Shale gas field yield prediction method

Abstract

The invention discloses a shale gas field yield prediction method. The method comprises the following steps: 1, establishing a shale gas field yield prediction initial model which only takes time as an independent variable parameter; 2, obtaining a yield fitting value of the shale gas field according to the shale gas field yield prediction initial model; 3, drawing a yield prediction curve according to the yield fitting value and the corresponding development time; 4, dividing the yield prediction curve into an ascending section yield prediction curve and a descending section yield prediction curve; 5, respectively processing the ascending section yield prediction curve and the descending section yield prediction curve to obtain a final shale gas field yield prediction model; and 6, predicting the yield of the shale gas field according to the final shale gas field yield prediction model. According to the method, a shale gas field yield prediction model is established through a mathematical method which only takes time as an independent variable parameter, so that more accurate yield prediction can be realized on the basis of fewer parameters, and the technical problems that the existing yield prediction method obtains more parameters, and the parameters are difficult to obtain and inaccurate to obtain are solved.

Description

Technical Field

[0001] The invention belongs to the technical field of shale gas exploration and development, and in particular relates to a shale gas field production prediction method. Background Art

[0002] Shale gas is a very important unconventional oil and gas resource. my country's Sichuan Basin and its surrounding areas contain huge potential shale gas resources. Since the oil company drilled the first shale gas well in 2010, it has successfully explored and developed a series of shale gas fields such as Weiyuan, Changning, Luzhou, Weirong, Jiaoshiba, Zhaotong, and Yongchuan. However, due to the short development time, these shale gas fields are still in the early stages of development. Since the future output of shale gas fields determines the economic benefits of oil companies, accurate prediction of the future output of shale gas fields can effectively guide oil companies in the exploitation of shale gas fields.

[0003] At present, the shale oil and gas production prediction methods mainly include the decline curve analysis method (DCA method), analytical model method, numerical simulation method, etc., but these methods require more parameters, and shale gas fields often cannot accurately obtain the required parameters, so the prediction results are not accurate.

[0004] In addition, the prior art with publication number CN107676085A discloses a method for predicting the production capacity of marine shale gas horizontal well logging, and its steps include: a. Data collection and collation: Establish a database of logging interpretation and fracturing construction parameters for the block to be predicted; b. Analysis of factors affecting shale gas production capacity: Conduct well logging interpretation of the tested wells and analysis of the relationship between fracturing parameters and production capacity, and conclude that the three factors of organic carbon content, brittleness index and fracturing fluid dosage are mainly considered; c. Use the test gas wells to establish a database of three factors: organic carbon content, brittleness index and fracturing fluid dosage; d. Use the tested gas wells to establish a production capacity prediction model: 1) Normalize the three parameters of organic carbon content, brittleness index and fracturing fluid dosage; 2) Use multivariate linear regression to establish a production capacity prediction formula; 3) Calculate the initial production capacity of a single well. This prediction method can better guide the fracturing transformation construction of shale gas, but in actual application, this technology still has the following technical problems:

[0005] 1. When using this technology to predict production capacity, many difficult-to-obtain parameters are required, resulting in high costs and difficulty in prediction.

[0006] 2. The prediction model proposed by this technology has many complex independent variable parameters, which makes the calculation difficult and the prediction process complicated.

[0007] 3. The prediction model proposed by this technology is only applicable to the calculation of the initial production capacity of a single well in marine shale gas fields, and its scope of application is relatively narrow. Summary of the invention

[0008] The purpose of the present invention is to overcome the above-mentioned problems existing in the prior art and to provide a shale gas field production prediction method. The method establishes a shale gas field production prediction model by a mathematical method using only time as an independent variable parameter, which can achieve more accurate production prediction based on fewer parameters, and solves the technical problems that the current production prediction method requires more parameters, the required parameters are difficult to obtain, and the obtained parameters are inaccurate.

[0009] To achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0010] A shale gas field production prediction method, characterized in that it comprises the following steps:

[0011] Step 1: Establish an initial model for predicting shale gas field production with time as the independent variable parameter;

[0012] Step 2: Obtain the production fitting value of the shale gas field according to the initial model for predicting the production of the shale gas field;

[0013] Step 3: Draw a yield prediction curve based on the yield fitting value and the corresponding development time;

[0014] Step 4: Divide the yield prediction curve into an ascending segment yield prediction curve and a descending segment yield prediction curve;

[0015] Step 5: Process the production prediction curve of the ascending section and the production prediction curve of the descending section respectively to obtain the final production prediction model of the shale gas field;

[0016] Step 6: The production of shale gas fields is predicted based on the final model.

