Prediction model generation method and prediction method for final recoverable reserves of shale gas

Abstract

The invention provides a prediction model generation method and prediction method for the final recoverable reserves of shale gas, and the prediction method comprises the steps: obtaining a parameter type and a parameter value of a to-be-predicted shale gas well; according to the parameter type and the parameter value of the to-be-predicted shale gas well, determining the approximate recoverable reserves of the to-be-predicted shale gas well by using a preset rule; performing normalization processing on the to-be-predicted shale gas well parameter values and the approximate recoverable reserves to obtain normalized shale gas well parameter values and normalized approximate recoverable reserves; calculating the correlation between the normalized shale gas well parameter value and the normalized approximate recoverable reserves; the to-be-predicted shale gas well parameter types with the correlation exceeding a preset fourth threshold value serve as main control factors of the to-be-predicted shale gas well; and substituting the main control factors of the to-be-predicted shale gas well into a preset prediction model of the final recoverable reserves of the shale gas to obtain the final recoverable reserves of the to-be-predicted shale gas well. According to the prediction method, the final recoverable reserves of the shale gas well can be accurately and efficiently obtained.

Description

Technical Field

[0001] The invention belongs to the technical field of oil and gas field development, and in particular relates to a prediction method and device for determining the final recoverable reserves of a shale gas well. Background Art

[0002] In recent years, with the proposal of the dual carbon goals, the importance of shale gas in primary energy has become increasingly prominent, especially in China, which is in urgent need of energy transformation. Unlike conventional gas, shale gas has the characteristics of dense reservoirs. Shale gas extraction requires multi-stage hydraulic fracturing of horizontal wells. Shale gas reservoirs have complex nano-scale intragranular pores and micron-scale intergranular pores. Shale gas is retained in the shale formation in the form of free gas and adsorbed gas. The multi-scale storage space and various occurrence states of shale reservoirs determine the complex flow mechanisms of adsorption, desorption and diffusion in shale gas reservoirs. These complex flow mechanisms have caused great trouble to shale gas generation. At home and abroad, more and more attention has been paid to the development of natural gas wells, and the concept of efficient development has also been deeply rooted in the hearts of the people. If the benefits of gas wells are to be accurately predicted, the ultimate recoverable reserves of shale gas wells must be accurately predicted. Summary of the invention

[0003] In order to enrich the prediction method of recoverable reserves of shale gas wells to be applicable to more application scenarios, the present invention proposes a prediction method for determining the total ultimate recoverable reserves of shale gas wells, including:

[0004] In a first aspect, an embodiment of the present invention provides a method for generating a shale gas well final recoverable reserves prediction model for the entire life cycle, comprising:

[0005] Obtain known shale gas well parameter types and parameter values;

[0006] Determine the approximate recoverable reserves of the shale gas well using preset rules according to the parameter type and parameter value of the shale gas well;

[0007] Normalizing the shale gas well parameter values ​​and the approximate recoverable reserves respectively to obtain normalized shale gas well parameter values ​​and normalized approximate recoverable reserves;

[0008] Calculating the correlation between the normalized shale gas well parameter value and the normalized approximate recoverable reserves;

[0009] The shale gas well parameter type whose correlation exceeds a preset first threshold is used as a main control factor;

[0010] The main controlling factors are used as input, the normalized approximate recoverable reserves are used as output, and a least squares support vector machine is used to train the model to obtain a prediction model for the ultimate recoverable reserves of a shale gas well over its entire life cycle.

[0011] In an optional embodiment, the determining the approximate recoverable reserves of the shale gas well by using a preset rule includes:

[0012] The recoverable reserves of multiple shale gas wells are obtained using a preset method;

[0013] When the difference between the recoverable reserves of any two shale gas wells is less than a preset second threshold, an average value of the recoverable reserves of the multiple shale gas wells is taken, and the average value is used as the approximate recoverable reserves of the shale gas wells.

[0014] In an optional embodiment, the determining the approximate recoverable reserves of the shale gas well by using a preset rule further includes:

[0015] When the difference between the recoverable reserves of any two shale gas wells is greater than a preset second threshold, the recoverable reserves of the two shale gas wells are removed and the average value of the recoverable reserves of the remaining other shale gas wells is taken, and the average value is used as the approximate recoverable reserves of the shale gas well.

[0016] In an optional embodiment, the determining the approximate recoverable reserves of the shale gas well by using a preset rule includes:

[0017] The recoverable reserves of multiple shale gas wells are obtained using a preset method;

[0018] If the standard deviation of the recoverable reserves of multiple shale gas wells is less than the preset third threshold, the recoverable reserves of the multiple shale gas wells are averaged, and the average value is used as the approximate recoverable reserves of the shale gas wells; otherwise, the recoverable reserves of the multiple shale gas wells obtained are sorted, and after removing the maximum value and / or the minimum value, the recoverable reserves of the remaining other shale gas wells are averaged, and the average value is used as the approximate recoverable reserves of the shale gas well.

[0019] In an optional embodiment, the method of obtaining recoverable reserves of multiple shale gas wells by using a preset method includes:

[0020] The preset method is at least two of the Blasingame plate method, the Wattenbarger plate method, the Fekotvich plate method, the Agarwal-Gardner plate method, the NPI yield integral plate method, the flow material balance method and the analytical model method.

[0021] In an optional embodiment, normalizing the shale gas well parameter value and the approximate recoverable reserves respectively to obtain the normalized shale gas well parameter value and the normalized approximate recoverable reserves includes:

[0022]

[0023] Among them, y is the normalized shale gas well parameter value or the normalized approximate recoverable reserves, x is the shale gas well parameter value or the approximate recoverable reserves, min is the minimum value of the shale gas well parameter value or the approximate recoverable reserves, and max is the maximum value of the shale gas well parameter value or the approximate recoverable reserves.

[0024] In an optional embodiment, the calculating the correlation between the normalized shale gas well parameter value and the normalized approximate recoverable reserves includes:

[0025] The distance correlation coefficient between the normalized shale gas well parameter value and the normalized approximate recoverable reserves is calculated using a distance correlation coefficient method.

