Method for predicting tripping-in property of small casing in horizontal well and related equipment

Abstract

The invention provides a method and related equipment for predicting the tripping-in performance of a small casing in a horizontal well, a prediction model among a well track, a well depth structure, a small casing parameter and the tripping-in performance of the small casing is constructed through integrated application of a casing tripping-in simulation and machine learning technology, well information is directly obtained after the model is established, and the running-in performance of the small casing in the horizontal well is predicted. Well information is input into the model, small casing running feasibility analysis can be rapidly carried out, the problem of small casing running feasibility analysis can be rapidly solved, the small casing well cementation shaft reconstruction technology is guaranteed, an XGBoost gradient boosting decision tree prediction model is used for training a data set, and compared with a BP artificial neural network and an SVM support vector machine algorithm, the method has the advantages that the method is simple and convenient to operate, and operation is easy and convenient. Compared with the prior art, the method has the advantages that the capability of processing complex nonlinear problems is higher, after algorithm parameters are optimized and adjusted, the prediction accuracy can reach 95% or above, and the accuracy of casing tripping-in prediction is further improved.

Description

Technical Field

[0001] The invention belongs to the technical field of overall repeated transformation of old oilfield blocks, and specifically relates to a method for predicting the insertability of a small casing in a horizontal well and related equipment. Background Art

[0002] Small casing cementing wellbore re-creation volume re-pressurization technology has become one of the key technical means of repeated fracturing. For horizontal wells with different wellbore trajectories and wellbore conditions, it is ensured that small casings of different sizes can be smoothly lowered into the existing 5 1 / 2 “The horizontal section of the wellbore is the key to wellbore reconstruction technology and the prerequisite for wellbore reconstruction volume re-pressurization technology;

[0003] However, the analysis of the insertability of a small casing in an old casing mostly relies on manual judgment, which results in large errors and low efficiency. Summary of the invention

[0004] The invention provides a method for predicting the runnability of a small casing in a horizontal well and related equipment, which solves the problem of large error and low efficiency in the runnability analysis of a small casing in an old casing.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A method for predicting the runnability of a small casing in a horizontal well, comprising:

[0007] Get well information;

[0008] Inputting the well information into a pre-trained small casing runnability analysis model to perform runnability analysis;

[0009] Among them, the small casing insertability analysis model is a insertability analysis model that has been trained and tested, and the small casing insertability analysis model is based on the XGBoost gradient boosting decision tree artificial intelligence algorithm.

[0010] Preferably, the small casing runnability analysis model training method is specifically as follows:

[0011] Construct a small casing running data sample set;

[0012] Based on the established data sample set, artificial neural network, support vector machine, gradient boosting decision tree and convolutional neural network are used to carry out data training and prediction to generate a small casing runnability analysis model.

[0013] Preferably, the construction of the small casing running data sample set specifically comprises:

[0014] Based on the horizontal wellbore trajectory and well depth structure characteristics of the target block, the Gaussian data generation idea is used to generate n groups of horizontal wellbore trajectory data sets, and small casing data sets of different sizes and lengths are generated at the same time. On this basis, a data interface is written and the software is called to complete the feasibility analysis simulation of n groups of examples, and the input data and analysis results are matched to generate a small casing feasibility sample set.

[0015] Preferably, the XGBoost gradient boosting decision tree artificial intelligence algorithm adopts a step-by-step parameter adjustment strategy to complete algorithm optimization and find the optimal parameter combination.

[0016] Preferably, the specific steps of algorithm optimization are:

[0017] According to conventional experience, a set of initial parameters is selected, and the number of decision trees is set to 50. On this basis, the depth of the decision tree and the node weight, i.e., the regularization coefficient, are adjusted;

[0018] Set different gamma coefficients;

[0019] Adjust the sample collection method;

[0020] Adjust the learning rate and compare the loss function to complete the learning rate parameter tuning.

[0021] Preferably, the sample sampling method of the adjustment is specifically based on column sampling and row sampling, and a heat map of the loss function is drawn according to the column sampling and the row sampling to find the best parameter combination.

[0022] Preferably, the algorithm optimization results are: decision tree depth is 6, regularization coefficient is 6, column sampling parameter is 0.8, row sampling parameter is 0.6, learning rate is 0.4, and the damage function of the optimization process changes with the parameters as follows: Figure 7 , the correlation coefficient R2 between the final prediction result and the true value is 0.9212.

[0023] A system for predicting the runnability of a small casing in a horizontal well, comprising:

[0024] Data acquisition module: used to obtain well information;

[0025] Analysis module: used for inputting the well information into a pre-trained small casing runnability analysis model;

[0026] Among them, the small casing insertability analysis model is a insertability analysis model that has been trained and tested, and the small casing insertability analysis model is based on the XGBoost gradient boosting decision tree artificial intelligence algorithm.