[0017] In step 1, the initial model for shale gas field production prediction is established as follows:

[0018]

[0019] T=(YY 0 )——1,2,3,……(years)

[0020] In the formula, Q T To fit the shale gas field production, 10 8 m 3 ; c, n, B are the first fitting coefficient, the second fitting coefficient and the third fitting coefficient of the shale gas field respectively; T is the development time, year; Y is the development year after the shale gas field is put into production, year; Y 0 is the year when the shale gas field was put into production, .

[0021] The calculation method of the first fitting coefficient and the second fitting coefficient is:

[0022] S1: By dividing the initial model of shale gas field production prediction in two consecutive years and taking the natural logarithm, the first relationship between the natural logarithm production ratio and the natural logarithm time ratio is obtained as follows:

[0023]

[0024] S2: Obtain the actual production and corresponding development time of the exploited shale gas fields in recent years, and make a linear relationship diagram of the natural logarithm production ratio and the natural logarithm time ratio based on the actual production and development time obtained;

[0025] S3: Fit the linear relationship graph to obtain the following linear function expression:

[0026]

[0027] S4: Calculate the first fitting coefficient and the second fitting coefficient by combining the first relational expression and the linear function expression in step S3.

[0028] The calculation method of the second fitting coefficient and the third fitting coefficient is:

[0029] S11: Taking the natural logarithm of the fitted shale gas field production in the initial model for shale gas field production prediction, the following second relationship is obtained after sorting:

[0030]

[0031] S22: Obtain the actual production and corresponding development time of the exploited shale gas fields in recent years, and make Linear relationship graph with lnT;

[0032] S33: Fitting the linear relationship graph yields the following linear function expression:

[0033]

[0034] S44: Calculate the second fitting coefficient and the third fitting coefficient by combining the second relational expression and the linear function expression in step S33;

[0035] S55: Compare the second fitting coefficient obtained by the first relationship formula with the second fitting coefficient obtained by the second relationship formula, and take the maximum value as the final second fitting coefficient.

[0036] The obtaining of the actual production of the exploited shale gas field in recent years and the corresponding development time refers to obtaining the actual production and the corresponding development time in the recent 15-30 years.

[0037] In step 4, the yield forecast curve is divided as follows: subtract the yield fitting value of the previous year from the yield fitting value of the next year. If the difference is negative, the subsequent yield forecast curve is divided into a descending segment yield forecast curve from the point where the yield fitting value corresponds to the negative difference.

[0038] In step 5, the final model for shale gas field production prediction is obtained as follows:

[0039]

[0040] In the formula, a1, b 1 are the fourth fitting coefficient and the fifth fitting coefficient of the rising stage production prediction curve respectively; a 2 、b 2 are the sixth fitting coefficient and the seventh fitting coefficient of the downward production prediction curve respectively; p is (Q T -Q T-1 )<0 corresponding to T.

[0041] The calculation method of the fourth fitting coefficient and the fifth fitting coefficient is: first use the least squares method to fit the rising section production prediction curve to obtain a fitting relationship based on the actual production, the production fitting value, the fourth fitting coefficient and the fifth fitting coefficient, and then calculate the fourth fitting coefficient and the fifth fitting coefficient according to the actual production and the production fitting value respectively.

[0042] The specific calculation method of the fourth fitting coefficient and the fifth fitting coefficient is:

[0043] Step 1: Set the fitting relationship as:

[0044]

[0045] in,

[0046]

[0047]

[0048] In the formula, is the average value of all rising segment production fitting values, is the average value of the actual output of all rising sections, Q U is the production fitting value of each rising section, Q U1 is the actual output of each ascending section;

[0049] Step 2: Calculate the covariance of the fitted values ​​of all rising segments and the covariance of the actual output of all rising segments. The specific calculation method is:

[0050]

[0051]

[0052] Where, L QU is the covariance of the fitted values ​​of all rising segment yields, L QUQU1 The covariance of actual production in all ascending segments;

[0053] Step 3: Calculate the fifth fitting coefficient based on the covariance of the fitted values ​​of all rising segments and the covariance of the actual output of all rising segments, and then calculate the fourth fitting coefficient based on the average value of the actual output of all rising segments, the average value of the fitted values ​​of the output of all rising segments and the fifth fitting coefficient. The specific calculation method is:

[0054]

[0055]

[0056] The calculation method of the sixth fitting coefficient and the seventh fitting coefficient is: firstly use the least squares method to fit the descending section production prediction curve to obtain a fitting relationship based on the actual production, the production fitting value, the sixth fitting coefficient and the seventh fitting coefficient, and then calculate the sixth fitting coefficient and the seventh fitting coefficient according to the actual production and the production fitting value respectively.

