where the number of clusters is the number of shale gas well pressure channeling risk classifications;

[0010] S4: sorting the classifications obtained after clustering, obtaining the risk classification of shale gas well pressure channeling, completing the establishment of a shale gas well pressure channeling risk assessment model, and obtaining a risk assessment model before pressure channeling and a risk assessment model after pressure channeling;

[0011] S5: using the shale gas well pressure channeling risk classification as a label, and using each of the pre-pressure channeling evaluation index and the post-pressure channeling evaluation index as a feature, to divide the training set and the test set;

[0012] S6: Based on the training set and the test set, a model is established using a classification method to obtain a risk prediction model before pressure channeling and a risk prediction model after pressure channeling.

[0013] Preferably, in step S1, when obtaining the pre-pressure channeling evaluation index and the post-pressure channeling evaluation index, it is determined whether pressure channeling occurs in the target well according to any one of the following rules:

[0014] If there is a new well being fracturing within the target well distance threshold, and the daily gas production drops sharply to the daily gas production threshold, it is determined that the target well is fracturing-induced;

[0015] If the increase in the daily water production of the target well exceeds the daily water production increase threshold, it is determined that pressure channeling occurs in the target well;

[0016] If the increase in the casing pressure of the target well exceeds the casing pressure increase threshold, it is determined that pressure channeling occurs in the target well.

[0017] Preferably, in step S1, the evaluation index before pressure channeling includes the daily production of the disturbed well before pressure channeling, the bottom hole pressure of the disturbed well before pressure channeling, the total fracturing fluid volume of the disturbed well, the degree of development of the pressure fractures in the disturbed well, the total fracturing fluid volume of the pressure channeling source well, and the degree of development of the pressure fractures in the pressure channeling source well;

[0018] The post-pressure channeling evaluation index includes the pre-pressure channeling evaluation index and the index after pressure channeling occurs in the shale gas well;

[0019] The indicators after the shale gas well is compressed and channeled include the number of affected days, compression and channeling distance, impact degree, recovery ratio, and the cumulative average daily production of the disturbed wells.

[0020] Preferably, the development degree of the pressure fractures in the disturbed well and the development degree of the pressure fractures in the pressure channeling source well are determined by the following method:

[0021] If microseismic testing is carried out on site, the degree of fracture development can be calculated by the following formula:

[0022]

[0023] Where: K frac is the degree of fracture development, pieces; n is the total number of fracture stages, pieces; i is the fracture stages at all levels, pieces; x i is the number of microseismic time responses corresponding to each level of fracturing stage, ;

[0024] If microseismic testing is not performed on site, the degree of fracture development is calculated using the following formula:

[0025]

[0026]

[0027]

[0028]

[0029] Where: L is the half length of the fracture, m; f(X) is the radius of the micro-unit of the fracture volume, m; is the porosity occupied by the fracture volume micro-unit; dX is the height of the fracture volume micro-unit, m; X is the location of the fracture volume micro-unit, m; m is the perforation aperture, m; It is the average porosity of the fractures in the core of each level of fracturing; is the porosity of the core after fracturing under formation pressure and temperature conditions; is the porosity of the core under formation pressure and temperature conditions; n is the number of cores, pieces.

[0030] Preferably, the degree of influence is calculated by the following formula:

[0031]

[0032] Where: D Y Q is the influence degree of pressure channeling; sq is the daily gas production of the shale gas well before the pressure channeling occurs, 10 4 m 3 ;Q sh is the daily gas production of the shale gas well after pressure channeling occurs, 10 4 m 3 ;

[0033] The recovery ratio is calculated by the following formula:

[0034]

[0035] Where: D H is the recovery ratio of pressure channeling; Q F is the daily gas production of the shale gas well after pressure recovery, 10 4 m 3 ;

[0036] The cumulative average daily production of the disturbed wells is calculated by the following formula:

[0037] Q sz =Q sq ×d y -Q shz (8)

[0038] Where: Q sz is the average daily production of the disturbed wells, 10 4 m 3 ;d y is the number of days affected by pressure channeling, d; Q shz is the cumulative gas production of the gas well after the pressure channeling occurs until the daily gas production of the gas well recovers to the threshold value, 10 4 m 3 .

