A method and device for predicting the production decline of a tight gas horizontal well with a variable declining exponent, and a storage medium

Abstract

The application discloses a method and equipment for predicting production decline of a tight gas horizontal well with a variable decreasing exponent, and a storage medium, and comprises the following steps: collecting historical gas production data of a target well; performing nonlinear regression analysis by taking a dimensionless time function as an independent variable and taking a logarithmic value of the production as a dependent variable, and determining a second key undetermined parameter and a linear expression by a characteristic curve fitting method; substituting the parameter and a theoretical initial production into a formula to obtain a time-varying prediction model of the gas well production; comparing and analyzing measured data in the field to obtain an absolute error value; if the absolute error value is greater than an absolute error threshold value, the next step is performed; repeating the steps based on historical gas production data of a measure section to obtain a time-varying prediction model of the gas well production of the measure section. The application can more accurately represent an actual percolation law of the tight gas reservoir, and the theoretical calculation result is more in line with a dynamic material balance principle of the gas reservoir, thereby providing a new theoretical basis and technical means for optimizing a development scheme of the tight gas horizontal well.

Description

Technical Field

[0001] This invention relates to a method, equipment, and storage medium for predicting production decline in tight gas horizontal wells with a variable decline index, belonging to the field of oil and gas field efficient development and production enhancement technology. Background Technology

[0002] Currently, the industry mainly uses numerical simulation, analytical methods, and empirical models (Arps, SEPD, PLE, Duong, etc.) to predict gas well production decline. However, these methods have significant technical limitations in the development of horizontal wells in tight gas reservoirs: numerical simulation is difficult to apply due to low computational efficiency and excessive parameter sensitivity; analytical methods are based on idealized geological assumptions (homogeneous reservoirs, isotropic permeability) and cannot characterize the strong heterogeneity of actual reservoirs; traditional empirical models (such as Arps) use static decline exponents, which are particularly unable to adapt to the dynamic evolution of reservoir properties under pressurized production conditions. This technical bottleneck is particularly prominent in the development of horizontal wells in tight gas reservoirs: on the one hand, pressurized production significantly exacerbates the reservoir stress sensitivity effect; on the other hand, the complex fracture network formed by multi-stage fracturing in horizontal wells leads to a significant enhancement of the time-varying characteristics of reservoir parameters. Existing models lack effective time-varying parameter coupling mechanisms, resulting in insufficient accuracy in predicting dynamic recoverable reserves and final recovery rate, which seriously restricts the efficient development of tight gas reservoirs.

[0003] Therefore, it is urgent to establish a variable exponential decline model that considers the "stress-pressure-seepage coupling" mechanism in order to achieve accurate prediction of the production dynamics of horizontal wells in tight gas reservoirs. Summary of the Invention

[0004] The purpose of this invention is to provide a method for predicting the decline in production of tight gas horizontal wells with a variable decline index, addressing the problems existing in the prior art.

[0005] The technical solution provided by this invention to solve the above-mentioned technical problems is: a method for predicting the decline in production of tight gas horizontal wells with a variable decline index, comprising the following steps:

[0006] Step S10: Collect historical gas production data of the target well and convert the time series data of the production decline stage into standard days.

[0007] Step S20: Preprocess historical gas production data;

[0008] Step S30: Use the dimensionless time function t b With lnt as the independent variable and the logarithm of output lnq as the dependent variable, a nonlinear regression analysis was performed. The second key undetermined parameter b and the linear expression were determined by the characteristic curve fitting method.

[0009] Step S40: Determine the first key undetermined parameter a and the theoretical initial yield q based on the slope and intercept of the linear expression. i ;

[0010] Step S50: The first key undetermined parameter a, the second key undetermined parameter b, and the theoretical initial output q are... i Substitute the formula to obtain the time-varying prediction model for gas well production;

[0011] Step S60: Based on the time-varying prediction model for gas well production, obtain the predicted gas production value for different production periods through numerical solution, and compare it with the field measured data to obtain the absolute error value. If the absolute error value is less than or equal to the absolute error threshold, then predict the target gas well production based on the time-varying prediction model for gas well production; if the absolute error value is greater than the absolute error threshold, proceed to the next step.