[0057] The specific calculation method of the sixth fitting coefficient and the seventh fitting coefficient is:

[0058] Step 11: Set the fitting relationship as:

[0059]

[0060] in,

[0061]

[0062]

[0063] In the formula, is the average of all the fitting values ​​of the production in the descending section, is the average value of the actual output of all descending sections, Q D is the production fitting value of each descending section, Q D1 is the actual output of each descending section;

[0064] Step 22: Calculate the covariance of the fitted values ​​of the output of all descending sections and the covariance of the actual output of all descending sections. The specific calculation method is:

[0065]

[0066]

[0067] Where, L QDis the covariance of the fitted values ​​of all the descending segments’ yields, L QDQD1 The covariance of actual production in all decline segments;

[0068] Step 33: First, calculate the seventh fitting coefficient according to the covariance of the actual output of all the descending sections and the covariance of the output fitting values ​​of all the descending sections, and then calculate the sixth fitting coefficient according to the average value of the actual output of all the descending sections, the average value of the output fitting values ​​of all the descending sections and the seventh fitting coefficient. The specific calculation method is:

[0069]

[0070]

[0071] By adopting the above technical solution, the beneficial technical effects of the present invention are:

[0072] 1. The present invention establishes a shale gas field production prediction model by a mathematical method using only time as an independent variable parameter, which can achieve more accurate production prediction based on fewer parameters. Compared with the prior art, the prediction model proposed by the present invention uses time as an independent variable parameter, which has the advantages of simple calculation process, low calculation difficulty, low prediction cost, high prediction efficiency and higher accuracy.

[0073] 2. The present invention is applicable not only to the prediction of the initial production capacity of a single well in marine shale gas fields, but also to the prediction of the initial production capacity of a single well in continental shale gas fields. Compared with the prior art, the scope of application is expanded and the universality is higher.

[0074] 3. The present invention can accurately and quantitatively predict the gas production of shale gas fields in different periods (production increase period, stable production period, adjustment period and process period) after they are put into production, and can provide a quantitative basis for the effective development of shale gas fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 It is a flowchart of the present invention;

[0076] Figure 2 It is a linear relationship diagram of the natural logarithm output ratio and the natural logarithm time ratio in Example 2;

[0077] Figure 3 is the shale gas field in Example 2 Linear relationship graph with lnT;

[0078] Figure 4 is a relationship diagram between the actual value and the fitted value of the annual production of the shale gas field in Example 2;

[0079] Figure 5 This is a comparative relationship diagram drawn in Example 2. DETAILED DESCRIPTION

[0080] Example 1

[0081] The present invention provides a shale gas field production prediction method, which is applicable to the prediction of marine / continental shale gas field production. Figure 1 As shown, it includes the following steps:

[0082] Step 1: Establish an initial model for predicting shale gas field production with time as the independent variable parameter. The established initial model for predicting shale gas field production is:

[0083]

[0084] T=(YY 0 )——1,2,3,……(years)

[0085] In the formula, Q T To fit the shale gas field production, 10 8 m 3 ; c, n, B are the first fitting coefficient, the second fitting coefficient and the third fitting coefficient of the shale gas field respectively; T is the development time, year; Y is the development year after the shale gas field is put into production, year; Y 0 is the year when the shale gas field was put into production, .

[0086] Among them, the calculation method of the first fitting coefficient and the second fitting coefficient is:

[0087] S1: By dividing the initial model of shale gas field production prediction in two consecutive years and taking the natural logarithm, the first relationship between the natural logarithm production ratio and the natural logarithm time ratio is obtained as follows:

[0088]

[0089] S2: Obtain the actual production and corresponding development time of the exploited shale gas fields in recent years, and make a linear relationship diagram of the natural logarithm production ratio and the natural logarithm time ratio based on the actual production and development time obtained.