[0026] In a second aspect, an embodiment of the present invention provides a method for predicting the ultimate recoverable reserves of a shale gas well during its entire life cycle, comprising the following steps:

[0027] Obtain parameter types and parameter values ​​of the shale gas well to be predicted;

[0028] Determine the approximate recoverable reserves of the shale gas well to be predicted by using preset rules according to the parameter type and parameter value of the shale gas well to be predicted;

[0029] Normalizing the shale gas well parameter value to be predicted and the approximate recoverable reserves respectively to obtain a normalized shale gas well parameter value and a normalized approximate recoverable reserves;

[0030] Calculating the correlation between the normalized shale gas well parameter value and the normalized approximate recoverable reserves;

[0031] The parameter type of the shale gas well to be predicted whose correlation exceeds a preset fourth threshold is used as the main controlling factor of the shale gas well to be predicted;

[0032] Substitute the main controlling factors of the shale gas well to be predicted into the prediction model of the final recoverable reserves generated as claimed in claim 6 to obtain the final recoverable reserves of the shale gas well to be predicted.

[0033] In an optional embodiment, the shale gas well to be predicted is a well with unknown final recoverable reserves among the shale gas wells that have been put into production, and the parameter types of the shale gas well include geological data, engineering data and production data.

[0034] In an optional embodiment, the shale gas well to be predicted is a shale gas well to be drilled, and the parameter types of the shale gas well include geological data and engineering data.

[0035] In a third aspect, an embodiment of the present invention provides a method for determining the main controlling factors of the final recoverable reserves of a shale gas well, comprising the following steps:

[0036] Obtain known shale gas well parameter types and parameter values;

[0037] Determine the approximate recoverable reserves of the shale gas well using preset rules according to the parameter type and parameter value of the shale gas well;

[0038] Normalizing the shale gas well parameter values ​​and the approximate recoverable reserves respectively to obtain normalized shale gas well parameter values ​​and normalized approximate recoverable reserves;

[0039] Calculating the correlation between the normalized shale gas well parameter value and the normalized approximate recoverable reserves;

[0040] The shale gas well parameter type whose correlation exceeds a preset first threshold is taken as a main controlling factor.

[0041] In a fourth aspect, an embodiment of the present invention provides a prediction model generation device for the final recoverable reserves of a shale gas well, comprising:

[0042] A parameter acquisition module, used to acquire the known shale gas well parameter type and parameter value, and determine the approximate recoverable reserves of the known shale gas well using preset rules;

[0043] A normalization module, used to normalize the shale gas well parameter value and the approximate recoverable reserves respectively to obtain a normalized shale gas well parameter value and a normalized approximate recoverable reserves;

[0044] A correlation calculation module, used for calculating the correlation between the normalized shale gas well parameter value and the normalized approximate recoverable reserves;

[0045] A main control factor determination module, used to take the shale gas well parameter type whose correlation exceeds a preset first threshold as the main control factor;

[0046] The model generation module is used to take the main controlling factors as input and the normalized approximate recoverable reserves as output, and use the least squares support vector machine to train the model to obtain a prediction model for the ultimate recoverable reserves of the shale gas well throughout its life cycle.

[0047] In a fifth aspect, an embodiment of the present invention provides a device for predicting the ultimate recoverable reserves of a shale gas well during its entire life cycle, comprising:

[0048] A parameter acquisition module, used to acquire the parameter type and parameter value of the shale gas well to be predicted, and determine the approximate recoverable reserves of the shale gas well to be predicted using preset rules;

[0049] A normalization module, used to normalize the shale gas well parameter value and the approximate recoverable reserves respectively to obtain a normalized shale gas well parameter value and a normalized approximate recoverable reserves;

[0050] A correlation calculation module, used for calculating the correlation between the normalized shale gas well parameter value and the normalized approximate recoverable reserves;

[0051] A main control factor determination module, used for taking the parameter type of the shale gas well to be predicted whose correlation exceeds a preset fourth threshold as the main control factor of the shale gas well to be predicted;

[0052] The final recoverable reserves determination module is used to substitute the main control factors of the shale gas well to be predicted into the above device to obtain the final recoverable reserves of the shale gas well to be predicted.

[0053] In a sixth aspect, an embodiment of the present invention provides a device for determining the main controlling factors of the final recoverable reserves of a shale gas well, comprising:

[0054] A parameter acquisition module, used to acquire known shale gas well parameter types and parameter values, and determine the approximate recoverable reserves of the shale gas well using preset rules;

[0055] A normalization module, respectively normalizing the shale gas well parameter value and the approximate recoverable reserves to obtain a normalized shale gas well parameter value and a normalized approximate recoverable reserves;

[0056] A correlation calculation module, used for calculating the correlation between the normalized shale gas well parameter value and the normalized approximate recoverable reserves;

[0057] The main controlling factor determination module is used to take the shale gas well parameter type whose correlation exceeds a preset first threshold as the main controlling factor.

[0058] Based on the above technical solution, the beneficial effects of the present invention compared with the prior art are as follows:

[0059] The present invention provides a method for determining the main controlling factors of the final recoverable reserves of a shale gas well. First, based on obtaining the type and value of shale gas well parameters, the approximate recoverable reserves of the shale gas well are determined using preset rules. Then, the shale gas well parameter values ​​and the approximate recoverable reserves are normalized to obtain the normalized shale gas well parameter values ​​and the normalized approximate recoverable reserves. Then, the correlation between the normalized shale gas well parameter values ​​and the normalized approximate recoverable reserves is determined, and the correlation coefficient is calculated by the distance correlation coefficient method. The shale gas well parameter type with a correlation coefficient exceeding a preset threshold is used as the main controlling factor. The method can systematically consider the main controlling factors affecting the final recoverable reserves of a shale gas well, can dig out the complex nonlinear relationship between shale gas well parameters, and can accurately obtain the main controlling factors affecting the final recoverable reserves of a shale gas well.