[0012] Step S70: The historical gas production data is divided into historical gas production data of the natural decline segment and historical gas production data of the treatment segment. Based on the historical gas production data of the treatment segment, steps S30-S50 are repeated to obtain the time-varying prediction model of gas well production in the treatment segment, and the production of the target gas well is predicted based on the time-varying prediction model of gas well production in the treatment segment.

[0013] A further technical solution is that the preprocessing in step S20 includes removing abnormal production data of the target well during periods such as shut-in, workover, and washing; excluding fluctuation data of the target well during the unstable flow stage and frequent parameter adjustment period in the early stage of production; screening out effective production points of the target well during the continuous and stable production stage; and correcting measurement errors and outliers.

[0014] A further technical solution is that the linear expression in step S30 is:

[0015] y = -ax + lnq i

[0016] In the formula: q i denoted as the theoretical initial output; 'a' is the first critical parameter to be determined.

[0017] A further technical solution is that the time-varying prediction model for gas well production in step S50 is:

[0018]

[0019] In the formula: q is the predicted output; q i t represents the theoretical initial output; a represents the first key undetermined parameter; b represents the second key undetermined parameter; and t represents the prediction time.

[0020] Based on the traditional Duong decline model, assuming that the production well maintains a stable fracture flow state (linear / bilinear flow) throughout its entire life cycle, its production rate satisfies:

[0021] q = q i t -λ

[0022] Where λ represents the model parameters.

[0023] To address the dynamic changes in flow state caused by the coupling effect of "stress-pressure-seepage" in horizontal well pressurization of tight gas reservoirs, this invention breaks through the limitation of the traditional model constant λ and constructs a time-varying coefficient function:

[0024] λ(t)=at b

[0025] Where a is the first key undetermined parameter; b is the second key undetermined parameter.

[0026] To accurately characterize the reservoir dynamics of pressurized production in horizontal tight gas wells, the time-varying coefficient λ(t) is introduced into the Duong model to construct an improved production prediction equation: We can obtain:

[0027]

[0028] Where, q i The theoretical initial production rate needs to be determined by data fitting based on actual gas well production data. To determine the undetermined characteristic parameters in the model, linearizing both sides of the above equation by taking the logarithm yields:

[0029] lnq = -at b lnt+lnq i

[0030] Furthermore, it can be seen from the above formula that lnq(t) and t b There is a linear relationship between lnt and lnt.

[0031] A further technical solution is that the formula for calculating the absolute error value in step S60 is:

[0032]

[0033] In the formula: z is the total number of days selected; q is the actual daily production of the gas well; q(t) is the daily production predicted by the model; and k is the absolute error value.

[0034] A further technical solution is that the absolute error threshold ranges from 5% to 15%.

[0035] A computer device includes a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to perform the method described above.

[0036] A computer-readable storage medium storing computer instructions for causing a computer to perform the above-described method.

[0037] The beneficial effects of this invention are as follows: This invention addresses the complex production decline patterns caused by dynamic changes in reservoir parameters during the pressurization and production of horizontal wells in tight gas reservoirs. Based on the traditional Duong model, it extends and improves upon it, establishing a variable decline index production prediction model that considers the coupling effect of "stress-pressurization-seepage". This model, by introducing a time-varying parameter mechanism, more accurately characterizes the dynamic seepage characteristics of tight gas reservoirs, theoretically solving the technical problem of overestimating dynamic reserves due to neglecting the time-varying effects of the reservoir in traditional methods. Compared to traditional constant parameter models, this method can more accurately characterize the actual seepage patterns of tight gas reservoirs, and its theoretical calculation results are more consistent with the dynamic material balance principle of gas reservoirs, providing a new theoretical basis and technical means for optimizing horizontal well development schemes in tight gas reservoirs. Attached Figure Description

[0038] Figure 1 This is a graph showing the entire fitting result of the present invention used to determine the parameters to be determined in the production expression of well J1 in a tight gas reservoir.