[0090] S3: Fit the linear relationship graph to obtain the following linear function expression:

[0091]

[0092] S4: Calculate the first fitting coefficient and the second fitting coefficient by combining the first relational expression and the linear function expression in step S3.

[0093] The calculation method of the second fitting coefficient and the third fitting coefficient is:

[0094] S11: Taking the natural logarithm of the fitted shale gas field production in the initial model for shale gas field production prediction, the following second relationship is obtained after sorting:

[0095]

[0096] S22: Obtain the actual production and corresponding development time of the exploited shale gas fields in recent years, and make Linear relationship plot with lnT.

[0097] S33: Fitting the linear relationship graph yields the following linear function expression:

[0098]

[0099] S44: Calculate the second fitting coefficient and the third fitting coefficient by combining the second relationship and the linear function expression in step S33.

[0100] S55: Compare the second fitting coefficient obtained by the first relationship formula with the second fitting coefficient obtained by the second relationship formula, and take the maximum value as the final second fitting coefficient.

[0101] In this step, obtaining the actual production of the exploited shale gas field in recent years and the corresponding development time refers to obtaining the actual production and the corresponding development time in the recent 15-30 years.

[0102] Step 2: After the initial model for predicting the production of shale gas fields is established, the production fitting value of the shale gas fields is obtained according to the initial model for predicting the production of shale gas fields.

[0103] Step 3: Draw the yield prediction curve based on the yield fitting value and the corresponding development time.

[0104] Step 4: Divide the yield forecast curve into an ascending segment yield forecast curve and a descending segment yield forecast curve. The division method is: use the yield fitting value of the next year to subtract the yield fitting value of the previous year. If the difference is negative, then divide the subsequent yield forecast curve into a descending segment yield forecast curve from the point where the yield fitting value corresponds to the negative difference.

[0105] It should be noted that since the yield prediction curve is drawn based on the yield fitting values ​​and the corresponding development time, the rising segment yield prediction curve contains multiple rising segments in which the yield fitting values ​​rise successively, and the falling segment yield prediction curve contains multiple falling segments in which the yield fitting values ​​fall successively.

[0106] Step 5: Process the production prediction curve of the ascending section and the production prediction curve of the descending section respectively to obtain the final model for shale gas field production prediction, as follows:

[0107]

[0108] In the formula, a 1 、b1 are the fourth fitting coefficient and the fifth fitting coefficient of the rising stage production prediction curve respectively; a 2 、b 2 are the sixth fitting coefficient and the seventh fitting coefficient of the downward production prediction curve respectively; p is (Q T -Q T-1 )<0 corresponding to T.

[0109] The calculation method of the above-mentioned fourth fitting coefficient and fifth fitting coefficient is: first use the least squares method to fit the rising section production prediction curve, and obtain the fitting relationship based on the actual production, production fitting value, the fourth fitting coefficient and the fifth fitting coefficient, and then calculate the fourth fitting coefficient and the fifth fitting coefficient according to the actual production and production fitting value respectively.

[0110] Specifically, the specific calculation method of the fourth fitting coefficient and the fifth fitting coefficient is:

[0111] Step 1: Set the fitting relationship as:

[0112]

[0113] in,

[0114]

[0115]

[0116] In the formula, is the average value of all rising segment production fitting values, is the average value of the actual output of all rising sections, Q U is the production fitting value of each rising section, Q U1 is the actual output of each rising section.

[0117] Step 2: Calculate the covariance of the fitted values ​​of all rising segments and the covariance of the actual output of all rising segments. The specific calculation method is:

[0118]

[0119]

[0120] Where, L QU is the covariance of the fitted values ​​of all rising segment yields, L QUQU1 The covariance of actual outputs for all upswings.

[0121] Step 3: Calculate the fifth fitting coefficient based on the covariance of the fitted values ​​of all rising segments and the covariance of the actual output of all rising segments, and then calculate the fourth fitting coefficient based on the average value of the actual output of all rising segments, the average value of the fitted values ​​of the output of all rising segments and the fifth fitting coefficient. The specific calculation method is:

[0122]

[0123]

[0124] The calculation method of the sixth fitting coefficient and the seventh fitting coefficient is as follows: firstly use the least squares method to fit the descending section production prediction curve, and obtain the fitting relationship based on the actual production, the production fitting value, the sixth fitting coefficient and the seventh fitting coefficient, and then calculate the sixth fitting coefficient and the seventh fitting coefficient according to the actual production and the production fitting value respectively.