[0027] A computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of a method for predicting the runnability of a small casing in a horizontal well are implemented.

[0028] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a method for predicting the runnability of a small casing in a horizontal well.

[0029] Compared with the prior art, the present invention has the following beneficial effects: the present invention provides a method and related equipment for predicting the insertability of small casing in a horizontal well. By integrating and applying "casing lowering simulation + machine learning technology", a prediction model between wellbore trajectory, well depth structure, small casing parameters and small casing insertability is constructed. After the model is established, the well information can be directly obtained, and the insertability analysis can be quickly performed by inputting the well information into the model. The present invention can quickly solve the problem of feasibility analysis of small casing lowering, thereby ensuring the small casing cementing wellbore reconstruction technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 A flow chart of a method for predicting the runnability of a small casing in a horizontal well according to the present invention;

[0031] Figure 2 A system block diagram of the present invention for predicting the runnability of a small casing in a horizontal well;

[0032] Figure 3 Optimized values ​​of eta and gamma parameters in the embodiment of the present invention;

[0033] Figure 4 The max_depth and min_child_weight parameters of the embodiment of the present invention are combined and optimized;

[0034] Figure 5 The colsample_bytree and subsample parameter combination optimization of the embodiment of the present invention;

[0035] Figure 6 The present invention is a flowchart of a method for predicting the runnability of a small casing in a horizontal well according to an embodiment of the present invention.

[0036] Figure 7 The present invention is a software analysis flow chart for realizing the feasibility of small-casing lowering by utilizing drilling engineering software. DETAILED DESCRIPTION

[0037] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Components of the embodiments of the present invention usually described and illustrated in the drawings here can be arranged and designed in various different configurations.

[0038] Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention.

[0039] It should be noted that like reference numerals and letters denote like items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0040] In the description of the embodiments of the present invention, it should be noted that if terms such as "upper", "lower", "horizontal", "inner", etc. indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, or the orientation or positional relationship in which the inventive product is usually placed during use, it is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus cannot be construed as a limitation of the present invention. In addition, terms such as "first", "second", etc. are only used for descriptive distinction and cannot be construed as indicating or implying relative importance.

[0041] In addition, if the term "horizontal" appears, it does not mean that the component is required to be absolutely horizontal, but it can be slightly inclined. For example, "horizontal" only means that its direction is more horizontal relative to "vertical", and does not mean that the structure must be completely horizontal, but it can be slightly inclined.

[0042] In the description of the embodiments of the present invention, it should also be noted that unless otherwise clearly specified and limited, if terms such as "set", "installed", "connected", "connected" are understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0043] To enable those skilled in the art to better understand the technical solutions of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings.

[0044] like Figure 1 The present invention provides a method for predicting the runnability of a small casing in a horizontal well, comprising:

[0045] S101 obtains well information;

[0046] S102 inputs the well information into a pre-trained small casing runnability analysis model;

[0047] Among them, the small casing insertability analysis model is a insertability analysis model that has been trained and tested, and the small casing insertability analysis model is based on the XGBoost gradient boosting decision tree artificial intelligence algorithm.

[0048] The specific training method of the small casing runnability analysis model is as follows:

[0049] Construct a small casing running data sample set;

[0050] Based on the established data sample set, artificial neural network, support vector machine, gradient boosting decision tree and convolutional neural network are used to carry out data training and prediction to generate a small casing runnability analysis model.

[0051] The construction of the small casing running data sample set is specifically as follows:

[0052] Based on the horizontal wellbore trajectory and well depth structure characteristics of the target block, the Gaussian data generation idea is used to generate n groups of horizontal wellbore trajectory data sets, and small casing data sets of different sizes and lengths are generated at the same time. On this basis, a data interface is written and the software is called to complete the feasibility analysis simulation of n groups of examples, and the input data and analysis results are matched to generate a small casing feasibility sample set.

[0053] The XGBoost gradient boosting decision tree artificial intelligence algorithm adopts a step-by-step parameter adjustment strategy to complete algorithm optimization and find the optimal parameter combination.

[0054] The specific steps of algorithm optimization are:

[0055] According to conventional experience, a set of initial parameters is selected, and the number of decision trees is set to 50. On this basis, the depth of the decision tree and the node weight, i.e., the regularization coefficient, are adjusted;

[0056] Set different gamma coefficients;

[0057] Adjust the sample collection method;

[0058] Adjust the learning rate and compare the loss function to complete the learning rate parameter tuning.

[0059] The sample sampling method for adjusting the sample is specifically based on column sampling and row sampling, and a heat map of the loss function is drawn according to the column sampling and row sampling to find the best parameter combination.

