Well leakage identification method and device, electronic equipment and storage medium
By acquiring and processing time series data during drilling, calculating feature parameters and inputting machine learning models, the problem of low accuracy of drilling well leakage recognition is solved, and higher recognition accuracy and risk control is achieved.
Patent Information
- Application Number
- CN202310761203.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-06-26
- Publication Date
- 2024-12-27
AI Technical Summary
The prior art has low accuracy in the identification of drilling well leakage, making it difficult to effectively apply to different regions and drilling rig environments.
By obtaining time series measurement data during drilling, the dispersion of characteristic parameters such as d-index, pressure calibration flow gain, moving average dispersion and moving slope average are calculated, the characteristic parameter mixing matrix is constructed, and input it to the pre-trained machine Learn the model to identify well leakage to improve the recognition accuracy.
It improves the accuracy of well leakage identification, can more effectively identify drilling well leakage, and reduces the risk costs in drilling projects.
Smart Images
Figure CN119195750A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of oil drilling engineering, and in particular to a method, device, electronic equipment and storage medium for identifying well leakage. Background Art
[0002] Well leakage refers to a complex underground situation in which various working fluids (including drilling fluid, cement slurry, completion fluid and other fluids) directly enter the formation under the action of pressure difference during underground operations such as drilling, cementing, testing or well repair. Well leakage includes permeability loss, fracture loss and cave loss. The main reasons include: 1) There are natural leakage channels in the drilled formation, such as high permeability formation, fractured formation and cave formation; poor drilling fluid performance or improper operation will also cause artificial leakage channels. 2) The pressure of the drilled formation is deficient, or the density of the drilling fluid is too high, resulting in a large pressure difference. 3) The viscosity and shear force of the drilling fluid are too large, resulting in excessive pump pressure, resulting in pressure shock and leakage in the formation. 4) The drilling fluid has poor sand carrying performance, the well wall is not clean, or the water loss is too large, the filter cake is thick, and improper operations such as drilling and pumping will cause pressure shock. During the drilling construction process at the drilling site, identifying and warning of well leakage through real-time (second) data from the well site is an extremely important part of drilling project risk prevention and control, and is also an important factor in reducing the risk cost of drilling projects. Summary of the invention
[0003] In order to solve the problems in the above-mentioned related technologies, the present application provides a method, device, electronic device and storage medium for identifying well leakage, which can improve the accuracy of well leakage identification.
[0004] The present application provides a method for identifying lost circulation, including:
[0005] Obtaining time series measurement data of the target well during the drilling process;
[0006] Calculating characteristic parameters based on the measurement data, wherein the characteristic parameters are used to characterize a change trend of the measurement data based on a time series;
[0007] Constructing a time series-based characteristic parameter mixing matrix based on the characteristic parameters and the measurement data;
[0008] The characteristic parameter mixing matrix is input into a pre-established machine learning model to identify lost circulation in the target well.
[0009] In some embodiments, the method further comprises:
[0010] Acquire initial sample measurement data based on time series during the drilling process of the sample well;
[0011] Determine sample measurement data from initial sample measurement data based on principal component analysis or Pearson correlation analysis;
[0012] Calculating sample characteristic parameters based on the sample measurement data;
[0013] Determine a time series-based sample characteristic parameter mixing matrix based on the sample characteristic parameters and the sample measurement data;
[0014] Adding a leakage label to the sample characteristic parameter mixing matrix to obtain training data;
[0015] The machine learning model is obtained by training the initial model based on the training data.
[0016] In some embodiments, the initial model is a gradient boosting tree model, and determining a time series-based sample feature parameter mixing matrix based on the sample feature parameters and the sample measurement data includes:
[0017] The sample characteristic parameters and the sample measurement data are combined based on the time points in the time series to obtain a sample characteristic parameter mixing matrix based on the time series.