[0125] Specifically, the specific calculation method of the sixth fitting coefficient and the seventh fitting coefficient is:

[0126] Step 11: Set the fitting relationship as:

[0127]

[0128] in,

[0129]

[0130]

[0131] In the formula, is the average of all the fitting values ​​of the production in the descending section, is the average value of the actual output of all descending sections, Q D is the production fitting value of each descending section, Q D1 is the actual output of each descending section.

[0132] Step 22: Calculate the covariance of the fitted values ​​of the output of all descending sections and the covariance of the actual output of all descending sections. The specific calculation method is:

[0133]

[0134]

[0135] Where, L QD is the covariance of the fitted values ​​of all the descending segments’ yields, L QDQD1 The covariance of actual outputs for all decline segments.

[0136] Step 33: First, calculate the seventh fitting coefficient according to the covariance of the actual output of all the descending sections and the covariance of the output fitting values ​​of all the descending sections, and then calculate the sixth fitting coefficient according to the average value of the actual output of all the descending sections, the average value of the output fitting values ​​of all the descending sections and the seventh fitting coefficient. The specific calculation method is:

[0137]

[0138]

[0139] Step 6: After obtaining the final shale gas field production prediction model according to step 5, the shale gas field production can be predicted using the final shale gas field production prediction model.

[0140] The present invention can achieve more accurate production prediction based on fewer parameters through the shale gas field production prediction model established with time as the independent variable parameter, and has the advantages of simple calculation process, low calculation difficulty, low prediction cost, high prediction efficiency and higher accuracy.

[0141] Example 2

[0142] The applicant further used the actual production of a shale gas field from 1995 to 2021 (Table 1 below) and calculated the fitting coefficients through the double linear regression method, thereby deriving the expression of the initial model for shale gas field production prediction with time as the independent variable parameter.

[0143] Table 1 Actual annual production value of a shale gas field, calculated production fitting value and related data

[0144]

[0145]

[0146] Specifically, the calculation process of the initial model expression for shale gas field production prediction is as follows:

[0147] 1. The initial models for predicting shale gas field production in two consecutive years are set as Formula 1 and Formula 2, as follows:

[0148]

[0149]

[0150] Dividing Formula 1 by Formula 2 and taking the natural logarithm, we get the first relationship between the natural logarithm of the yield ratio and the natural logarithm of the time ratio as follows:

[0151]

[0152] 2. According to the actual output and corresponding development time obtained in Table 1, and The linear relationship diagram is as follows: Figure 2 shown.

[0153] 3. Fit the linear relationship graph to obtain the linear function expression y 1 =1.8556x 1 -0.0808 (where y 1 express x 1 express ), the intercept is -0.0808, which is expressed as The first fitting coefficient c is calculated to be 12.3762; and the slope of the straight line is 1.8556, so the second fitting coefficient n is 1.8556.

[0154] 4. Take the natural logarithm of the fitted shale gas field production in the initial model for shale gas field production prediction to obtain the following formula 4:

[0155]

[0156] After sorting, the second relationship is obtained as follows, formula 5:

[0157]

[0158] 5. According to the actual output and corresponding development time obtained in Table 1, The linear relationship diagram with l%T is as follows Figure 3 shown.

[0159] 6. Fit the linear relationship graph to obtain the linear function expression y 2 =1.8839x 2 +2.1608 (where y 2 express x 2 Indicates l%T), similarly, the slope of the straight line is 1.8839, which is the second fitting coefficient n value. The second fitting coefficient obtained by the first relationship and the second fitting coefficient obtained by the second relationship are compared, and the maximum value is taken as the final second fitting coefficient. In this embodiment, the second fitting coefficient n value is 1.8839, the first fitting coefficient c value is 12.3762 obtained by the first linear regression, and the intercept of the straight line is 2.1608. Substituting the above values ​​into Thus, the third fitting coefficient B is obtained to be 992.5026.

[0160] 7. Substitute the first fitting coefficient c, the second fitting coefficient n, and the third fitting coefficient B obtained above into Formula 1 to obtain the expression of the initial model for shale gas field production prediction with time as a parameter, as shown in Formula 6:

[0161]

[0162] 8. Substitute T=1~27 into Formula 6 to calculate the yield fitting value (for specific data, see the yield fitting value calculated in Table 1), and then make the following calculation: Figure 4 From the relationship diagram between the actual annual production value and the fitted value, it can be seen that, except for the first 5 values, which are significantly different from the actual production values, the other values ​​are basically consistent. The average relative error of the last 22 points is only 4.58%, which has a high accuracy and can meet the production prediction of shale gas fields.