[0060] The present invention also provides a method for predicting the final recoverable reserves of a shale gas well. First, based on obtaining the parameter type and parameter value of the shale gas well, the approximate recoverable reserves of the shale gas well are determined using a preset rule. Then, the parameter value of the shale gas well and the approximate recoverable reserves are normalized to obtain the normalized shale gas well parameter value and the normalized approximate recoverable reserves. Then, the correlation between the normalized shale gas well parameter value and the normalized approximate recoverable reserves is determined, and the correlation coefficient is calculated by the distance correlation coefficient method. The shale gas well parameter type with a correlation coefficient exceeding a preset threshold is used as a main control factor, the parameter value of the main control factor is used as input, and the normalized approximate recoverable reserves is used as output. The least squares support vector machine training model is used to obtain a prediction model for the final recoverable reserves of the shale gas well in the entire life cycle, and the parameter type and parameter value of the shale gas well to be predicted are substituted into the recoverable reserves prediction model to obtain the final recoverable reserves of the shale gas well. The present invention can accurately and efficiently predict the final recoverable reserves of a shale gas well. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0062] Figure 1 A schematic diagram of a process for determining the main controlling factors of the final recoverable reserves of a shale gas well provided in an embodiment of the present invention;

[0063] Figure 2 A schematic diagram of a process for determining the approximate recoverable reserves of the shale gas well using preset rules is provided for an embodiment of the present invention;

[0064] Figure 3 A schematic flow chart of a method for generating a prediction model for the final recoverable reserves of a shale gas well during its entire life cycle is provided for an embodiment of the present invention;

[0065] Figure 4 Provide a distribution map of the number of shale gas wells and recoverable reserves for embodiments of the present invention;

[0066] Figure 5 Provide a least squares support vector machine training result diagram for an embodiment of the present invention;

[0067] Figure 6 Provide a least squares support vector machine prediction result diagram for an embodiment of the present invention;

[0068] Figure 7A schematic flow chart of a method for predicting the ultimate recoverable reserves of a shale gas well over its entire life cycle is provided for an embodiment of the present invention;

[0069] Figure 8 A schematic diagram of a device for determining the main controlling factors of the final recoverable reserves of a shale gas well is provided for an embodiment of the present invention;

[0070] Fig. 9 A schematic diagram of a device for generating a prediction model for the final recoverable reserves of a shale gas well is provided for an embodiment of the present invention;

[0071] Fig.10 A schematic diagram of a device for predicting the ultimate recoverable reserves of a shale gas well over its entire life cycle is provided for an embodiment of the present invention. DETAILED DESCRIPTION

[0072] At present, there are several methods for predicting the ultimate recoverable reserves of shale gas wells: the first is to use empirical methods for prediction, mainly including the Arps method, the Duong method and the improved Duong method, the SEPD method and the improved SEPD method, and the PLED method; the second is to use the modern production decline method for prediction; the third is to use the analytical model method for prediction; the fourth is to use the numerical simulation method for prediction; the fifth is to use the probability method for prediction. In the process of prediction using the empirical method, it is widely used in oil and gas fields because of its simplicity and ease of operation, but it requires a long production history to better fit the production dynamics and obtain accurate ultimate recoverable reserves. Modern production decline methods, analytical models, and numerical simulations usually rely on more geological and engineering parameters, which are costly and time-consuming to obtain, and are not suitable for the rapid assessment of the ultimate recoverable reserves of multiple wells. The probability method is an uncertainty evaluation method. The assessment of the ultimate recoverable reserves is a range value, but it is easy to cause the problem of difficulty in evaluating the correlation between input parameters.

[0073] In order to at least partially overcome the problems of the prior art, the inventors have invented a method for predicting the ultimate recoverable reserves of shale gas wells through extensive research. In order to better understand the above technical solution, the above technical solution will be described in detail below in combination with specific implementation methods.

[0074] Embodiment 1

[0075] The inventors studied the factors affecting the final recoverable reserves of existing shale gas wells and found that the final recoverable reserves of shale gas wells are affected by some geological, engineering and production data. Therefore, the embodiment of the present invention provides a method for determining the main controlling factors of the final recoverable reserves of shale gas wells, combined with Figure 1 As shown, it includes steps S110 to S150:

[0076] Step S110, obtaining known shale gas well parameter types and parameter values.

[0077] In one embodiment, the known shale gas well parameters include geological data, engineering data and production data. The geological data can be obtained by means of gas reservoir description, seismic, logging and the like. The geological data include the thickness of the shale gas reservoir layer, the depth of the shale gas reservoir, the temperature of the shale gas reservoir, porosity, permeability, gas saturation, pressure coefficient, etc. The engineering data can be obtained by means of drilling and completion, logging and the like. The engineering data include the length of the horizontal section, the number of fracturing stages, the intensity of the sand, the backflow rate, etc. The production data is the statistical data of the daily gas production, water production, static and dynamic liquid level of the shale gas well. In the process of data collection, the integrity of the data is ensured as much as possible, so as to avoid the processing of missing data values, thereby reducing the impact on the results. If there is still data missing, the data can be processed by direct deletion or data interpolation according to the missing data. In order to eliminate the large deviation caused by outliers, a box plot is used to eliminate outliers.

[0078] Taking a shale gas reservoir as an example, data from 169 shale gas wells were collected, and a total of 16 parameter types and parameter values ​​for geology, engineering and production were obtained. The different parameter types are shown in Table 1.

[0079] Table 1 Shale gas well parameter types

[0080] Serial number Influencing factors Serial number Influencing factors 1 Vertical Depth 9 Fracturing stages 2 TOC 10 Liquid Strength 3 Porosity 11 Sand Adding Strength 4 High-quality reservoir thickness 12 Average displacement 5 Gas Saturation 13 Drilling length of type I reservoir 6 Pressure coefficient 14 High-quality reservoir drilling length 7 Brittle mineral content 15 360-day return rate 8 Fracturing section length 16 Peak gas flowback rate

[0081] Step S120, determining the approximate recoverable reserves of the shale gas well according to the parameter type and parameter value of the shale gas well using preset rules.

[0082] Specifically, the approximate recoverable reserves of the shale gas well are determined using preset rules, such as Figure 2 As shown, it includes step S1201 and step S1203:

[0083] Step S1201, using a preset method to obtain recoverable reserves of multiple shale gas wells.

[0084] Specifically, the recoverable reserves of multiple shale gas wells are obtained by using a preset method, and the preset method is at least two of the Blasingame plate method, the Wattenbarger plate method, the Fekotvich plate method, the Agarwal-Gardner plate method, the NPI production integral plate method, the flowing material balance method and the analytical model method. The above methods are all means for calculating the recoverable reserves of oil and gas wells in conventional oil and gas reservoirs. Considering the complex seepage mechanism of shale gas wells, conventional recoverable reserves calculation methods can only obtain approximate recoverable reserves of shale gas wells.

[0085] Step S1202, taking an average value of the recoverable reserves of the multiple shale gas wells.

