Computer readable medium, drilling well bottom hole pressure calculation system and intelligent drilling platform

Abstract

The invention discloses a computer readable medium, a drilling well bottom hole pressure calculating system and an intelligent drilling platform, a computer program is stored in the computer readable medium, and when the computer program is executed by a processor, the following steps are achieved: dividing a wellbore structure into a plurality of cells according to wellbore structure parameters; according to the measured drilling fluid parameters, pressure values and well bore structure parameters at the first cell, calculating drilling fluid parameters and pressure values at the other one or more cells through a preset friction resistance coefficient network model and a drilling fluid rheological parameter network model; and correcting a preset friction resistance coefficient network model and a drilling fluid rheological parameter network model according to the calculated drilling fluid parameters and pressure values at one or more cells and a preset loss function. According to the computer program, a calculated model is optimized in real time in a data and physical coupling mode, multi-source data of a well site is fully utilized, and the prediction precision of the model is improved.

Description

Technical Field

[0001] This application belongs to the technical field of oil and gas development, and particularly relates to a computer-readable medium, a drilling bottom hole pressure calculation system, and an intelligent drilling platform. Background Art

[0002] With the exploitation of energy by humans, the world's energy has shown a trend of continuous depletion. The newly discovered oil reservoirs by humans are also continuously developing towards deeper strata and deeper seawater. In China's oil energy reserves, the proportion of deepwater and deep strata fields is very heavy. Although deepwater oil reservoirs bring rich resources, they are accompanied by increasingly complex geological conditions and increasingly expensive exploitation costs.

[0003] As the water depth increases, the overburden pressure decreases, resulting in a decrease in the fracture pressure. Moreover, abnormal pressures are generally developed in deepwater strata, and drilling faces the problem of a narrow safety window, which easily leads to malignant events such as blowouts and lost circulation.

[0004] To solve this problem, it is necessary to adopt pressure control drilling technology. By adjusting the wellhead back pressure in real time, the wellbore pressure can always be controlled within the safety window between the formation fracture pressure and the formation pore pressure. The most core issue lies in the accurate prediction of the wellbore pressure.

[0005] Most of the existing pressure control drilling hydraulics software systems are constructed based on traditional physical methods. These traditional methods are based on existing physical laws and establish corresponding physical models in combination with drilling working conditions characteristics to predict the pressure in the wellbore. However, traditional physical methods usually have problems such as overly idealized models and key parameters relying on experience, which lead to insufficient accuracy of the prediction results and are unable to fully adapt to the actual situation when facing complex and variable wellsite working conditions.

[0006] With the advancement of wellsite informatization, the technologies for collecting and processing wellsite data have been greatly improved, which provides the possibility for the application of machine learning in wellbore pressure prediction. Although there are currently some prediction methods and systems based on machine learning, these methods usually adopt simple models, lack physical constraints, and overall show "black box" characteristics, lacking interpretability, and are unable to meet the requirements of high-precision and high-reliability drilling operations. Nor can they optimize the system in real time during operation using the construction wellsite data. Summary of the Invention

[0007] To solve the above problems, this application provides a computer-readable medium, a drilling bottom hole pressure calculation system, and an intelligent drilling platform.

[0008] A computer-readable medium stores a computer program thereon. When the computer program is executed by a processor, the following steps are implemented:

[0009] Divide the wellbore structure into several cells according to the wellbore structure parameters;

[0010] According to the measured drilling fluid parameters, pressure values and wellbore structure parameters at the first cell, calculate the drilling fluid parameters and pressure values at another cell or multiple cells through a preset friction coefficient network model and a drilling fluid rheological parameter network model;

[0011] According to the calculated drilling fluid parameters and pressure values at one or more cells, and a preset loss function, correct the preset friction coefficient network model and drilling fluid rheological parameter network model.

[0012] As a further optimization of the present invention, the step of calculating the drilling fluid parameters and pressure values at another cell or multiple cells through a preset friction coefficient network model and a drilling fluid rheological parameter network model according to the measured drilling fluid parameters, pressure values and wellbore structure parameters at the first cell includes:

[0013] Calculate the friction coefficient of the first cell according to the measured drilling fluid parameters, pressure values and wellbore structure parameters of the first cell;

[0014] Calculate the pressure at the second cell according to the friction coefficient, drilling fluid parameters, wellbore structure parameters of the first cell and the measured first pressure;

[0015] Calculate the drilling fluid parameters at the second cell according to the pressure at the second cell and the drilling fluid parameters at the first cell.