[0163] Furthermore, it can be seen from the above content that although the production fitting value predicted by the initial model for shale gas field production prediction has a high accuracy, there is still a certain error between the two. In order to minimize the error and improve the accuracy, the present invention divides the production prediction curve into an ascending segment production prediction curve and a descending segment production prediction curve on the basis of the initial model for shale gas field production prediction, and uses the least squares method to process the production fitting value of the initial model for shale gas field production prediction, and summarizes the final model for shale gas field production prediction, so as to further reduce the error.

[0164] Specifically, the calculation process of the final model expression for shale gas field production prediction is as follows:

[0165] 1. Since the actual output has an ascending section and a descending section, these two sections are fitted separately.

[0166] For the rising section, the fitting relationship is set as:

[0167]

[0168] in,

[0169]

[0170]

[0171] In the formula, is the average value of all rising segment production fitting values, is the average value of the actual output of all rising sections, Q U is the production fitting value of each rising section, Q U1 is the actual output of each rising section.

[0172] 2. Take the ascending interval (12~510×10 8 m3 )The data is as follows:

[0173]

[0174] By calculation, we get:

[0175]

[0176]

[0177] 3. Obtain the covariance of the fitted values ​​of the output of all rising sections and the covariance of the actual output of all rising sections. The specific calculation method is:

[0178]

[0179]

[0180] Where, L QU is the covariance of the fitted values ​​of all rising segment yields, L QUQU1 The covariance of actual outputs for all upswings.

[0181] By calculation, we get:

[0182]

[0183]

[0184] 4. Calculate the fifth fitting coefficient based on the covariance of the output fitting values ​​of all rising sections and the covariance of the actual output of all rising sections, and then calculate the fourth fitting coefficient based on the average value of the actual output of all rising sections, the average value of the output fitting values ​​of all rising sections and the fifth fitting coefficient. The specific calculation method is:

[0185]

[0186]

[0187] By calculation, we get:

[0188]

[0189]

[0190] That is y 3 =1.084x 3 -17.479 (where y 3 Indicates Q T , x 3 express ); the second fitting formula in the rising stage is:

[0191]

[0192] 5. For the descending section, the fitting relationship is set as:

[0193]

[0194] in,

[0195]

[0196]

[0197] In the formula, is the average of all the fitting values ​​of the production in the descending section, is the average value of the actual output of all descending sections, Q D is the production fitting value of each descending section, Q D1 is the actual output of each descending section;

[0198] 6. Take the descending section (500~450×10 8 m 3 ) data are as follows:

[0199]

[0200] By calculation, we get:

[0201]

[0202]

[0203] 7. Obtain the covariance of the fitted values ​​of all the declining production and the covariance of the actual production in the declining section. The specific calculation method is:

[0204]

[0205]

[0206] Where, L QD is the covariance of the fitted values ​​of all the descending segments’ yields, L QDQD1 The covariance of actual production in all decline segments;

[0207] By calculation, we get:

[0208]

[0209]

[0210] 8. First, the eighth fitting coefficient is calculated according to the covariance of the actual output of all the descending sections and the covariance of the output fitting values ​​of all the descending sections, and then the seventh fitting coefficient is calculated according to the average value of the actual output of all the descending sections, the average value of the output fitting values ​​of all the descending sections and the eighth fitting coefficient. The specific calculation method is:

[0211]

[0212]

[0213] By calculation, we get:

[0214]

[0215]

[0216] That is 4 3 =4.483x 3 -1737.134 (where y 3 Indicates Q T , x 3 express ); then the second fitting formula for the descending section is:

[0217]

[0218] 9. Combining Formula 14 and Formula 22, the final model for shale gas field production prediction is obtained:

[0219]

[0220] Finally, the applicant also used the actual values ​​from 1997 to 2021, the first fitting values ​​calculated by the initial model for shale gas field production prediction, and the second fitting values ​​calculated by the final model for shale gas field production prediction to draw the following figure: Figure 5 The comparison relationship diagram shown in the figure and the comparison table shown in Table 2 below were established.