[0039] Preferably, in step S2, the preprocessing specifically includes the following sub-steps:

[0040] S21: using a box-type method to perform outlier detection on the original data set, deleting outliers in the original data set, and obtaining a first processed data set;

[0041] S22: using the Spearman correlation coefficient to calculate the correlation coefficient between each feature in the first processed data set to obtain a first correlation coefficient result;

[0042] S23: interpolating missing values ​​on the first processed data set to obtain a second processed data set; when interpolating missing values:

[0043] If the number of missing values ​​is less than 5% of the total data, delete the column data;

[0044] If the number of missing values ​​accounts for 5% to 10% of the total data, the mean interpolation method or median interpolation method is used to interpolate missing values;

[0045] If the number of missing values ​​is greater than 10% of the total data, multiple imputation is used to impute missing values;

[0046] S24: using a box-type method to perform outlier detection on the second processed data set, deleting outliers when they exist, and interpolating using a multiple interpolation method to obtain a third processed data set;

[0047] S25: using the Mahalanobis distance method to perform outlier detection on the third processed data set, deleting outliers when they exist, and interpolating using a multiple interpolation method to obtain a fourth processed data set;

[0048] S26: repeating steps S24-S25 until there are no more outliers, and obtaining a fifth processed data set;

[0049] S27: using the Spearman correlation coefficient to calculate the correlation coefficient between the features in the fifth processed data set to obtain a second correlation coefficient result;

[0050] S28: Determine whether the correlation between the first correlation coefficient result and the second correlation coefficient result is consistent;

[0051] If not, return to step S23, use only the multiple interpolation method to interpolate missing values ​​to obtain the second processed data set, and repeat steps S24-S28;

[0052] If they are consistent, proceed to step S29;

[0053] S29: performing normalization processing on the fifth processed data set to obtain a preprocessed data set.

[0054] Preferably, in step S4, when sorting:

[0055] For the classification obtained after the three-clustering of the data set, the daily production of the disturbed well before the pressure channeling is used as a pressure channeling risk classification measurement indicator;

[0056] For the classification obtained after the four clustering of the data set, if the value of the data above the impact degree threshold in the original data corresponding to the impact degree exceeds 70%, the cumulative impact average daily production of the disturbed wells is used as the pressure channeling risk classification measurement indicator, otherwise the impact degree is used as the pressure channeling risk classification measurement indicator;

[0057] The pressure channeling risk grading measurement indicators are sorted to obtain the shale gas well pressure channeling risk grading.

[0058] Preferably, in step S3, clustering processing specifically includes the following sub-steps:

[0059] S31: Each row of data in the data set is represented as a data point in the multidimensional space. According to the required number of categories, data points are randomly selected as the center of different categories.

[0060] S32: Calculate the distances between the remaining data points in the data set and different classification centers, and classify each data point and the center closest to it into the same category according to the calculated distances;

[0061] S33: After all data are classified, the center of each category is calculated as a new classification center;

[0062] S34: Compare the new classification center with the old classification center to see if they are the same:

[0063] If they are different, the old classification center is replaced with the new classification center, and steps S32-S34 are repeated;

[0064] If they are the same, clustering is completed;

[0065] In step S6, using the classification method to establish a model specifically includes the following sub-steps:

[0066] S61: Calculate the distance between the data to be classified and each training sample, and sort the distances;

[0067] S62: Select the first X samples that are closest to the data to be classified, count the categories of the first X training samples, and select the category with the highest probability as the category of the data to be classified;

[0068] S63: Repeat steps S61-S62 until the classification prediction of all samples to be classified is completed.

[0069] On the other hand, a shale gas well pressure channeling risk evaluation and prediction model established by using any of the above-mentioned establishment methods is also provided, and its application in the evaluation and prediction of shale gas well pressure channeling risk before or after pressure channeling occurs.