[0016] As a further optimization of the present invention, the step of calculating the friction coefficient of the first cell according to the measured drilling fluid parameters, pressure values and wellbore structure parameters of the first cell includes:

[0017] Construct the initial vector (l 1 , v 1 , d h1 , d p1 , k 1 , T 1 , V 1 , ρ 1 , P 1 , H 1 ) of the first cell;

[0018] Through the friction coefficient network model f(d h1 , d p1 , K 1 , ρ 1 , v 1 ) = f 1 Calculate the friction coefficient f 1 of the first cell;

[0019] wherein, l 1 is the length of the first cell, v 1 is the flow rate of the drilling fluid in the first cell, d h1 is the outer diameter of the annulus of the first cell, d p1 is the inner diameter of the annulus of the first cell, k 1 is the viscosity coefficient of the drilling fluid in the first cell, T 1 is the formation temperature in the first cell, V 1 is the mechanical rotation speed in the first cell, ρ 1 is the density of the drilling fluid in the first cell, P 1 is the pressure in the first cell, H 1 is the well depth at the outlet of the first cell.

[0020] As a further optimization of the present invention, the step of calculating the pressure at the second cell according to the friction coefficient, drilling fluid parameters, wellbore structure parameters of the first cell and the measured first pressure includes:

[0021] According to the length l of the first cell 1 , well depth H 1 , inclination angle, density ρ of the drilling fluid 1 , friction coefficient f 1 , and the physical information derivation equation PI l 1 , ρ 1 , P 1 , H 1 = P 2 , calculate the second pressure P at the second cell 2 .

[0022] As a further optimization of the present invention, the step of calculating the drilling fluid parameters at the second cell according to the pressure of the second cell and the drilling fluid parameters of the first cell includes:

[0023] According to the liquid rheology network ρ 2 is the density of the drilling fluid at the second cell, k 2 is the viscosity coefficient of the drilling fluid at the second cell.

[0024] As a further optimization of the present invention, the step of calculating the drilling fluid parameters and pressure values at one or more other cells according to the drilling fluid parameters, pressure values and wellbore structure parameters measured at the first cell through a preset friction coefficient network model and drilling fluid rheological parameter network model includes:

[0025] Construct the initial vector of the nth cell (l n , v n , d hn , dpn , k n , T n , V n , ρ n , P n , H n );

[0026] After passing through the friction coefficient network f(d hn , d pn , K n , ρ n , v n ) = f n The friction coefficient f of the nth cell is obtained n ;

[0027] According to the length l of the nth cell n , well depth H n , inclination angle, drilling fluid density ρ n , friction coefficient f n , and the physical information derivation equation PIl n , ρ n , P n , H n = P n+1 , the pressure P at the (n + 1)th cell is calculated n+1；

[0028] According to the liquid rheology network ρ n+1 is the drilling fluid density at the (n + 1)th cell, k n+1 is the viscosity coefficient of the drilling fluid at the (n + 1)th cell, n = 2, 3, 4, 5...

[0029] As a further optimization of the present invention, when the computer program is executed by a processor, the following steps are implemented:

[0030] Format the received data.

[0031] The present invention also provides a drilling bottom hole pressure calculation system, including a processor and a computer-readable medium as described in any one of the above.

[0032] The present invention also provides an intelligent drilling platform, including the drilling bottom hole pressure calculation system as described above and a logging instrument electrically connected to the computer-readable medium.

[0033] The computer-readable medium provided by the present application, when executed, can optimize the calculation model in a data-physics coupling manner, not only complying with the constraints of traditional physical laws but also making full use of multi-source data on the well site, further improving the prediction accuracy of the model and being able to calibrate the formation pressure in real time, thereby facilitating the regulation of the bottom hole pressure. Brief Description of the Drawings

[0034] The drawings herein are incorporated into and constitute a part of this specification, showing embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application.

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0036] Figure 1 System architecture diagram of the computer program described in the present invention;

[0037] Figure 2 Schematic diagram of a cell in the present invention;

[0038] Figure 3 Schematic diagram of the wellsite layout.

[0039] Description of the reference numerals in the drawings:

[0040] 1. Data processing module; 2. Core computing module; 3. Real-time computing module; 4. Application module; 5. Online learning module; 6. Computer; 7. Blowout preventer group; 8. Choke manifold; 9. Flowmeter; 10. Logging tool; 11. Wellbore. Detailed Description of the Embodiments

[0041] In order to enable those skilled in the art of the present technology to better understand the solutions of the present application, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present application.