[0086] Step S1203: When the difference between the recoverable reserves of the multiple shale gas wells and the average does not exceed the preset second threshold, the average value is used as the approximate recoverable reserves of the shale gas well; otherwise, the recoverable reserves with a deviation greater than the preset third threshold are removed, and the average value of the remaining multiple recoverable reserves is used as the approximate recoverable reserves of the shale gas well.

[0087] Specifically, after obtaining multiple shale gas recoverable reserves by at least two of the Blasingame plate method, Wattenbarger plate method, Fekotvich plate method, Agarwal-Gardner plate method, NPI production integral plate method, flowing material balance method and analytical model method, if the differences between the results obtained by various recoverable reserve calculation methods are not very large, the recoverable reserves obtained by various methods are directly averaged as the approximate recoverable reserves of the shale gas well; if there is a large deviation between the results obtained by various recoverable reserve calculation methods, the values ​​with large deviations are removed, and the average of the remaining results is taken as the approximate recoverable reserves of shale gas.

[0088] Step S130, normalizing the shale gas well parameter values ​​and the approximate recoverable reserves respectively to obtain normalized shale gas well parameter values ​​and normalized approximate recoverable reserves.

[0089] Specifically, the shale gas well parameter values ​​and the approximate recoverable reserves are normalized to obtain normalized shale gas well parameter values ​​and normalized approximate recoverable reserves, which can be calculated using the following normalization formula:

[0090]

[0091] Among them, y is the normalized shale gas well parameter value or the normalized approximate recoverable reserves, x is the shale gas well parameter value or the approximate recoverable reserves, min is the minimum value of the shale gas well parameter value or the approximate recoverable reserves, and max is the maximum value of the shale gas well parameter value or the approximate recoverable reserves.

[0092] Step S140, calculating the correlation between the normalized shale gas well parameter value and the normalized approximate recoverable reserves.

[0093] Specifically, the factors that affect the final recoverable reserves of shale gas wells are numerous and complex, and the common Pearson linear correlation coefficient and sensitivity analysis cannot fully tap the nonlinear relationship between factors. In order to overcome the limitations of the Pearson correlation coefficient, the distance correlation coefficient can well measure the nonlinear correlation between variables, and does not need to obey the assumption that it must obey the normal distribution. Therefore, it is selected for the screening of the main controlling factors of the final recoverable reserves. The distance correlation coefficient is used to calculate the correlation coefficient between the obtained geological, engineering, and production parameter data and the final recoverable reserves value obtained by comprehensive evaluation of multiple methods.

[0094] In the process of calculating the correlation between the normalized shale gas well parameter value and the normalized approximate recoverable reserves using the distance correlation coefficient method, for two random variables, the distance correlation coefficient of and can be defined as:

[0095]

[0096] Among them, θ n (X,Y) is called the empirical distance covariance, Likewise, is the k-th row average, is the mean value of the lth column, is the mean of the distance matrix of the X samples.

[0097] Step S150: taking the shale gas well parameter type whose correlation exceeds a preset first threshold as a main control factor.

[0098] Specifically, after obtaining the correlation coefficient between the normalized shale gas well parameter value and the normalized approximate recoverable reserves, the parameters are sorted according to the size of the correlation coefficient. The main controlling factor can be determined by selecting the first few parameters with a larger correlation coefficient with the final recoverable reserves. A first threshold value can be set here. When the correlation coefficient exceeds the first threshold value, this parameter type is determined as the main controlling factor.

[0099] Table 1 lists the 16 characteristic parameters that affect the final recoverable reserves of shale gas wells. The correlation coefficients of each characteristic parameter and the approximate recoverable reserves are calculated to obtain 16 correlation coefficients. These 16 correlation coefficients are sorted by size. According to the preset first threshold, six main control factors are determined, namely, the drilling length of high-quality reservoirs, the drilling length of Class I reservoirs, the length of fracturing sections, the thickness of high-quality reservoirs, the number of fracturing sections, and the gas saturation. It is worth noting that the number of main control factors is related to the set first threshold. In practical applications, the number of main control factors can be reduced or expanded.

[0100] Through extensive research, the inventors have found that the final recoverable reserves of shale gas wells are affected by many factors such as geology, engineering, and production. It is crucial to clarify the main controlling factors and further optimize the development plan to increase the production and efficiency of shale gas. In the process of shale gas exploitation, the implementation of production-increasing measures such as shutting in wells, repeated fracturing, and CO2 injection will bring difficulties to the accurate assessment of the final recoverable reserves of shale gas wells. If the above factors are evaluated and analyzed using the Pearson-MIC comprehensive evaluation method, the gray correlation analysis method, and the random forest importance analysis method, it is still impossible to determine the main controlling factors affecting the final recoverable reserves. In addition, the field data that can be obtained in different production stages of shale gas are different, and the method of evaluating the final recoverable reserves is also different. Therefore, in order to establish a method for quickly and accurately predicting the final recoverable reserves of shale gas wells throughout their life cycle, the present invention first provides a method for determining the main controlling factors of the final recoverable reserves of shale gas wells. The method for determining the main controlling factors of the final recoverable reserves of a shale gas well includes firstly, on the basis of obtaining the parameter type and parameter value of the shale gas well, determining the approximate recoverable reserves of the shale gas well using a preset rule, then, normalizing the parameter value of the shale gas well and the approximate recoverable reserves to obtain the normalized shale gas well parameter value and the normalized approximate recoverable reserves, then, judging the correlation between the normalized shale gas well parameter value and the normalized approximate recoverable reserves, calculating the correlation coefficient by the distance correlation coefficient method, and taking the shale gas well parameter type with a correlation coefficient exceeding a preset threshold as the main controlling factor. The method can systematically consider the main controlling factors affecting the final recoverable reserves of a shale gas well, can dig out the complex nonlinear relationship between the parameters of a shale gas well, and can accurately obtain the main controlling factors affecting the final recoverable reserves of a shale gas well.

[0101] Embodiment 2

[0102] The embodiment of the present invention provides a method for generating a prediction model for the ultimate recoverable reserves of a shale gas well over its entire life cycle, such as Figure 3 As shown, it includes steps S210 to S260:

[0103] Step S210, obtaining known shale gas well parameter types and parameter values.