[0060] The optimization results of the algorithm are: the decision tree depth is 6, the regularization coefficient is 6, the column sampling parameter is 0.8, the row sampling parameter is 0.6, and the learning rate is 0.4. The damage function of the optimization process changes with the parameters as follows Figure 7 , the correlation coefficient R between the final prediction result and the true value 2 It is 0.9212.

[0061] Example:

[0062] The present invention provides a method for predicting the runnability of a small casing in a horizontal well. Figure 6 FIG. 1 is a flow chart of a method according to an embodiment of the present invention. Figure 6 The present invention will be described in detail.

[0063] Step 1) establishing a feasibility analysis model for small-casing drilling based on drilling engineering analysis software;

[0064] Step 2) Generate n groups of data samples to write a software interface and construct a small set of downloaded data samples;

[0065] Step 3) Generate a small set entry feasibility proxy model based on artificial intelligence algorithm;

[0066] Step 4) Input the data of any well and quickly analyze and determine the feasibility of small casing installation.

[0067] The specific process is as follows:

[0068] In step 1, a feasibility analysis model for small casing lowering is established using drilling engineering analysis software. The wellbore trajectory data, well depth structure data, small casing size and lowering length, friction coefficient between double-layer casing and other parameters of the target well are input. The software is run to obtain the effective stress, friction and hook load line during the lifting and lowering process, which are compared with the safety line of bending and damage to determine whether the small casing can be safely lowered.

[0069] In step 2, based on the horizontal wellbore trajectory and well depth structure characteristics of the target block, a Gaussian data generator is used to design and generate n groups of horizontal wellbore trajectory data sets, and small casing data sets of different sizes and lengths are generated at the same time. On this basis, an intelligent interface between sample data and software operation is written to automatically call the software to complete the feasibility analysis simulation of n groups of examples, correspond the input data and analysis results, and generate a small casing feasibility database.

[0070] In step 3, based on the established database, the XGBoost gradient boosting decision tree artificial intelligence algorithm is used to carry out data training and prediction, and a feasibility prediction proxy model for small casing running is generated.

[0071] For the proxy model based on the XGBoost algorithm described in step 3, analyzing its algorithm principle, the optimization tuning process involves multiple parameter combinations. If the conventional grid search optimization method is used to optimize by traversing all parameters, the workload of the method prediction process is huge. The present invention designs to adopt a strategy of step-by-step parameter tuning to complete the algorithm optimization and find the optimal parameter combination. The specific steps for parameter adjustment are as follows:

[0072] (1) Select a set of initial parameters according to conventional experience, set the number of decision trees to 50, and on this basis, adjust the depth of the decision tree (max_depth) and the node weight, that is, the regularization coefficient (min_child_weight). The depth of the decision tree and the regularization coefficient determine the complexity of the tree. By using the depth of the tree and the regularization coefficient to draw the heat map of the loss function, the optimal tree parameter combination can be found;

[0073] (2) Set different gamma coefficients. This parameter determines when the loss function splits. The smaller the parameter, the lower the risk of overfitting. Therefore, gamma should be taken as small as possible on the premise of ensuring a reasonable loss function;

[0074] (3) Adjust the sample sampling method, mainly involving column sampling (colsample_bytree) and row sampling (subsample). Similarly, by using these two parameters to draw the heat map of the loss function, the best parameter combination can be found;

[0075] (4) Adjust the learning rate eta, and complete the tuning of the eta parameter by comparing the loss function.

[0076] Finally, it is optimized to a decision tree depth of 6, a regularization coefficient of 6, a column sampling parameter of 0.8, a row sampling parameter of 0.6, and a learning rate (eta) of 0.4. The variation of the damage function with parameters during the optimization process is as Figure 7 , and the correlation coefficient R between the final prediction result and the true value 2 is 0.9212.

[0077] In step 4, based on the established proxy model, by inputting the relevant data of any well, it can be quickly predicted whether the small casing can be safely run in.

[0078] Through this method, the feasibility analysis of small casing running in more than 40 horizontal wells indoors has been completed. The actual situation of small casing running in the field wells shows that the small casings of the wells analyzed through prediction are run in smoothly. This method has important guiding significance for the wellbore reconstruction and repeated fracturing of small casing cementing in horizontal wells.

[0079] Such asFigure 2 As shown, the present invention also provides a system for predicting the runnability of a small casing in a horizontal well, comprising:

[0080] Data acquisition module: used to obtain well information;

[0081] Analysis module: used for inputting the well information into a pre-trained small casing runnability analysis model;

[0082] Among them, the small casing insertability analysis model is a insertability analysis model that has been trained and tested, and the small casing insertability analysis model is based on the XGBoost gradient boosting decision tree artificial intelligence algorithm.