[0042] It should be noted that the terms "first", "second", etc. in the specification, claims and the above-mentioned drawings of this application are used to distinguish similar objects, and do not necessarily describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of this application described here can be implemented in an order other than those illustrated or described here. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0043] As Figure 1 shown, the present invention provides a computer-readable medium on which a computer program is stored, which includes a data processing module 1, a core computing module 2, a real-time computing module 3, an application module 4, and an online learning module 5.

[0044] The data processing module 1 is used to format the received data.

[0045] The core computing module 2 includes a first model, and the first model includes a friction coefficient network model for calculating the friction coefficient and a drilling fluid rheological parameter network model for calculating the rheological parameters of the drilling fluid.

[0046] Among them, the friction coefficient network model fd h ,d p ,K,ρ,v=f,

[0047] In the formula, d h : the outer diameter of the annulus, d p : the inner diameter of the annulus, K: consistency coefficient, ρ: drilling fluid density, v: drilling fluid flow rate, f: friction coefficient.

[0048] The drilling fluid rheological parameter network model

[0049] In the formula, ρ: the drilling fluid density vector; k: viscosity coefficient, T: temperature, P: pressure, ρ 1 : the density under this condition, k 1 : the viscosity coefficient under this condition.

[0050] The above two models are pre-trained in the laboratory, so that the two models learn the variation laws of the rheological parameters of the drilling fluid under different temperatures and pressures, and the pressure losses, that is, the friction pressure losses, under different drilling fluids, fluid flow rates and casing sizes. These data can be truly obtained under laboratory conditions, and the two models can effectively learn the corresponding laws.

[0051] The real-time calculation module 3 collaborates with the core calculation module to deduce and calculate the drilling fluid parameters and pressure values of each cell, and transfers the relevant data to the application module and the online learning module.

[0052] In this embodiment, the real-time calculation module 3 adopts the cell method of traditional physical methods, and discretizes the wellbore from the wellhead to the bottom into multiple cells, such as Figure 2 shown, State is the state of each cell, that is, the average pressure of each cell. Param represents the parameters required for calculating the pressure of the next cell, including: wellbore length (i.e., the length of cell division), drilling fluid density, drilling fluid viscosity, drilling fluid flow behavior index, drilling fluid dynamic shear stress, and friction coefficient. Among them, parameters such as flow velocity and wellbore radius are numerical values that can be determined. The wellbore radius can be directly obtained according to the casing size, and the flow velocity can be calculated based on the displacement and the volume of the wellbore. Among them, the rheological parameters and friction coefficient of the liquid are obtained from the output of the network. PI is physical information, and the following recurrence formula can be derived

[0053] P n+1 = P n + ρgh + F

[0054] Finally, the pressure of the next cell is obtained and output.

[0055] The well depth is divided into Figure 2 shown in multiple such cell links. The number of cells is the number of discrete divisions of the wellbore. That is, for each calculated cell pressure, it and the parameters of this cell (in this embodiment, the approximate value of the finite difference method is used instead) are brought into the friction coefficient network model and the drilling fluid rheological parameter network model, and the corresponding network modules output the corresponding results. By iterating according to the above process, the bottom hole pressure can be obtained.

[0056] The application module 4 is used to upload data to the cloud, display terminal or control terminal.

[0057] The online learning module 5 calculates the loss value according to the data calculated by the real-time calculation module and the loss function, and feeds back the loss value to the friction coefficient network model and the drilling fluid rheological parameter network model to optimize the friction coefficient network model and the drilling fluid rheological parameter network model.

[0058] Regarding the bottom hole pressure, it is also a numerical value that can be obtained from the sensor. Constraining the final desired result, that is, the loss function.

[0059]

[0060] After derivation and simplification, it is

[0061] Ppre = P real

[0062] Regarding the frictional pressure loss, from the wellhead to the bottom of the drill pipe and then from the annulus to the wellhead, the frictional pressure losses of each cell are added to obtain the casing standpipe pressure. Therefore, there will be a constraint, that is, the loss function:

[0063]

[0064] In the formula, t is the time after the start of injection and displacement, A is the cross-sectional area (casing or annulus), ρ is the density of the fluid, z is the distance of the unit body from the bottom of the well, and F is the external force received by the unit body. Ap is the cross-sectional area of the casing, m2; α is the well deviation angle, °. After derivation and simplification, it is as follows

[0065]

[0066] According to the displacement, inlet flow rate and outlet flow rate, the total mass of the drilling fluid discharged into the wellbore can be known. The mass of the drilling fluid in each cell can be calculated and summed up cumulatively to obtain this constraint, that is, the loss function:

[0067]

[0068] After derivation and simplification, it is

[0069]

[0070] When the computer program in the computer-readable medium in this embodiment is executed, the specific steps implemented are as follows:

[0071] After receiving the real-time data of the well site, the data processing module will format the data and regularize it into a data format that the frictional coefficient network model and the drilling fluid rheological parameter network model can accept, and transfer the regularized real-time data to the real-time calculation module and the core calculation module.