[0221] Table 2 Actual value of annual production of a shale gas field and two fittings and relative errors

[0222] <![CDATA[Actual value Q T (10 8 m 3 )]]> <![CDATA[First fitted value Q T1 (10 8 m 3 )]]> <![CDATA[Q T1 Relative error]]> <![CDATA[Second fitted value Q T2 (10 8 m 3 )]]> <![CDATA[Q T2 Relative error]]> 45 53.9586 19.91％ 44.5073 1.09％ 67 85.5782 27.73％ 78.4415 17.08％ 100 120.1884 20.19％ 115.5852 15.59％ 140 156.3013 11.64％ 154.3416 10.24％ 180 192.7589 7.09％ 193.4678 7.48％ 240 228.6631 4.72％ 232.0003 3.33％ 270 263.3256 2.47％ 269.2000 0.30％ 310 296.2281 4.44％ 304.5110 1.77％ 335 326.9916 2.39％ 337.5263 0.75％ 378 355.3506 5.99％ 367.9613 2.66％ 410 381.1326 7.04％ 395.6305 3.50％ 430 404.2401 5.99％ 420.4295 2.23％ 445 424.6370 4.58％ 442.3194 0.60％ 465 442.3355 4.87％ 461.3134 0.79％ 480 457.3867 4.71％ 477.4664 0.53％ 487 469.8717 3.52％ 490.8653 0.79％ 500 479.8948 4.02％ 501.6221 0.32％ 505 487.5769 3.45％ 509.8666 0.96％ 510 493.0516 3.32％ 515.7419 1.13％ 510 496.4599 2.65％ 519.3997 1.84％ 500 497.9478 0.41％ 494.7019 1.06％ 493 497.6630 0.95％ 493.4255 0.09％ 480 495.7528 3.28％ 484.8643 1.01％ 470 492.3625 4.76％ 469.6688 0.07％ 450 487.6335 8.36％ 448.4734 0.34％

[0223] According to Table 2 and Figure 5 It can be seen that the second fitting value calculated by the final model for shale gas field production prediction has a smaller error with the actual production, and the corresponding accuracy is higher. This experiment mainly uses the shale gas field production prediction model with time as a parameter to determine the predicted production expression and predict the future production of the developed shale gas field, and the prediction effect is good.

[0224] The above description is only a specific implementation mode of the present invention. Any feature disclosed in this specification, unless otherwise stated, can be replaced by other alternative features that are equivalent or have similar purposes; all the disclosed features, or all the steps in the methods or processes, except for mutually exclusive features and / or steps, can be combined in any way.

Claims

1. A shale gas field production prediction method, characterized in that: The following steps are involved: Step 1: Establish an initial model for predicting shale gas field production with time as the independent variable parameter; Step 2: Obtain the production fitting value of the shale gas field according to the initial model for predicting the production of the shale gas field; Step 3: Draw a yield prediction curve based on the yield fitting value and the corresponding development time; Step 4: Divide the yield prediction curve into an ascending segment yield prediction curve and a descending segment yield prediction curve; Step 5: Process the production prediction curve of the ascending section and the production prediction curve of the descending section respectively to obtain the final production prediction model of the shale gas field; Step 6: The production of shale gas fields is predicted based on the final model.

2. A shale gas field production prediction method according to claim 1, characterized in that: In step 1, the initial model for shale gas field production prediction is established as follows: T=(Y-Y0)——1,2,3,……(years) In the formula, Q T To fit the shale gas field production, 10 8 m 3 ; c, n, B are the first fitting coefficient, the second fitting coefficient and the third fitting coefficient of the shale gas field respectively; T is the development time, year; Y is the development year after the shale gas field is put into production, year; Y0 is the year when the shale gas field is put into production, year.

3. A shale gas field production prediction method according to claim 2, characterized in that: The calculation method of the first fitting coefficient and the second fitting coefficient is: S1: By dividing the initial model of shale gas field production prediction in two consecutive years and taking the natural logarithm, the first relationship between the natural logarithm production ratio and the natural logarithm time ratio is obtained as follows: S2: Obtain the actual production and corresponding development time of the exploited shale gas fields in recent years, and make a linear relationship diagram of the natural logarithm production ratio and the natural logarithm time ratio based on the actual production and development time obtained; S3: Fit the linear relationship graph to obtain the following linear function expression: S4: Calculate the first fitting coefficient and the second fitting coefficient by combining the first relational expression and the linear function expression in step S3.