[0070] The beneficial effects of the present invention are:

[0071] The present invention has established two more complete evaluation systems for shale gas wells before and after pressure channeling. It has innovatively established a shale gas well pressure channeling risk evaluation and prediction model by using a new data preprocessing method combined with clustering and classification algorithms. The established model can be applied to the prediction needs of shale gas well pressure channeling risks before and after pressure channeling occurs. According to the evaluation and prediction results of the model established by the present invention, production strategies can be formulated in advance on site, dynamic simulation of shale gas well pressure channeling can be performed, and pressure channeling control measures can be optimized to reduce the impact of pressure channeling on gas well production and achieve safe and efficient development of shale gas reservoirs. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0073] Figure 1 It is the structural schematic diagram of the hydraulic fracture;

[0074] Figure 2 It is a structural schematic diagram of a core fracturing device according to a specific embodiment;

[0075] Figure 3 It is a schematic diagram of the pressure channeling distance of a specific embodiment. DETAILED DESCRIPTION

[0076] The present invention is further described below in conjunction with the accompanying drawings and embodiments. It should be noted that, in the absence of conflict, the embodiments in this application and the technical features in the embodiments can be combined with each other. It should be noted that, unless otherwise specified, all technical and scientific terms used in this application have the same meanings as those generally understood by those of ordinary skill in the art to which this application belongs. The words "including" or "comprising" and the like used in the disclosure of the present invention mean that the elements or objects appearing before the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects.

[0077] In one aspect, the present invention provides a method for establishing a shale gas well pressure channeling risk assessment and prediction model, comprising the following steps:

[0078] S1: obtaining a pre-pressure channeling evaluation index and a post-pressure channeling evaluation index, and establishing a data set 1 and a data set 2 according to the pre-pressure channeling evaluation index and the post-pressure channeling evaluation index, respectively.

[0079] In a specific embodiment, when obtaining the pre-pressure channeling evaluation index and the post-pressure channeling evaluation index, it is determined whether pressure channeling occurs in the target well according to any one of the following rules:

[0080] If there is a new well being fracturing within the target well distance threshold, and the daily gas production drops sharply to the daily gas production threshold, it is determined that the target well is fracturing-induced;

[0081] If the increase in the daily water production of the target well exceeds the daily water production increase threshold, it is determined that pressure channeling occurs in the target well;

[0082] If the increase in the casing pressure of the target well exceeds the casing pressure increase threshold, it is determined that pressure channeling occurs in the target well.

[0083] In a specific embodiment, the daily gas production threshold is 0 million cubic meters / day, the daily water production increase threshold is 200%, and the casing pressure increase threshold is 150%.

[0084] In a specific embodiment, the evaluation index before pressure channeling includes the daily production of the disturbed well before pressure channeling, the bottom hole pressure of the disturbed well before pressure channeling, the total fracturing fluid volume of the disturbed well, the degree of development of the pressure fractures in the disturbed well, the total fracturing fluid volume of the pressure channeling source well, and the degree of development of the pressure fractures in the pressure channeling source well;

[0085] The post-pressure channeling evaluation index includes the pre-pressure channeling evaluation index and the index after pressure channeling occurs in the shale gas well;

[0086] The indicators after the shale gas well is compressed and channeled include the number of affected days, compression and channeling distance, impact degree, recovery ratio, and the cumulative average daily production of the disturbed wells.

[0087] In a specific embodiment, the development degree of the pressure fractures in the disturbed well and the development degree of the pressure fractures in the pressure channeling source well are determined by the following method:

[0088] If microseismic testing is carried out on site, the degree of fracture development can be calculated by the following formula:

[0089]

[0090] Where: K frac is the degree of fracture development, pieces; n is the total number of fracture stages, pieces; i is the fracture stages at all levels, pieces; x i is the number of microseismic time responses corresponding to each level of fracturing stage, ;

[0091] If microseismic testing is not performed on site, the degree of fracture development is calculated using the following formula:

[0092]

[0093]

[0094]

[0095]

[0096] Where: L is the half length of the fracture, m; f(X) is the radius of the micro-unit of the fracture volume, m; is the porosity occupied by the fracture volume micro-unit; dX is the height of the fracture volume micro-unit, m; X is the location of the fracture volume micro-unit, m; m is the perforation aperture, m; It is the average porosity of the fractures in the core of each level of fracturing; is the porosity of the core after fracturing under formation pressure and temperature conditions; is the porosity of the core under formation pressure and temperature conditions; n is the number of cores, pieces.