[0104] In one embodiment, the known shale gas well parameters include geological data, engineering data and production data. The geological data can be obtained through gas reservoir description, seismic, logging and other means, the engineering data can be obtained through drilling and completion, logging and other means, and the production data is the statistical data of the daily gas production, water production, static and dynamic liquid level of the shale gas well. In the process of data collection, the integrity of the data is ensured as much as possible, so as to avoid the processing of missing data values, thereby reducing the impact on the results. If there is still missing data, the data can be processed by direct deletion or data interpolation according to the missing data. In order to eliminate the large deviation caused by outliers, a box plot is used to eliminate outliers.

[0105] Taking a shale gas reservoir as an example, data from 169 shale gas wells were collected, and a total of 16 parameter types and parameter values ​​for geology, engineering and production were obtained. The different parameter types are shown in Table 1.

[0106] Table 1 Shale gas well parameter types

[0107] Serial number Influencing factors Serial number Influencing factors 1 Vertical Depth 9 Fracturing stages 2 TOC 10 Liquid Strength 3 Porosity 11 Sand Adding Strength 4 High-quality reservoir thickness 12 Average displacement 5 Gas Saturation 13 Drilling length of type I reservoir 6 Pressure coefficient 14 High-quality reservoir drilling length 7 Brittle mineral content 15 360-day return rate 8 Fracturing section length 16 Peak gas flowback rate

[0108] Step S220, determining the approximate recoverable reserves of the shale gas well according to the parameter type and parameter value of the shale gas well using preset rules.

[0109] Specifically, the approximate recoverable reserves of the shale gas well are determined using preset rules, such as Figure 2 As shown, it includes step S2201 and step S2202:

[0110] Step S2201, using a preset method to obtain the recoverable reserves of multiple shale gas wells.

[0111] Specifically, the recoverable reserves of multiple shale gas wells are obtained by using a preset method, and the preset method is at least two of the Blasingame plate method, the Wattenbarger plate method, the Fekotvich plate method, the Agarwal-Gardner plate method, the NPI production integral plate method, the flowing material balance method and the analytical model method. The above methods are all means for calculating the recoverable reserves of oil and gas wells in conventional oil and gas reservoirs. Considering the complex seepage mechanism of shale gas wells, conventional recoverable reserves calculation methods can only obtain approximate recoverable reserves of shale gas wells.

[0112] Step S2202, taking an average value of the recoverable reserves of the plurality of shale gas wells, and using the average value as the approximate recoverable reserves of the shale gas wells.

[0113] Specifically, after obtaining multiple shale gas recoverable reserves by at least two of the Blasingame plate method, Wattenbarger plate method, Fekotvich plate method, Agarwal-Gardner plate method, NPI production integral plate method, flowing material balance method and analytical model method, if the differences between the results obtained by various recoverable reserve calculation methods are not very large, the recoverable reserves obtained by various methods are directly averaged as the approximate recoverable reserves of the shale gas well; if there is a large deviation between the results obtained by various recoverable reserve calculation methods, the values ​​with large deviations are removed, and the average of the remaining results is taken as the approximate recoverable reserves of shale gas.

[0114] Step S230, normalizing the shale gas well parameter values ​​and the approximate recoverable reserves respectively to obtain normalized shale gas well parameter values ​​and normalized approximate recoverable reserves.

[0115] Specifically, the shale gas well parameter values ​​and the approximate recoverable reserves are normalized to obtain normalized shale gas well parameter values ​​and normalized approximate recoverable reserves, which can be calculated using the following normalization formula:

[0116]

[0117] Among them, y is the normalized shale gas well parameter value or the normalized approximate recoverable reserves, x is the shale gas well parameter value or the approximate recoverable reserves, min is the minimum value of the shale gas well parameter value or the approximate recoverable reserves, and max is the maximum value of the shale gas well parameter value or the approximate recoverable reserves.

[0118] Step S240, calculating the correlation between the normalized shale gas well parameter value and the normalized approximate recoverable reserves.

[0119] Specifically, the factors that affect the final recoverable reserves of shale gas wells are numerous and complex, and the common Pearson linear correlation coefficient and sensitivity analysis cannot fully tap the nonlinear relationship between factors. In order to overcome the limitations of the Pearson correlation coefficient, the distance correlation coefficient can well measure the nonlinear correlation between variables, and does not need to obey the assumption that it must obey the normal distribution. Therefore, it is selected for the screening of the main controlling factors of the final recoverable reserves. The distance correlation coefficient is used to calculate the correlation coefficient between the obtained geological, engineering, and production parameter data and the final recoverable reserves value obtained by comprehensive evaluation of multiple methods.

[0120] In the process of calculating the correlation between the normalized shale gas well parameter value and the normalized approximate recoverable reserves using the distance correlation coefficient method, for two random variables, the distance correlation coefficient of and can be defined as:

[0121]

[0122] Among them, θ n (X,Y) is called the empirical distance covariance, Likewise, is the k-th row average, is the mean value of the lth column, is the mean of the distance matrix of the X samples.

[0123] Step S250: taking the shale gas well parameter type whose correlation exceeds a preset first threshold as a main control factor.

[0124] Specifically, after obtaining the correlation coefficient between the normalized shale gas well parameter value and the normalized approximate recoverable reserves, the parameters are sorted according to the size of the correlation coefficient. The main controlling factor can be determined by selecting the first few parameters with a larger correlation coefficient with the final recoverable reserves. A first threshold value can be set here. When the correlation coefficient exceeds the first threshold value, this parameter type is determined as the main controlling factor.

[0125] Table 1 lists the 16 characteristic parameters that affect the final recoverable reserves of shale gas wells. The correlation coefficients of each characteristic parameter and the approximate recoverable reserves are calculated to obtain 16 correlation coefficients. These 16 correlation coefficients are sorted by size. According to the preset first threshold, six main control factors are determined, namely, the drilling length of high-quality reservoirs, the drilling length of Class I reservoirs, the length of fracturing sections, the thickness of high-quality reservoirs, the number of fracturing sections, and the gas saturation. It is worth noting that the number of main control factors is related to the set first threshold. In practical applications, the number of main control factors can be reduced or expanded.

[0126] Step S260, taking the main controlling factors as input and the normalized approximate recoverable reserves as output, and using a least squares support vector machine to train a model to obtain a prediction model for the ultimate recoverable reserves of a shale gas well over its entire life cycle.