[0083] A terminal device is provided in one embodiment of the present invention. The terminal device of this embodiment includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps in the above-mentioned method embodiments are implemented. Alternatively, when the processor executes the computer program, the functions of the modules / units in the above-mentioned device embodiments are implemented.

[0084] The computer program may be divided into one or more modules / units, and the one or more modules / units are stored in the memory and executed by the processor to accomplish the present invention.

[0085] The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0086] The processor can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0087] The memory may be used to store the computer programs and / or modules, and the processor implements various functions of the terminal device by running or executing the computer programs and / or modules stored in the memory and calling the data stored in the memory.

[0088] If the modules / units integrated in the terminal device are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, to implement all or part of the processes in the above-described embodiment methods of the present invention, it can also be completed by a computer program instructing relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-described various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0089] Although the embodiments of the present invention have been described above in conjunction with the accompanying drawings, the present invention is not limited to the above specific embodiments and application fields. The above specific embodiments are merely illustrative and guiding, rather than restrictive. Under the inspiration of the specification, those of ordinary skill in the art can also make many forms without departing from the scope protected by the claims of the present invention, and these all fall within the scope of protection of the present invention.

Claims

1. A method for predicting the runnability of small casing in horizontal wells. It is characterized in that include: Get well information; Inputting the well information into a pre-trained small casing runnability analysis model to perform runnability analysis; Among them, the small casing insertability analysis model is a insertability analysis model that has been trained and tested, and the small casing insertability analysis model is based on the XGBoost gradient boosting decision tree artificial intelligence algorithm.

2. A method for predicting the runnability of a small casing in a horizontal well according to claim 1, It is characterized in that The specific training method of the small casing runnability analysis model is as follows: Construct a small casing running data sample set; Based on the established data sample set, artificial neural network, support vector machine, gradient boosting decision tree and convolutional neural network are used to carry out data training and prediction to generate a small casing runnability analysis model.

3. A method for predicting the runnability of a small casing in a horizontal well according to claim 2, It is characterized in that The construction of the small casing running data sample set is specifically as follows: Based on the horizontal wellbore trajectory and well depth structure characteristics of the target block, the Gaussian data generation idea is used to generate n groups of horizontal wellbore trajectory data sets, and small casing data sets of different sizes and lengths are generated at the same time. On this basis, a data interface is written and the software is called to complete the feasibility analysis simulation of n groups of examples, and the input data and analysis results are matched to generate a small casing feasibility sample set.

4. A method for predicting the runnability of a small casing in a horizontal well according to claim 1, It is characterized in that The XGBoost gradient boosting decision tree artificial intelligence algorithm adopts a step-by-step parameter adjustment strategy to complete algorithm optimization and find the optimal parameter combination.

5. A method for predicting the runnability of a small casing in a horizontal well according to claim 4, It is characterized in that The specific steps of algorithm optimization are: According to conventional experience, a set of initial parameters is selected, and the number of decision trees is set to 50. On this basis, the depth of the decision tree and the node weight, i.e., the regularization coefficient, are adjusted; Set different gamma coefficients; Adjust the sample collection method; Adjust the learning rate and compare the loss function to complete the learning rate parameter tuning.

6. A method for predicting the runnability of a small casing in a horizontal well according to claim 5, It is characterized in that The sample sampling method for adjusting the sample is specifically based on column sampling and row sampling, and a heat map of the loss function is drawn according to the column sampling and row sampling to find the best parameter combination.

7. A method for predicting the runnability of a small casing in a horizontal well according to claim 5, It is characterized in that The optimization results of the algorithm are as follows: the depth of the decision tree is 6, the regularization coefficient is 6, the column sampling parameter is 0.8, the row sampling parameter is 0.6, and the learning rate is 0.

4. The variation of the loss function with parameters during the optimization process is shown in Figure 7. The correlation coefficient R 2 between the final prediction result and the true value is 0.9212.

8. A system for predicting the runnability of small casing in horizontal wells. It is characterized in that include: Data acquisition module: used to obtain well information; Analysis module: used for inputting the well information into a pre-trained small casing runnability analysis model; Among them, the small casing insertability analysis model is a insertability analysis model that has been trained and tested, and the small casing insertability analysis model is based on the XGBoost gradient boosting decision tree artificial intelligence algorithm.

9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, It is characterized in that When the processor executes the computer program, the steps of a method for predicting the insertability of a small casing in a horizontal well as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program. It is characterized in that When the computer program is executed by a processor, the steps of a method for predicting the insertability of a small casing in a horizontal well as described in any one of claims 1 to 7 are implemented.

Images