[0097] In the above embodiment, in view of the difficulty in determining the degree of fracture development on site, the present invention innovatively proposes a method for calculating the degree of fracture development when microseismic testing is not performed on site. The calculation method is derived by the following steps:

[0098] To simplify the problem, Figure 1 As shown, it is assumed that the degree of fracture development gradually decreases to 0 from the perforation point to the formation, and the decreasing trend conforms to linear change.

[0099] Take multiple (greater than or equal to 5) unfractured cores near the target fracturing section of the shale gas well and number them corresponding to each level of fracturing section. Take a core as an example, wrap the core with heat shrink film, and measure the core porosity under formation pressure and temperature conditions. Since shale gas reservoirs usually have poor natural fracture development, in order to simplify the problem, it is assumed that the degree of fracture development is mainly contributed by hydraulic fracturing.

[0100] use Figure 2 The core fracturing device shown is used to fractur e the core, and the cross-sectional area of ​​the contact end between the stress applying device of the core fracturing device and the core is the size of the perforation aperture of the shale gas well. After fracturing, the core porosity is measured again under the formation pressure and temperature conditions. The difference between the two porosities is the porosity occupied by the rock fractures. is the average porosity of the fractures in multiple cores. Under formation conditions, it is considered that the fracture shape is a cone with the perforation diameter as the base and the half length of the fracture as the height. Then the shape of a certain fracture volume micro-unit is a truncated cone, and its size is:

[0101]

[0102] Where: V fracture is the size of the fracture volume microunit;

[0103] To simplify the problem, assume that f(X) is a linear function of the volume unit radius and the volume unit position. is a linear function of the porosity occupied by the fracture and the volume unit position, and the calculation methods of the two are shown in equations (3) and (4). In this way, the calculation method of the degree of fracture development when no microseismic test is performed on site as shown in the above equations (2)-(5) can be obtained.

[0104] In a specific embodiment, the number of days of impact is the time from when the old well is compressed and channeled to when the production is restored to the production threshold. The production threshold can be determined according to the on-site demand, and can generally be 100%.

[0105] The pressure channeling distance is the pressure channeling section length between the old well being pressure channeled and the new well being pressure channeling source fracturing section after pressure channeling occurs. When multiple fracturing sections of a new well are simultaneously fractured, causing pressure channeling in the old well, the pressure channeling distance is the minimum pressure channeling section length. The specific calculation method is: draw the horizontal section trajectory of the old well being pressure channeled and the new well being pressure channeling source in the three-dimensional coordinate axis, such as Figure 3 As shown in the figure, the horizontal trajectory of the old well is a straight line L1, the endpoints of L1 are A1 and A2 respectively, and the midpoint of the new well fracturing section where the pressure is coming from is A3. Let A3 and L1 form a plane B1, and the normal vector of B1 is but The plane formed with A1 is the normal plane B2 of B1, and the intersection of B1 and B2 is L2. Solve the vertical distance d1 from A3 to L2 and the vertical distance d2 from A4 to L2. The distance between A3 and A5 is the required compression distance.

[0106] The degree of impact is calculated by the following formula:

[0107]

[0108] Where: D Y Q is the influence degree of pressure channeling; sq is the daily gas production of the shale gas well before the pressure channeling occurs, 10 4 m 3 ;Q sh is the daily gas production of the shale gas well after pressure channeling occurs, 10 4 m 3 ;

[0109] The recovery ratio is calculated by the following formula:

[0110]

[0111] Where: D H is the recovery ratio of pressure channeling; Q F is the daily gas production of the shale gas well after pressure recovery, 10 4 m 3 ;

[0112] The cumulative average daily production of the disturbed wells is calculated by the following formula:

[0113] Q sz =Q sq ×d y -Q shz (8)

[0114] Where: Q sz is the average daily production of the disturbed wells, 10 4 m 3 ;d y is the number of days affected by pressure channeling, d; Q shzis the cumulative gas production of the gas well after the pressure channeling occurs until the daily gas production of the gas well recovers to the threshold value, 10 4 m 3 .