[0127] Specifically, in order to prevent overfitting of the model and improve the generalization ability of the model, the main controlling factors and approximate recoverable reserves are generally divided into a training set and a test set. The training set is used to train the model, and the test set is used to test the quality of the model. The present invention adopts a random sampling method to select the number of training data and the number of predicted data. During the selection process, it is necessary to ensure that the number of data used for training is more than the number of predicted data. According to a large number of experiments, the ratio of the number of training data to the number of predicted data can be a variety of situations, such as 8:2 and 7:3. In order to make the training effect better, the data is divided in a ratio of 8:2. The mean absolute error can better reflect the actual situation of the prediction value error because the error will not be absolute. It is used to evaluate the effect of the model. The calculation formula is as follows:

[0128]

[0129] Among them, y i ′ is the predicted ultimate recoverable reserves, y i is the approximate recoverable reserves, and n is the number of shale gas wells collected. For different sample data volumes, the modeling models that can be selected include: BP neural network model, random forest algorithm, support vector machine, etc. BP neural network model and random forest algorithm are suitable for large sample data volumes. For small sample data sets, support vector machine is a better choice for regression modeling. Support vector machine is widely used in nonlinear regression and classification problems due to the use of kernel techniques. By statistically analyzing the approximate recoverable reserves data of shale gas wells, the distribution of recoverable reserves is obtained. The number of shale gas wells and the distribution of recoverable reserves are shown in Figure 2. Figure 4 As shown. The least squares support vector machine is an improved support vector machine with an advanced complete knowledge theory system, which can simplify the solution of quadratic optimization problems. With the selected main control factors as input and the final recoverable reserves of the gas well as output, the least squares support vector machine is used to train a final recoverable reserves evaluation model.

[0130] Assume that the input data in the collected data is {x1, x2, x3, x4, x5, ..., x N}, the corresponding output data is {y1, y2, y3, y4, y5, ..., y N}, will collect {x1, x2, x3, ..., x n} and {y1, y2, y3, ..., y n} as the training set to train the model, and take the input data {x1, x2, x3, ..., x n} into the model, and the model output {Y1, Y2, Y3, ..., Y n}, respectively {y1, y2, y3, . . . . , y n} and {Y1, Y2, Y3, ..., Y n} is normalized to obtain the normalized {y1, y2, y3, . . . . , y n} and the normalized {Y1, Y2, Y3, ..., Y n}, the normalized {y1, y2, y3, . . . . , y n} as the normalized true value, and the normalized {Y1, Y2, Y3, ..., Y n} as the normalized predicted value, the normalized true value and the normalized predicted value are plotted in the figure, and the comparison results of the normalized true value and the normalized predicted value are as follows Figure 5As shown in the figure, it can be seen that the data points are concentrated near the diagonal line, indicating that the error between the normalized true value and the normalized predicted value in the training set is very small. (The inventor has conducted a large number of experiments and statistically obtained that the error between the normalized true value and the normalized predicted value is 4.59%, which shows that the model is reliable.

[0131] The {x n+1 , x n+2 ,......,x N} and {y n+1 ,y n+1 ,......y N} as the prediction set to predict the model, and the input data {x n+1 , x n+2 ,......,x N} into the model, and the model output {Y n+1 , Y n+2 ,......,Y N}, respectively {y n+1 ,y n+1 ,......y N} and {Y n+1 , Y n+2 ,......,Y N} normalized to obtain the normalized {y n+1 ,y n+1 ,......y N} and normalized {Y n+1 , Y n+2 ,......,Y N}, the normalized {y n+1 ,y n+1 ,......y N} as the normalized true value, and {Y n+1 , Y n+2 ,......,Y N} as the normalized predicted value, the normalized true value and the normalized predicted value are plotted in the figure, and the comparison results of the normalized true value and the normalized predicted value are as follows Figure 6 As shown in the figure, it can be seen that the data points are concentrated near the diagonal line, indicating that the error between the normalized true value and the normalized predicted value in the prediction set is very small. (The inventor has conducted a large number of experiments and statistically obtained that the error between the normalized true value and the normalized predicted value is 13.05%, which shows that the model is reliable.

[0132] The embodiment of the present invention provides a method for predicting the ultimate recoverable reserves of a shale gas well during its entire life cycle. Figure 7 As shown, it includes steps S210' to S260':

[0133] Step S210', obtaining the parameter type and parameter value of the shale gas well to be predicted.

[0134] Specifically, the shale gas wells to be predicted include shale gas wells that have been put into production but whose ultimate recoverable reserves are unknown and shale gas wells that are planned to be drilled. The parameter types of the shale gas wells that have been put into production but whose ultimate recoverable reserves are unknown include geological data, engineering data and production data, and the parameter types of the shale gas wells that are planned to be drilled include geological data and engineering data.

[0135] Step S220': determining the approximate recoverable reserves of the shale gas well to be predicted by using preset rules according to the parameter type and parameter value of the shale gas well to be predicted.

[0136] Step S230', respectively normalizing the shale gas well parameter values ​​to be predicted and the approximate recoverable reserves to obtain normalized shale gas well parameter values ​​and normalized approximate recoverable reserves.

[0137] Step S240', calculating the correlation between the normalized shale gas well parameter value and the normalized approximate recoverable reserves.

[0138] Step S250': taking the parameter type of the shale gas well to be predicted whose correlation exceeds the preset fourth threshold as the main controlling factor of the shale gas well to be predicted.

[0139] Step S260', substituting the main controlling factors of the shale gas well to be predicted into the prediction model of the final recoverable reserves generated as described above, to obtain the final recoverable reserves of the shale gas well to be predicted.

[0140] The specific implementation methods of step S210' to step S260' can refer to the steps in embodiment 1, and will not be described in detail here.

[0141] The present invention also provides a method for predicting the final recoverable reserves of a shale gas well. First, based on the acquisition of the shale gas well parameter type and parameter value, the approximate recoverable reserves of the shale gas well are determined using a preset rule. Then, the shale gas well parameter value and the approximate recoverable reserves are normalized to obtain the normalized shale gas well parameter value and the normalized approximate recoverable reserves. Then, the correlation between the normalized shale gas well parameter value and the normalized approximate recoverable reserves is determined, and the correlation coefficient is calculated by the distance correlation coefficient method. The shale gas well parameter type with a correlation coefficient exceeding a preset threshold is used as the main control factor, the parameter value of the main control factor is used as input, and the normalized approximate recoverable reserves is used as output. The least squares support vector machine training model is used to obtain a prediction model for the final recoverable reserves of the shale gas well throughout its life cycle. The established prediction model for the final recoverable reserves of a shale gas well can be used for the prediction and evaluation of the final recoverable reserves of a shale gas well throughout its life cycle. Compared with the existing recoverable reserves prediction model, the model established by the present invention has higher prediction accuracy.