4. A shale gas field production prediction method according to claim 3, characterized in that: The calculation method of the second fitting coefficient and the third fitting coefficient is: S11: Taking the natural logarithm of the fitted shale gas field production in the initial model for shale gas field production prediction, the following second relationship is obtained after sorting: S22: Obtain the actual production and corresponding development time of the exploited shale gas fields in recent years, and make Linear relationship diagram with lnT; S33: Fitting the linear relationship graph yields the following linear function expression: S44: Calculate the second fitting coefficient and the third fitting coefficient by combining the second relational expression and the linear function expression in step S33; S55: Compare the second fitting coefficient obtained by the first relationship formula with the second fitting coefficient obtained by the second relationship formula, and take the maximum value as the final second fitting coefficient.

5. The shale gas field production prediction method according to claim 1, characterized in that: In step 4, the yield forecast curve is divided as follows: subtract the yield fitting value of the previous year from the yield fitting value of the next year. If the difference is negative, the subsequent yield forecast curve is divided into a descending segment yield forecast curve from the point where the yield fitting value corresponds to the negative difference.

6. A shale gas field production prediction method according to any one of claims 2 to 5, characterized in that: In step 5, the final model for shale gas field production prediction is obtained as follows: Where a1 and b1 are the fourth and fifth fitting coefficients of the yield prediction curve of the ascending stage, respectively; a2 and b2 are the sixth and seventh fitting coefficients of the yield prediction curve of the descending stage, respectively; p is (Q T -Q T-1 )<0 corresponding to T.

7. A shale gas field production prediction method according to claim 6, characterized in that: The calculation method of the fourth fitting coefficient and the fifth fitting coefficient is: first use the least squares method to fit the rising section production prediction curve to obtain a fitting relationship based on the actual production, the production fitting value, the fourth fitting coefficient and the fifth fitting coefficient, and then calculate the fourth fitting coefficient and the fifth fitting coefficient according to the actual production and the production fitting value respectively.

8. A shale gas field production prediction method according to claim 7, characterized in that: The specific calculation method of the fourth fitting coefficient and the fifth fitting coefficient is: Step 1: Set the fitting relationship as: in, In the formula, is the average value of all rising segment production fitting values, is the average value of the actual output of all rising sections, Q U is the production fitting value of each rising section, Q U1 is the actual output of each ascending section; Step 2: Calculate the covariance of the fitted values ​​of all rising segments and the covariance of the actual output of all rising segments. The specific calculation method is: Where, L QU is the covariance of the fitted values ​​of all rising segment yields, L QUQU1 The covariance of actual production in all ascending segments; Step 3: Calculate the fifth fitting coefficient based on the covariance of the fitted values ​​of all rising segments and the covariance of the actual output of all rising segments, and then calculate the fourth fitting coefficient based on the average value of the actual output of all rising segments, the average value of the fitted values ​​of the output of all rising segments and the fifth fitting coefficient. The specific calculation method is:

9. A shale gas field production prediction method according to claim 6, characterized in that: The calculation method of the sixth fitting coefficient and the seventh fitting coefficient is: firstly use the least squares method to fit the descending section production prediction curve to obtain a fitting relationship based on the actual production, the production fitting value, the sixth fitting coefficient and the seventh fitting coefficient, and then calculate the sixth fitting coefficient and the seventh fitting coefficient according to the actual production and the production fitting value respectively.

10. A shale gas field production prediction method according to claim 7, characterized in that: The specific calculation method of the sixth fitting coefficient and the seventh fitting coefficient is: Step 11: Set the fitting relationship as: in, In the formula, is the average of all the fitting values ​​of the production in the descending section, is the average value of the actual output of all descending sections, Q D is the production fitting value of each descending section, Q D1 is the actual output of each descending section; Step 22: Calculate the covariance of the fitted values ​​of the output of all descending sections and the covariance of the actual output of all descending sections. The specific calculation method is: Where, L QD is the covariance of the fitted values ​​of all the descending segments’ yields, L QDQD1 The covariance of actual production in all decline segments; Step 33: First, calculate the seventh fitting coefficient according to the covariance of the actual output of all the descending sections and the covariance of the output fitting values ​​of all the descending sections, and then calculate the sixth fitting coefficient according to the average value of the actual output of all the descending sections, the average value of the output fitting values ​​of all the descending sections and the seventh fitting coefficient. The specific calculation method is:

Images