[0072] The real-time calculation module divides the wellbore structure into several cells according to the wellbore structure parameters;

[0073] The core calculation module calculates the drilling fluid parameters and pressure values at one or more other cells through the preset frictional coefficient network model and drilling fluid rheological parameter network model according to the drilling fluid parameters, pressure values and wellbore structure parameters measured at the first cell;

[0074] The online learning module corrects the preset frictional coefficient network model and drilling fluid rheological parameter network model according to the calculated drilling fluid parameters and pressure values at one or more cells and the preset loss function.

[0075] The core computing module calculates the drilling fluid parameters and pressure values at one or more other cells based on the measured drilling fluid parameters, pressure values, and wellbore structure parameters at the first cell, specifically including:

[0076] Construct the initial vector of the first cell (l 1 , v 1 , d h1 , d p1 , k 1 , T 1 , V 1 , ρ 1 , P 1 , H 1 );

[0077] Through the friction coefficient network model fd h1 , d p1 , K 1 , ρ 1, , v 1 = f 1 Calculate the friction coefficient f of the first cell 1 ;

[0078] Among them, l 1 is the length of the first cell, v 1 is the flow rate of the drilling fluid in the first cell, d h1 is the outer diameter of the annulus of the first cell, d p1 is the inner diameter of the annulus of the first cell, k 1 is the viscosity coefficient of the drilling fluid in the first cell, T 1 is the formation temperature in the first cell, V 1 is the mechanical rotation speed in the first cell, ρ 1 is the density of the drilling fluid in the first cell, P 1 is the pressure in the first cell, H 1 is the well depth at the first cell. Among them, the cell length, outer diameter of the annulus, inner diameter of the annulus, formation temperature, mechanical rotation speed, and pressure of the first cell can all be measured, and the flow rate of the drilling fluid can also be calculated through the displacement and casing size.

[0079] According to the length l 1 of the first cell, well depth H 1 , inclination angle, density ρ 1 of the drilling fluid, friction coefficient f 1 , and the physical information derivation equation PI(l 1 , ρ 1 , P 1 , H 1 ) = P 2, calculate the second pressure P at the second cell 2 ;

[0080] According to the liquid rheology network ρ 2 is the density of the drilling fluid at the second cell, and k 2 is the viscosity coefficient of the drilling fluid at the second cell;

[0081] And so on, construct the initial vector (l n , v n , d hn , d pn , k n , T n , V n , ρ n , P n , H n ) of the nth cell;

[0082] Through the friction coefficient network f(d hn , d pn , K n , ρ n , v n ) = f n obtain the friction coefficient f of the nth cell n ;

[0083] According to the length l n of the nth cell, well depth H n , inclination angle, drilling fluid density ρ n , friction coefficient f n , and the physical information derivation equation PI(l n , ρ n , P n , H n ) = P n+1 , calculate the pressure P at the (n + 1)th cell n+1；

[0084] According to the liquid rheology network ρ n+1 is the density of the drilling fluid at the (n + 1)th cell, and k n+1 is the viscosity coefficient of the drilling fluid at the (n + 1)th cell, n = 2, 3, 4, 5...

[0085] Finally, obtain the pressures and densities of all cells, substitute the pressures and densities of each cell into the loss function, and the online learning module corrects the friction coefficient network model and the drilling fluid rheological parameter network model.

[0086] After correction, repeat the above steps again. Finally, an accurate friction coefficient network model and a drilling fluid rheological parameter network model are obtained, and the bottom hole pressure is continuously calculated based on the above models and the measured drilling fluid parameters, pressure values, and wellbore structure.

[0087] This application also proposes a bottom hole pressure calculation system for drilling, including a processor and the computer-readable medium as described above.

[0088] The present invention also provides an intelligent drilling platform, including the bottom hole pressure calculation system for drilling as described above, and a logging instrument 10 electrically connected to the computer-readable medium.

[0089] The computer-readable medium, the bottom hole pressure calculation system for drilling, and the intelligent drilling platform provided by this application can optimize the calculated model in real time through the method of data and physical coupling. It not only complies with the constraints of traditional physical laws but also makes full use of the multi-source data on the well site, further improving the prediction accuracy of the model and enabling real-time calibration of the formation pressure, thus facilitating the regulation of the bottom hole pressure.