[0115] S2: Preprocessing the data set 1 and the data set 2 respectively to obtain a preprocessed data set 3 and a preprocessed data set 4.

[0116] In the modeling process, the processing of the data set is very important. At present, most research results ignore the processing of the data set, or only use simple methods such as deletion and average value replacement to process the data set, resulting in low data set quality and poor modeling effect. To solve this technical problem, in a specific embodiment, the preprocessing specifically includes the following sub-steps:

[0117] S21: using a box-type method to perform outlier detection on the original data set, deleting outliers in the original data set, and obtaining a first processed data set;

[0118] S22: using the Spearman correlation coefficient to calculate the correlation coefficient between each feature in the first processed data set to obtain a first correlation coefficient result;

[0119] S23: interpolating missing values ​​on the first processed data set to obtain a second processed data set; when interpolating missing values:

[0120] If the number of missing values ​​is less than 5% of the total data, delete the column data;

[0121] If the number of missing values ​​accounts for 5% to 10% of the total data, the mean interpolation method or median interpolation method is used to interpolate missing values;

[0122] If the number of missing values ​​is greater than 10% of the total data, multiple imputation is used to impute missing values;

[0123] S24: using a box-type method to perform outlier detection on the second processed data set, deleting outliers when they exist, and interpolating using a multiple interpolation method to obtain a third processed data set;

[0124] S25: using the Mahalanobis distance method to perform outlier detection on the third processed data set, deleting outliers when they exist, and interpolating using a multiple interpolation method to obtain a fourth processed data set;

[0125] S26: repeating steps S24-S25 until there are no more outliers, and obtaining a fifth processed data set;

[0126] S27: using the Spearman correlation coefficient to calculate the correlation coefficient between the features in the fifth processed data set to obtain a second correlation coefficient result;

[0127] S28: Determine whether the correlation between the first correlation coefficient result and the second correlation coefficient result is consistent;

[0128] If not, return to step S23, use only the multiple interpolation method to interpolate missing values ​​to obtain the second processed data set, and repeat steps S24-S28;

[0129] If they are consistent, proceed to step S29;

[0130] S29: performing normalization processing on the fifth processed data set to obtain a preprocessed data set.

[0131] In the above embodiments, the present invention processes the data set from four major directions: outlier detection, missing value interpolation, correlation analysis, and data normalization, which can establish a more accurate data set, thereby improving the accuracy of subsequent modeling.

[0132] S3: clustering the data set 3 and the data set 4 respectively to obtain the number of clusters, which is the number of shale gas well pressure channeling risk classifications.

[0133] In a specific embodiment, clustering processing specifically includes the following sub-steps:

[0134] S31: Each row of data in the data set is represented as a data point in the multidimensional space. According to the required number of categories, data points are randomly selected as the center of different categories.

[0135] S32: Calculate the distances between the remaining data points in the data set and different classification centers, and classify each data point and the center closest to it into the same category according to the calculated distances;

[0136] S33: After all data are classified, the center of each category is calculated as a new classification center;

[0137] S34: Compare the new classification center with the old classification center to see if they are the same:

[0138] If they are different, the old classification center is replaced with the new classification center, and steps S32-S34 are repeated;

[0139] If they are the same, clustering is completed.

[0140] S4: Sort the classifications obtained after clustering to obtain the risk classification of shale gas well pressure channeling, complete the establishment of a shale gas well pressure channeling risk assessment model, and obtain a risk assessment model before pressure channeling and a risk assessment model after pressure channeling.

[0141] In a specific embodiment, when sorting:

[0142] For the classification obtained after the three-clustering of the data set, the daily production of the disturbed well before the pressure channeling is used as a pressure channeling risk classification measurement indicator;

[0143] For the classification obtained after the four clustering of the data set, if the value of the data above the impact degree threshold in the original data corresponding to the impact degree exceeds 70%, the cumulative impact average daily production of the disturbed wells is used as the pressure channeling risk classification measurement indicator, otherwise the impact degree is used as the pressure channeling risk classification measurement indicator;

[0144] The pressure channeling risk grading measurement indicators are sorted to obtain the shale gas well pressure channeling risk grading.