[0142] For wells that have been put into production, the final recoverable reserves of the wells that have been put into production can be accurately obtained through geological, engineering and production parameters, and the final recoverable reserves prediction method of shale gas wells established by the present invention can be used. For wells to be drilled, the final recoverable reserves of the wells to be drilled can also be accurately obtained through the geological parameters of the planned drilling and the engineering parameters to be designed. The final recoverable reserves of the wells that have been put into production and the wells to be drilled can be obtained, which can effectively conduct economic evaluation of shale gas reservoir development plans and quickly implement development plan adjustments.

[0143] Embodiment 3

[0144] The embodiment of the present invention provides a device for determining the main controlling factors of the final recoverable reserves of a shale gas well, such as Figure 8 As shown, the device comprises:

[0145] The parameter acquisition module 301 is used to obtain known shale gas well parameter types and parameter values, and determine the approximate recoverable reserves of the shale gas well using preset rules.

[0146] The normalization module 302 performs normalization processing on the shale gas well parameter values ​​and the approximate recoverable reserves, respectively, to obtain normalized shale gas well parameter values ​​and normalized approximate recoverable reserves.

[0147] The correlation calculation module 303 is used to calculate the correlation between the normalized shale gas well parameter value and the normalized approximate recoverable reserves.

[0148] The main controlling factor determination module 304 is used to take the shale gas well parameter type whose correlation exceeds a preset first threshold as the main controlling factor.

[0149] The embodiment of the present invention provides a prediction model generation device for the final recoverable reserves of a shale gas well, such as Fig. 9 As shown, the device comprises:

[0150] The parameter acquisition module 305 is used to obtain the known shale gas well parameter type and parameter value, and determine the approximate recoverable reserves of the known shale gas well using preset rules.

[0151] The normalization module 306 is used to normalize the shale gas well parameter values ​​and the approximate recoverable reserves respectively to obtain normalized shale gas well parameter values ​​and normalized approximate recoverable reserves.

[0152] The correlation calculation module 307 is used to calculate the correlation between the normalized shale gas well parameter value and the normalized approximate recoverable reserves.

[0153] The main controlling factor determination module 308 is used to take the shale gas well parameter type whose correlation exceeds a preset first threshold as the main controlling factor.

[0154] The model generation module 309 is used to take the main controlling factors as input and the normalized approximate recoverable reserves as output, and use the least squares support vector machine to train the model to obtain a prediction model for the ultimate recoverable reserves of the shale gas well throughout its life cycle.

[0155] The embodiment of the present invention provides a device for predicting the ultimate recoverable reserves of a shale gas well during its entire life cycle. Fig.10 As shown, the device comprises:

[0156] The parameter acquisition module 310 is used to obtain the parameter type and parameter value of the shale gas well to be predicted, and determine the approximate recoverable reserves of the shale gas well to be predicted using preset rules.

[0157] The normalization module 311 is used to normalize the shale gas well parameter values ​​and the approximate recoverable reserves respectively to obtain normalized shale gas well parameter values ​​and normalized approximate recoverable reserves.

[0158] The correlation calculation module 312 is used to calculate the correlation between the normalized shale gas well parameter value and the normalized approximate recoverable reserves.

[0159] The main controlling factor determination module 313 is used to take the parameter type of the shale gas well to be predicted whose correlation exceeds a preset fourth threshold as the main controlling factor of the shale gas well to be predicted.

[0160] The final recoverable reserves determination module 314 is used to substitute the main control factors of the shale gas well to be predicted into the above-mentioned device to obtain the final recoverable reserves of the shale gas well to be predicted.

[0161] In the above detailed description, various features are grouped together in a single embodiment to simplify the disclosure. This method of disclosure should not be interpreted as reflecting an intention that the embodiments of the claimed subject matter require more features than are clearly stated in each claim. On the contrary, as reflected in the appended claims, the invention is in a state of having less than all the features of the disclosed individual embodiments. Therefore, the appended claims are hereby expressly incorporated into the detailed description, with each claim standing on its own as a separate preferred embodiment of the invention.

[0162] The above description includes examples of one or more embodiments. Of course, it is not possible to describe all possible combinations of components or methods for the purpose of describing the above embodiments, but it should be recognized by those skilled in the art that the various embodiments may be further combined and arranged. Therefore, the embodiments described herein are intended to cover all such changes, modifications and variations that fall within the scope of protection of the appended claims. In addition, with respect to the term "comprising" used in the specification or claims, the word is encompassed in a manner similar to the term "including", as explained by "including," used as a transitional word in the claims. In addition, any term "or" used in the specification of the claims is intended to mean "non-exclusive or".

Claims

1. A method for generating a prediction model for the ultimate recoverable reserves of shale gas, characterized in that: The following steps are involved: Obtain known shale gas well parameter types and parameter values; Determining the approximate recoverable reserves of the shale gas well using preset rules based on the parameter type and parameter value of the shale gas well; Normalizing the shale gas well parameter values ​​and the approximate recoverable reserves respectively to obtain normalized shale gas well parameter values ​​and normalized approximate recoverable reserves; Calculating the correlation between the normalized shale gas well parameter value and the normalized approximate recoverable reserves; The shale gas well parameter type whose correlation exceeds a preset first threshold is used as a main controlling factor; The main controlling factors are used as input and the normalized approximate recoverable reserves are used as output. A least squares support vector machine is used to train a model to obtain a prediction model for the ultimate recoverable reserves of a shale gas well throughout its life cycle.

2. The method according to claim 1, wherein Determining the approximate recoverable reserves of the shale gas well using preset rules includes: Obtain recoverable reserves of multiple shale gas wells using a preset method; When the difference between the recoverable reserves of any two shale gas wells is less than a preset second threshold, the recoverable reserves of the multiple shale gas wells are averaged and the average is used as the approximate recoverable reserves of the shale gas wells.

3. The method according to claim 2, wherein The method of determining the approximate recoverable reserves of the shale gas well using preset rules further includes: When the difference between the recoverable reserves of any two shale gas wells is greater than a preset second threshold, the recoverable reserves of the two shale gas wells are removed and the average value of the recoverable reserves of the remaining shale gas wells is taken as the approximate recoverable reserves of the shale gas well.