[0039] Figure 2 This is a piecewise fitting result diagram of the production expression of well J1 in a tight gas reservoir used in this invention to determine the parameters to be determined in the natural decline segment;

[0040] Figure 3 This is a piecewise fitting result diagram of the production expression of well J1 in a tight gas reservoir, used in this invention to determine the parameters to be determined in the measure reduction segment;

[0041] Figure 4 This is a comparison chart of the entire prediction results of the traditional model for a specific tight gas reservoir, J1 well, as used in this invention.

[0042] Figure 5 This is a segmented comparison of the prediction results of the traditional model with those of the J1 well in a tight gas reservoir, as used in this invention;

[0043] Figure 6 This is a segmented comparison chart of the predicted results of the traditional model and the measures reduction stage in a tight gas reservoir J1 well, which is an example of the present invention. Detailed Implementation

[0044] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0045] This invention provides a method for predicting production decline in tight gas horizontal wells with a variable decline index, comprising the following steps:

[0046] Step S1: Data collection and preparation steps, collect historical gas production data of a tight gas horizontal well J1, and convert the time series data of the production decline stage into standard days.

[0047] Step S2: Data preprocessing step, remove abnormal production data of well J1 during well shut-in, well workover, well washing, etc., exclude fluctuation data of well J1 in the unstable flow stage and frequent parameter adjustment period in the early stage of production, and screen out the effective production points of well J1 in the continuous stable production stage. At the same time, correct measurement errors and outliers to ensure the accuracy of the decline pattern analysis.

[0048] Step S3: Based on the actual gas production data of well J1 processed in step S2, construct lnq-t b The lnt characteristic curve was optimized using the least squares algorithm. Through iterative calculation and adjustment of the deformation parameter b, it was finally determined that b = 0.4 resulted in the best characteristic curve fitting, and a linear expression was obtained.

[0049] y = -0.0035x - 0.1272

[0050] Step S4: Based on the linear regression equation obtained in step S3, obtain the slope parameter -a and the intercept term lnq. i This allows us to determine the first key undetermined parameter a = 0.0035 and the theoretical initial yield q. i =0.880. Substituting this parameter set into the variable exponential decline model shown in the formula, the time-varying production prediction equation for well J1 is finally established:

[0051]

[0052] Step S5: Based on the established time-varying production prediction model for well J1, the predicted gas production values ​​for different production periods are obtained through numerical solution. These values ​​are then compared with the actual field measurement data to obtain the absolute error value. The results show that the absolute error value is greater than the absolute error threshold, which does not meet the accuracy requirements. Therefore, it is necessary to return to the data preprocessing stage and use a piecewise fitting method for optimization.

[0053] Step S6: Based on the actual gas production data of well J1 processed in step S2, the decline period of well J1 is divided into a natural decline phase and a controlled decline phase, and two lnq-t values ​​are constructed. b The LNT characteristic curves were optimized using a least squares algorithm. The second key undetermined parameter, b, was adjusted through iterative calculation. Ultimately, it was determined that b = 0.4 resulted in the best fitting effect for the two characteristic curve segments, yielding two linear expressions:

[0054] y1 = -0.256x1 + 1.0258

[0055] y² = -0.0082x² - 0.0314

[0056] The slope is -a, and the intercept is lnq. i The first formula is the characteristic curve of the naturally decreasing segment, and the second formula is the characteristic curve of the naturally decreasing segment.

[0057] Step S7: Based on the two linear regression equations obtained in Step S6, obtain the slope parameters -a1, -a2 and the intercept term lnq respectively. i1 lnq i2 This allows us to determine the key undetermined parameters a1 = 0.256, a2 = 0.0082, and the theoretical initial yield q. i1 =2.789, q i2 =0.969. Substituting the two parameter sets into the variable exponential decreasing model shown in the formula, the time-varying production prediction equation suitable for well J1 is finally established:

[0058]

[0059] Where a1 and a2 are the first key undetermined parameters of the natural decline segment and the measure decline segment, respectively, and q i1 q i2 These are the theoretical initial yields for the natural decline phase and the reduced yield phase, respectively.