[0018] In some embodiments, the measurement data includes: mechanical drilling speed of the drill bit, the rotation speed of the drill bit, the drill bit diameter, the drilling pressure, the inlet flow rate and outlet flow rate of the drilling fluid, and the standpipe pressure; the characteristic parameters include: d index, pressure calibration flow gain, moving average dispersion and moving slope average dispersion; the characteristic parameters are calculated based on the measurement data, including:
[0019] The d index is calculated based on the mechanical drilling speed, the rotation speed, the diameter and the drilling pressure of the drill bit, where the d index is a value describing the relationship between the drilling speed and the engineering parameters;
[0020] Calculate the pressure calibration flow gain based on the standpipe pressure, the outlet flow rate and the inlet flow rate of the drilling fluid;
[0021] Based on the d index, the moving average dispersion of the d index is obtained by subtracting the long and short sliding time windows;
[0022] The dispersion of the moving slope average is calculated based on the moving average dispersion.
[0023] In some embodiments, the calculating of the d index based on the mechanical drilling speed, the rotation speed, the diameter and the bit pressure of the drill bit includes:
[0024] The machine speed, the drill bit speed, the drill bit diameter, and the drilling pressure are input into a first calculation formula to calculate the d index, wherein the first calculation formula includes:
[0025]
[0026] Among them, ROP is the mechanical drilling speed, N is the drill bit speed, WOB is the drilling pressure, D bit is the drill bit diameter.
[0027] In some embodiments, the step of calculating the pressure calibration flow gain based on the standpipe pressure, the outlet flow rate and the inlet flow rate of the drilling fluid comprises:
[0028] The standpipe pressure, the outlet flow rate and the inlet flow rate of the drilling fluid are input into a second calculation formula to calculate the pressure calibration flow gain, wherein the second calculation formula includes:
[0029]
[0030] Where FPG is the pressure calibration flow gain, SPP is the standpipe pressure, FlowIn is the drilling fluid inlet flow, FlowOut is the drilling fluid outlet flow, and P ref is the standard reference pressure constant, c r is the compression ratio of drilling fluid.
[0031] In some embodiments, the step of obtaining the moving average dispersion of the d index by smoothing and subtracting a long window and a short window based on the d index includes:
[0032] Determine a moving average of the d index; calculate the moving average dispersion using a third calculation formula based on the long and short time windows and the moving average, wherein the third calculation formula includes:
[0033] DMA t =MA t ,w1-MA t ,w2(w1 <w2);
[0034] Among them, MA t It is the moving average value at a specific time in the time series, w1 is a short time window, and w2 is a long time window; DMA t is the moving average dispersion;
[0035] The step of calculating the dispersion of the moving slope average value based on the moving average dispersion comprises:
[0036] The dispersion of the moving slope average value is calculated using a fourth calculation formula based on the dispersion of the moving average value, wherein the fourth calculation formula includes:
[0037] DMSA t =MSA t ,w1-MSA t ,w2(w1 <w2);
[0038] Among them, DMSA tis the dispersion of the moving slope average in a specific time window, MSA t It is the moving slope average of the moving average dispersion in a specific time window.
[0039] The present application provides a device for identifying a lost circulation, including:
[0040] An acquisition module, used to acquire measurement data of a target well based on a time series during the drilling process;
[0041] A calculation module, used for calculating characteristic parameters based on the measurement data, wherein the characteristic parameters are used for characterizing the change trend of the measurement data based on the time series;
[0042] A construction module, used for constructing a time series-based characteristic parameter mixing matrix based on the characteristic parameters and the measurement data;
[0043] The identification module is used to input the characteristic parameter mixing matrix into a pre-established machine learning model to identify the well leakage of the target well.
[0044] An embodiment of the present application provides an electronic device, including:
[0045] A memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, any one of the above-mentioned methods for identifying lost circulation is executed.
[0046] An embodiment of the present application provides a storage medium, which stores a computer program that can be executed by one or more processors and can be used to implement any of the above-mentioned methods for identifying lost circulation.