[0145] In a specific embodiment, the impact threshold is 80%. It should be noted that the larger the impact threshold is, the less clear the classification and sorting after clustering will be when it is used as a pressure channeling risk classification indicator. Those skilled in the art can select other impact thresholds according to accuracy requirements, such as 70%, 60%, etc.

[0146] When using the present invention, the actual knowledge on site and the calculation results of the risk assessment model before pressure channeling and the risk assessment model after pressure channeling can be combined to determine which model to use as the risk assessment model for pressure channeling of shale gas wells. When pressure channeling has not occurred in a shale gas well, the risk assessment model before pressure channeling can be used to make a preliminary judgment on the possible future pressure channeling risk of the shale gas well. After pressure channeling occurs in a shale gas well, since the risk assessment model after pressure channeling has more evaluation indicators, a larger amount of data, and a higher model accuracy, it is recommended to use the risk assessment model after pressure channeling to evaluate the degree of pressure channeling risk of the shale gas well.

[0147] S5: Using the shale gas well pressure channeling risk classification as a label, and using each of the pre-pressure channeling evaluation index and the post-pressure channeling evaluation index as a feature, to divide the training set and the test set.

[0148] In a specific embodiment, 80% of the data is used as a training set and 20% of the data is used as a test set. It should be noted that the data set can also be divided into other proportions such as 70% and 30% or 90% and 10%. The more training set data, the more accurate the result.

[0149] S6: Based on the training set and the test set, a model is established using a classification method to obtain a risk prediction model before pressure channeling and a risk prediction model after pressure channeling.

[0150] In a specific embodiment, using the classification method to establish a model specifically includes the following sub-steps:

[0151] S61: Calculate the distance between the data to be classified and each training sample, and sort the distances;

[0152] S62: Select the first X samples that are closest to the data to be classified, count the categories of the first X training samples, and select the category with the highest probability as the category of the data to be classified;

[0153] S63: Repeat steps S61-S62 until the classification prediction of all samples to be classified is completed.

[0154] On the other hand, a shale gas well pressure channeling risk evaluation and prediction model established by using any of the above-mentioned establishment methods is also provided, and its application in the evaluation and prediction of shale gas well pressure channeling risk before or after pressure channeling occurs.

[0155] In a specific embodiment, the method for establishing a shale gas well pressure channeling risk assessment and prediction model of the present invention is used to perform modeling, which specifically includes the following steps:

[0156] According to the actual shale gas reservoir field data, the data were sorted according to the evaluation indicators before and after pressure channeling to obtain data set 1 and data set 2. The two data sets were preprocessed according to the preprocessing method of the present invention to obtain modeling data sets (data set 3 and data set 4). There are 102 data in the data set. The original data set and part of the modeling data set are shown in Table 1 and Table 2:

[0157] Table 1. Raw data of some wells

[0158]

[0159]

[0160] Table 2 Preprocessing results of raw data of some wells

[0161]

[0162]

[0163] Clustering is performed, and the classifications obtained after clustering are sorted to obtain a risk assessment model before pressure channeling and a risk assessment model after pressure channeling. In this embodiment, the original data on site show that the data with a value exceeding 70% in the original data corresponding to the "degree of impact" accounts for more than 90%, so the "cumulative average daily production of disturbed wells affected" is used as a pressure channeling risk grading measurement indicator in the post-pressure channeling evaluation index system. For the evaluation index before pressure channeling of shale gas wells, the "daily production of disturbed wells before pressure channeling" is used as a pressure channeling risk grading measurement indicator in the known evaluation index system before pressure channeling.

[0164] In this embodiment, the pressure channeling risk is divided into five levels (0-low degree, 1-low degree, 2-medium degree, 3-high degree, 4-high degree), among which the prediction result of the post-pressure channeling risk evaluation model is more consistent with the actual field observation situation (the pressure channeling risk is generally normally distributed, and the gas wells with low pressure channeling risk account for the majority), so the post-pressure channeling risk evaluation model is selected as the shale gas well pressure channeling risk evaluation model.