[0060] Step S8: Based on the established time-varying production prediction model of well J1, the predicted gas production value for different production periods is obtained through numerical solution. At the same time, the Arps model and Duong model are used to predict the production decline of well J1, and the results are compared and analyzed with the field measured data.

[0061] In the above embodiments, compared with actual gas well data, the average error of the Duong model in the natural decline phase and the controlled decline phase is 20.23%; the average error of the Arps model in the natural decline phase and the controlled decline phase is 12.31%; and the average error of the extended Duong model of the present invention in the natural decline phase and the controlled decline phase is 9.73%. It can be seen that the production decline model of the present invention has significantly improved the prediction accuracy compared with the traditional model when used for production decline prediction under the pressure boosting conditions of horizontal wells in tight gas reservoirs.

[0062] A computer device includes a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to perform the method described above.

[0063] A computer-readable storage medium storing computer instructions for causing a computer to perform the above-described method.

[0064] The above description is not intended to limit the present invention in any way. Although the present invention has been disclosed through the above embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some changes or modifications to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall fall within the scope of the present invention.

Claims

1. A method for predicting production decline in tight gas horizontal wells with a variable decline index, characterized in that, Includes the following steps: Step S10: Collect historical gas production data of the target well and convert the time series data of the production decline stage into standard days. Step S20: Preprocess historical gas production data; Step S30: Use a dimensionless time function t b lnt As the independent variable, the logarithm of output ln q A nonlinear regression analysis was performed on the dependent variable, and the second key undetermined parameter was determined by characteristic curve fitting. And linear expressions; The linear expression is: In the formula: This represents the theoretical initial yield; This is the first key parameter to be determined; Step S40: Determine the first key undetermined parameter based on the slope and intercept of the linear expression. and theoretical initial output ; Step S50: Set the first key undetermined parameter Second key undetermined parameter and theoretical initial output Substitute the formula to obtain the time-varying prediction model for gas well production; The time-varying prediction model for gas well production is as follows: In the formula: The daily output predicted by the model; This is the second key parameter to be determined; For predicting time; Step S60: Based on the time-varying prediction model for gas well production, obtain the predicted gas production value for different production periods through numerical solution, and compare it with the field measured data to obtain the absolute error value. If the absolute error value is less than or equal to the absolute error threshold, then predict the target gas well production based on the time-varying prediction model for gas well production; if the absolute error value is greater than the absolute error threshold, proceed to the next step. Step S70: The historical gas production data is divided into historical gas production data of the natural decline segment and historical gas production data of the treatment segment. Based on the historical gas production data of the treatment segment, steps S30-S50 are repeated to obtain the time-varying prediction model of gas well production in the treatment segment, and the production of the target gas well is predicted based on the time-varying prediction model of gas well production in the treatment segment.

2. The method for predicting production decline in tight gas horizontal wells with a variable decline index according to claim 1, characterized in that, The preprocessing in step S20 includes removing abnormal production data of the target well during periods of shut-in, workover, and washing; excluding fluctuation data of the target well during the unstable flow stage and frequent parameter adjustment period in the early stage of production; screening out effective production points of the target well during the continuous and stable production stage; and correcting measurement errors and outliers.

3. The method for predicting production decline in tight gas horizontal wells with a variable decline index according to claim 1, characterized in that, The formula for calculating the absolute error value in step S60 is as follows: In the formula: The total number of days selected; This represents the actual daily production of the gas well. The daily output predicted by the model; This is the absolute error value.

4. The method for predicting production decline in tight gas horizontal wells with a variable decline index according to claim 1, characterized in that, The absolute error threshold ranges from 5% to 15%.

5. A computer device, characterized in that, include: A memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, the processor executing the computer instructions to perform the method of any one of claims 1 to 4.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing a computer to perform the method of any one of claims 1 to 4.

Images