[0090] The above are only the preferred embodiments of this application. It should be noted that for those of ordinary skill in the art, without departing from the principle of this application, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of this application.

Claims

1. A computer readable medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the following steps are implemented: Dividing the wellbore structure into a number of cells according to wellbore structure parameters; According to the measured drilling fluid parameters, pressure values ​​and wellbore structure parameters at the first cell, the drilling fluid parameters and pressure values ​​at another one or more cells are calculated by using a preset friction coefficient network model and a drilling fluid rheological parameter network model; According to the calculated drilling fluid parameters and pressure values ​​at one or more cells and the preset loss function, the preset friction coefficient network model and the drilling fluid rheological parameter network model are modified.

2. The computer-readable medium according to claim 1, wherein: The step of calculating the drilling fluid parameters and pressure values ​​at another one or more cells by using a preset friction coefficient network model and a drilling fluid rheological parameter network model according to the measured drilling fluid parameters, pressure values ​​and wellbore structure parameters at the first cell comprises: Calculating the friction coefficient of the first cell according to the measured drilling fluid parameters, pressure values ​​and wellbore structure parameters of the first cell; Calculate the pressure at the second cell according to the friction coefficient of the first cell, the drilling fluid parameters, the wellbore structure parameters and the measured first pressure; The drilling fluid parameter at the second cell is calculated according to the pressure of the second cell and the drilling fluid parameter of the first cell.

3. The computer-readable medium according to claim 2, wherein: The step of calculating the friction coefficient of the first cell according to the measured drilling fluid parameters, pressure values ​​and wellbore structure parameters of the first cell comprises: Construct the initial vector (l1,v1,d h1 ,d p1 ,k1,T1,V1,ρ1,P1,H1); After the coefficient network model f(d h1 ,d p1 ,K1,ρ 1, v1)=f1 to calculate the friction coefficient f1 of the first unit cell; Where l1 is the length of the first cell, v1 is the drilling fluid velocity in the first cell, d h1 is the outer diameter of the annulus of the first unit cell, d p1 is the inner diameter of the annulus of the first cell, k1 is the viscosity coefficient of the drilling fluid in the first cell, T1 is the formation temperature in the first cell, V1 is the mechanical speed in the first cell, ρ1 is the drilling fluid density in the first cell, P1 is the pressure in the first cell, and H1 is the well depth of the first cell.

4. The computer-readable medium according to claim 3, wherein: The step of calculating the pressure at the second cell according to the friction coefficient of the first cell, the drilling fluid parameters, the wellbore structure parameters and the measured first pressure comprises: According to the length l1 of the first cell, the well depth H1, the inclination angle, the drilling fluid density ρ1, the friction coefficient f1, and the physical information, the equation PI(l1,ρ1,P1,H1)=P2 is derived to calculate the second pressure P2 at the second cell.

5. The computer-readable medium of claim 4, wherein: The step of calculating the drilling fluid parameter at the second cell according to the pressure of the second cell and the drilling fluid parameter of the first cell comprises: According to the liquid rheological network ρ2 is the drilling fluid density at the second unit cell, and k2 is the viscosity coefficient of the drilling fluid at the second unit cell.

6. The computer-readable medium of claim 5, wherein: The step of calculating the drilling fluid parameters and pressure values ​​at another one or more cells by using a preset friction coefficient network model and a drilling fluid rheological parameter network model according to the measured drilling fluid parameters, pressure values ​​and wellbore structure parameters at the first cell comprises: Construct the initial vector (l) of the nth cell n ,v n ,d hn ,d pn ,k n ,T n ,V n ,ρ n ,P n ,H n ); After the coefficient network f(d hn ,d pn ,K n ,ρ n, v n ) = f n Get the friction coefficient f of the nth cell n ; According to the length l of the nth cell n , Jing Shen H n , inclination angle, drilling fluid density ρ n , friction coefficient f n , and the physical information derivation equation PI(l n ,ρ n ,P n ,H n )=P n+1 , calculate the pressure P at the n+1th cell n+1 ; According to the liquid rheological network ρ n+1 is the drilling fluid density at the n+1th cell, k n+1 is the viscosity coefficient of the drilling fluid at the n+1th cell, n=2, 3, 4, 5...

7. The computer-readable medium of claim 6, wherein: When the computer program is executed by a processor, the following steps are implemented: Format the received data.

8. A system for calculating bottom hole pressure during drilling, comprising a processor and the computer readable medium as claimed in any one of claims 1 to 7.

9. An intelligent drilling platform, characterized in that: It comprises the drilling bottom hole pressure calculation system as claimed in claim 8, and a logging instrument electrically connected to the computer readable medium.

Images