[0165] The post-pressure channeling evaluation index is used as the modeling feature, and the corresponding post-pressure channeling risk evaluation model prediction results are used as the modeling labels. The data set is divided into 1:5 ratios (i.e., 21 gas wells are randomly selected as the test set to test the model accuracy), and the post-pressure channeling risk prediction model is established. The prediction results of the model are shown in Table 3:

[0166] Table 3 Prediction results of the risk prediction model after pressure channeling

[0167] well name J1 J2 J3 J4 J5 J6 J7 True category Low Medium to high Medium to high high Medium to low Low high Prediction Category Low high Medium to high high Medium to low Low high well name J8 J9 J10 J11 J12 J13 J14 True category Medium to high Medium to high Medium to high Medium to low high Low Medium to low Prediction Category Medium to high Medium to high Medium to high Medium to low high Low Medium to low well name J15 J16 J17 J18 J19 J20 J21 True category Medium to low high Medium to low Low high high Medium to high Prediction Category Medium to low high Medium to low Low high high Medium to low

[0168] It can be seen from Table 3 that the prediction accuracy of the post-pressure channeling risk prediction model established by the present invention is 90.48% (19 out of 21 ports are predicted correctly), which is relatively high.

[0169] In summary, the present invention can evaluate and predict the risk of shale gas well pressure channeling before and after the shale gas well pressure channeling occurs. Compared with the prior art, the present invention has significant progress.

[0170] The above description is only a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Although the present invention has been disclosed as a preferred embodiment as above, it is not intended to limit the present invention. Any technician familiar with the profession can make some changes or modifications to equivalent embodiments of equivalent changes using the technical contents disclosed above without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention still falls within the scope of the technical solution of the present invention.

Claims

1. A method for establishing a shale gas well pressure channeling risk assessment and prediction model, characterized in that: The following steps are involved: S1: obtaining a pre-pressure channeling evaluation index and a post-pressure channeling evaluation index, and establishing a data set 1 and a data set 2 according to the pre-pressure channeling evaluation index and the post-pressure channeling evaluation index respectively; The evaluation indexes before pressure channeling include the daily production of the disturbed well before pressure channeling, the bottom hole pressure of the disturbed well before pressure channeling, the total fracturing fluid volume of the disturbed well, the degree of development of the pressure fractures in the disturbed well, the total fracturing fluid volume of the pressure channeling source well, and the degree of development of the pressure fractures in the pressure channeling source well; The post-pressure channeling evaluation index includes the pre-pressure channeling evaluation index and the index after pressure channeling occurs in the shale gas well; The indicators after the shale gas wells are affected include the number of days, distance, degree of impact, recovery ratio, and the cumulative average daily production of the affected wells; The development degree of the pressure fractures in the disturbed well and the development degree of the pressure fractures in the pressure channeling source well are determined by the following method: If microseismic testing is carried out on site, the degree of fracture development can be calculated by the following formula: Where: K frac is the degree of fracture development, pieces; n is the total number of fracture stages, pieces; i is the fracture stages at all levels, pieces; x i is the number of microseismic time responses corresponding to each level of fracturing stage, ; If microseismic testing is not performed on site, the degree of fracture development is calculated using the following formula: Where: L is the half length of the fracture, m; f(X) is the radius of the micro-unit of the fracture volume, m; is the porosity occupied by the fracture volume micro-unit; dX is the height of the fracture volume micro-unit, m; X is the location of the fracture volume micro-unit, m; m is the perforation aperture, m; It is the average porosity of the fractures in the core of each level of fracturing; is the porosity of the core after fracturing under formation pressure and temperature conditions; is the porosity of the core under the conditions of formation pressure and temperature; n is the number of cores, pieces; The degree of impact is calculated by the following formula: Where: D Y Q is the influence degree of pressure channeling; sq is the daily gas production of the shale gas well before the pressure channeling occurs, 10 4 m 3 ;Q sh is the daily gas production of the shale gas well after pressure channeling occurs, 10 4 m 3 ; The recovery ratio is calculated by the following formula: Where: D H Q is the recovery ratio of pressure channeling; F is the daily gas production of the shale gas well after pressure recovery, 10 4 m 3 ; The cumulative average daily production of the disturbed wells is calculated by the following formula: Q sz =Q sq ×d y -Q shz (8) Where: Q sz is the average daily production of the disturbed wells, 10 4 m 3 ;d y is the number of days affected by pressure channeling, d; Q shz is the cumulative gas production of the gas well after the pressure channeling occurs until the daily gas production of the gas well recovers to the threshold value, 10 4 m 3 ; S2: preprocessing the data set 1 and the data set 2 respectively to obtain a preprocessed data set 3 and a preprocessed data set 4; S3: performing clustering processing on the data set 3 and the data set 4 respectively to obtain the number of clusters, where the number of clusters is the number of shale gas well pressure channeling risk classifications; S4: sorting the classifications obtained after clustering, obtaining the risk classification of shale gas well pressure channeling, completing the establishment of a shale gas well pressure channeling risk assessment model, and obtaining a risk assessment model before pressure channeling and a risk assessment model after pressure channeling; S5: using the shale gas well pressure channeling risk classification as a label, and using each of the pre-pressure channeling evaluation index and the post-pressure channeling evaluation index as a feature, to divide the training set and the test set; S6: Based on the training set and the test set, a model is established using a classification method to obtain a risk prediction model before pressure channeling and a risk prediction model after pressure channeling.