[0047] The present application provides a method, device, electronic device and storage medium for identifying well leakage, which obtains measurement data of a target well based on a time series during drilling; calculates characteristic parameters based on the measurement data, wherein the characteristic parameters are used to characterize the changing trend of the measurement data based on the time series; constructs a characteristic parameter mixing matrix based on the time series based on the characteristic parameters and the measurement data; and inputs the characteristic parameter mixing matrix into a pre-established machine learning model to identify well leakage in the target well, thereby realizing the identification of well leakage and improving the accuracy of well leakage identification. BRIEF DESCRIPTION OF THE DRAWINGS
[0048] Hereinafter, the present application will be described in more detail based on embodiments and with reference to the accompanying drawings.
[0049] Figure 1 A schematic diagram of a flow chart of a method for identifying a lost circulation well provided in an embodiment of the present application;
[0050] Figure 2A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.
[0051] In the drawings, the same reference numerals are used for the same components, and the drawings are not drawn to scale. DETAILED DESCRIPTION
[0052] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.
[0053] In the following description, reference is made to “some embodiments”, which describe a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.
[0054] If similar descriptions of "first\second\third" appear in the application documents, the following instructions will be added. In the following description, the terms "first\second\third" involved are merely used to distinguish similar objects and do not represent a specific ordering of the objects. It can be understood that "first\second\third" can be interchanged in a specific order or sequence where permitted, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein.
[0055] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.
[0056] Before introducing the embodiments of the present application, a method for identifying lost circulation in the related art is briefly introduced:
[0057] For a long time, the real-time prediction of drilling leakage has mainly focused on the characteristics of leakage in different regions, establishing the identification and discrimination rules of leakage, and determining whether leakage has occurred through the rules. This method is difficult to be applied on a large scale due to problems such as "low discrimination success rate" and "large differences in discrimination rules in different regions". Therefore, in recent years, a large number of studies have begun to turn to the research of methods based on big data and machine learning. This kind of research uses a large amount of leakage case data and fits the classification model established by machine learning to avoid the mechanical nature brought by the rule-based model. Therefore, it can establish more targeted machine learning rules according to the differences in different regions and drilling rigs.
[0058] For example, in order to solve the shortcomings of manual monitoring of well leakage accidents in terms of timeliness and accuracy, the relevant technology uses logging observation logs and automatic logging data as data support, selects and constructs reasonable data features through the study of engineering logging parameters, and establishes a well leakage accident warning model based on the XGBoost algorithm. However, only the real-time parameters of comprehensive logging are used in this method.
[0059] In the related technology, in the well leakage risk warning technology based on neural network, the phenomenon of well leakage occurrence and the parameters required for well leakage risk warning were explored. In addition, a neural network warning model with self-learning properties was established using a neural network algorithm, and the model was tested using field data. However, the five comprehensive logging parameters used in this method, such as standpipe pressure, total pool volume, drilling time, inlet and outlet flow, and temperature, and two geological parameters, such as rock formation and cuttings characterization, are not included.
[0060] Related technologies also propose a method for predicting well leakage based on support vector regression, which is related to parameters such as drilling fluid type, rheological parameters and density, as well as formation characteristics, including in situ stress, pore pressure, rock strength and primary fractures. Support vector regression is proposed as a machine learning model to integrate a large amount of valid data that causes well leakage, thereby building an effective early warning system for well leakage.
[0061] In summary, the above-mentioned real-time prediction and warning methods for well leakage focus more on the selection of different machine learning methods. In the selection of relevant parameters of data samples, they directly use the data collected and measured at the well site, and lack the secondary processing of data based on the calculation method of drilling engineering. Based on research, in the process of applying machine learning to predict drilling well leakage, the application of various conventional machine learning methods is not much different, and the most effective machine learning and even deep learning methods such as decision trees, random forests, support vector machines, BP neural networks, XGBOOST, etc., under the conditions of the same characteristic parameters and total sample volume, the difference in prediction accuracy is already very small. Therefore, in order to further improve the prediction accuracy, a new method needs to be provided.
[0062] Based on the problems existing in the related art, an embodiment of the present application provides a method for identifying well leakage, which is applied to an electronic device, which may be a mobile terminal, a computer, a server, etc. The electronic device may be a controller of a mobile terminal, a computer, etc.
[0063] The functions implemented by the well leakage identification method provided in the embodiment of the present application can be implemented by calling a program code by a processor of an electronic device, wherein the program code can be stored in a computer storage medium.