4. The method according to claim 1, wherein Determining the approximate recoverable reserves of the shale gas well using preset rules includes: Obtain recoverable reserves of multiple shale gas wells using a preset method; If the standard deviation of the recoverable reserves of multiple shale gas wells is less than a preset third threshold, the recoverable reserves of the multiple shale gas wells are averaged and the average value is used as the approximate recoverable reserves of the shale gas well; otherwise, the recoverable reserves of the multiple shale gas wells obtained are sorted, and after removing the maximum value and / or the minimum value, the recoverable reserves of the remaining shale gas wells are averaged and the average value is used as the approximate recoverable reserves of the shale gas well.

5. The method according to claim 2, wherein The method of obtaining recoverable reserves of multiple shale gas wells using a preset method includes: The preset method is at least two of the Blasingame plate method, the Wattenbarger plate method, the Fekotvich plate method, the Agarwal-Gardner plate method, the NPI yield integral plate method, the flow material balance method and the analytical model method.

6. The method according to claim 1, wherein Normalizing the shale gas well parameter values ​​and the approximate recoverable reserves to obtain normalized shale gas well parameter values ​​and normalized approximate recoverable reserves, respectively, includes: Where y is the normalized shale gas well parameter value or the normalized approximate recoverable reserves, x is the shale gas well parameter value or the approximate recoverable reserves, min is the minimum value of the shale gas well parameter value or the approximate recoverable reserves, and max is the maximum value of the shale gas well parameter value or the approximate recoverable reserves.

7. The method according to claim 1, wherein Calculating the correlation between the normalized shale gas well parameter value and the normalized approximate recoverable reserves includes: The distance correlation coefficient between the normalized shale gas well parameter value and the normalized approximate recoverable reserves is calculated using a distance correlation coefficient method.

8. A method for predicting the ultimate recoverable reserves of a shale gas well, characterized in that: The following steps are involved: Obtain parameter types and parameter values ​​of the shale gas well to be predicted; Determining the approximate recoverable reserves of the shale gas well to be predicted using preset rules according to the parameter type and parameter value of the shale gas well to be predicted; Normalizing the shale gas well parameter values ​​to be predicted and the approximate recoverable reserves to obtain normalized shale gas well parameter values ​​and normalized approximate recoverable reserves; Calculating the correlation between the normalized shale gas well parameter value and the normalized approximate recoverable reserves; The parameter type of the shale gas well to be predicted whose correlation exceeds a preset fourth threshold is used as the main controlling factor of the shale gas well to be predicted; Substitute the main controlling factors of the shale gas well to be predicted into the prediction model of final recoverable reserves generated by the method according to any one of claims 1 to 7 to obtain the final recoverable reserves of the shale gas well to be predicted.

9. The method for predicting ultimate recoverable reserves according to claim 8, wherein: The shale gas well to be predicted is a well whose final recoverable reserves are unknown among the shale gas wells that have been put into production. The parameter types of the shale gas well include geological data, engineering data and production data.

10. The method for predicting ultimate recoverable reserves according to claim 8, wherein: The shale gas well to be predicted is a shale gas well to be drilled, and the parameter types of the shale gas well include geological data and engineering data.

11. A method for determining the main controlling factors of the ultimate recoverable reserves of shale gas wells, characterized in that: The following steps are involved: Obtain known shale gas well parameter types and parameter values; Determining the approximate recoverable reserves of the shale gas well using preset rules based on the parameter type and parameter value of the shale gas well; Normalizing the shale gas well parameter values ​​and the approximate recoverable reserves respectively to obtain normalized shale gas well parameter values ​​and normalized approximate recoverable reserves; Calculating the correlation between the normalized shale gas well parameter value and the normalized approximate recoverable reserves; The shale gas well parameter type whose correlation exceeds a preset first threshold is taken as a main controlling factor.

12. A device for generating a prediction model for the ultimate recoverable reserves of a shale gas well, characterized in that: include: A parameter acquisition module is used to obtain the type and value of known shale gas well parameters and determine the approximate recoverable reserves of the known shale gas well using preset rules; a normalization module, configured to perform normalization processing on the shale gas well parameter values ​​and the approximate recoverable reserves, respectively, to obtain normalized shale gas well parameter values ​​and normalized approximate recoverable reserves; a correlation calculation module, configured to calculate the correlation between the normalized shale gas well parameter value and the normalized approximate recoverable reserves; a main controlling factor determination module, configured to take the shale gas well parameter type whose correlation exceeds a preset first threshold as the main controlling factor; The model generation module is used to take the main controlling factors as input and the normalized approximate recoverable reserves as output, and use the least squares support vector machine to train the model to obtain a prediction model for the ultimate recoverable reserves of the shale gas well throughout its life cycle.

13. A device for predicting the ultimate recoverable reserves of a shale gas well, characterized in that: include: A parameter acquisition module is used to obtain the parameter type and parameter value of the shale gas well to be predicted, and determine the approximate recoverable reserves of the shale gas well to be predicted using preset rules; a normalization module, configured to perform normalization processing on the shale gas well parameter values ​​and the approximate recoverable reserves, respectively, to obtain normalized shale gas well parameter values ​​and normalized approximate recoverable reserves; a correlation calculation module, configured to calculate the correlation between the normalized shale gas well parameter value and the normalized approximate recoverable reserves; a main controlling factor determination module, configured to use the parameter type of the shale gas well to be predicted whose correlation exceeds a preset fourth threshold as the main controlling factor of the shale gas well to be predicted; The final recoverable reserves determination module is used to substitute the main control factors of the shale gas well to be predicted into the device as claimed in claim 12 to obtain the final recoverable reserves of the shale gas well to be predicted.

14. A device for determining the main controlling factors of the final recoverable reserves of a shale gas well, characterized in that: include: A parameter acquisition module is used to obtain the known shale gas well parameter types and parameter values, and determine the approximate recoverable reserves of the shale gas well using preset rules; a normalization module, performing normalization processing on the shale gas well parameter values ​​and the approximate recoverable reserves, respectively, to obtain normalized shale gas well parameter values ​​and normalized approximate recoverable reserves; a correlation calculation module, configured to calculate the correlation between the normalized shale gas well parameter value and the normalized approximate recoverable reserves; The main controlling factor determination module is used to take the shale gas well parameter type whose correlation exceeds a preset first threshold as the main controlling factor.

Images