2. The method for establishing a shale gas well pressure channeling risk assessment and prediction model according to claim 1, characterized in that: In step S1, when obtaining the pre-pressure channeling evaluation index and the post-pressure channeling evaluation index, it is determined whether pressure channeling occurs in the target well according to any of the following rules: If there is a new well being fracturing within the target well distance threshold, and the daily gas production drops sharply to the daily gas production threshold, it is determined that the target well is fracturing-induced; If the increase in the daily water production of the target well exceeds the daily water production increase threshold, it is determined that pressure channeling occurs in the target well; If the increase in the casing pressure of the target well exceeds the casing pressure increase threshold, it is determined that pressure channeling occurs in the target well.

3. The method for establishing a shale gas well pressure channeling risk assessment and prediction model according to claim 1, characterized in that: In step S4, when sorting: For the classification obtained after the three-clustering of the data set, the daily production of the disturbed well before the pressure channeling is used as a pressure channeling risk classification measurement indicator; For the classification obtained after the four clustering of the data set, if the value of the data above the impact degree threshold in the original data corresponding to the impact degree exceeds 70%, the cumulative impact average daily production of the disturbed wells is used as the pressure channeling risk classification measurement indicator, otherwise the impact degree is used as the pressure channeling risk classification measurement indicator; The pressure channeling risk grading measurement indicators are sorted to obtain the shale gas well pressure channeling risk grading.

4. The method for establishing a shale gas well pressure channeling risk assessment and prediction model according to any one of claims 1 to 3, characterized in that: In step S3, clustering processing specifically includes the following sub-steps: S31: Each row of data in the data set is represented as a data point in the multidimensional space. According to the required number of categories, data points are randomly selected as the center of different categories. S32: Calculate the distances between the remaining data points in the data set and different classification centers, and classify each data point and the center closest to it into the same category according to the calculated distances; S33: After all data are classified, the center of each category is calculated as a new classification center; S34: Compare the new classification center with the old classification center to see if they are the same: If they are different, the old classification center is replaced with the new classification center, and steps S32-S34 are repeated; If they are the same, clustering is completed; In step S6, using the classification method to establish a model specifically includes the following sub-steps: S61: Calculate the distance between the data to be classified and each training sample, and sort the distances; S62: Select the first X samples that are closest to the data to be classified, count the categories of the first X training samples, and select the category with the highest probability as the category of the data to be classified; S63: Repeat steps S61-S62 until the classification prediction of all samples to be classified is completed.

5. A shale gas well pressure channeling risk assessment and prediction model, characterized in that: The shale gas well pressure channeling risk assessment and prediction model is established by using the method for establishing the shale gas well pressure channeling risk assessment and prediction model as described in any one of claims 1 to 4.

6. Application of the shale gas well pressure channeling risk assessment and prediction model as claimed in claim 5 in the assessment and prediction of the pressure channeling risk of shale gas wells before or after pressure channeling occurs.

Images