[0064] The present application provides a method for identifying a lost circulation. Figure 1A schematic diagram of the implementation flow of a method for identifying a lost circulation well provided in an embodiment of the present application is shown in FIG. Figure 1 As shown, including:
[0065] Step S1, obtaining measurement data of a target well based on a time series during the drilling process.
[0066] In the embodiment of the present application, the target well can be any well, and the measurement data is the data that can be directly measured by the instrument during the drilling process of the target well. The measurement data may include: mechanical drilling speed, rotation speed, drill bit diameter, drilling pressure, standpipe pressure, drilling fluid outlet flow rate and inlet flow rate, etc.
[0067] In an embodiment of the present application, the electronic device can be connected to a measuring instrument for communication, and the measuring instrument is used to obtain the measurement data of the well based on the time series during the drilling process. In some embodiments, the electronic device can obtain the measurement data based on the time series during the drilling process from the Internet.
[0068] Step S2, calculating characteristic parameters based on the measurement data, wherein the characteristic parameters are used to characterize the change trend of the measurement data based on a time series.
[0069] In the embodiment of the present application, the characteristic parameters may include: d index, pressure-calibrated flow gain, dispersion of moving average value and dispersion of moving slope average value.
[0070] In an embodiment of the present application, the d index can be calculated based on the mechanical drilling speed, rotation speed, diameter and drilling pressure of the drill bit, wherein the d index is a value describing the relationship between the drilling speed and the engineering parameters; the pressure calibration flow gain is calculated based on the riser pressure, the outlet flow rate and the inlet flow rate of the drilling fluid; based on the d index, the moving average dispersion of the d index is obtained by subtracting the long and short sliding time windows; and the dispersion of the moving slope average is calculated based on the moving average dispersion.
[0071] Step S3: constructing a time series-based characteristic parameter mixing matrix based on the characteristic parameters and the measurement data.
[0072] In the embodiment of the present application, the characteristic parameters can be constructed into a characteristic parameter matrix based on the time points, and then the characteristic parameters and the measurement data are merged based on the time points in the time series to obtain a characteristic parameter mixed matrix based on the time series.
[0073] Step S4, inputting the characteristic parameter mixing matrix into a pre-established machine learning model to identify lost circulation in the target well.
[0074] In the embodiment of the present application, the machine learning model is obtained by training with sample data. The machine learning model may be a gradient boosting tree model.
[0075] In the embodiment of the present application, the input of the machine learning model is a feature parameter mixing matrix, and the output of the machine learning is a lost circulation label. The lost circulation label is output to identify the target well for lost circulation identification.
[0076] The present application provides a method for identifying well leakage, which obtains measurement data based on a time series of a target well during drilling; calculates characteristic parameters based on the measurement data, wherein the characteristic parameters are used to characterize a change trend of the measurement data based on the time series; constructs a characteristic parameter mixing matrix based on the time series based on the characteristic parameters and the measurement data; and inputs the characteristic parameter mixing matrix into a pre-established machine learning model to identify well leakage in the target well, thereby realizing the identification of well leakage and improving the accuracy of well leakage identification.
[0077] In some embodiments, step S2 may be implemented by the following steps:
[0078] Step S21, calculating the d index based on the mechanical drilling speed, the rotation speed, the diameter and the bit pressure of the drill bit.
[0079] In the embodiments of the present application, DPG has been used in the oil and gas industry to identify abnormal pressure formations and detect abnormal pore pressure, namely, the d index.
[0080] In the embodiment of the present application, the machine speed, the drill speed, the drill diameter, and the drilling pressure are input into a first calculation formula to calculate the d index, wherein the first calculation formula includes:
[0081]
[0082] Among them, ROP is the mechanical drilling speed, N is the drill bit speed, WOB is the drilling pressure, D bit is the drill bit diameter.
[0083] In the embodiment of the present application, in a normal pressure formation, the d index shows an upward trend. When the drill bit encounters an overpressure formation, the d index will deviate from the original growth trend and follow a slow growth or decline trend.
[0084] Step S22, calculating the pressure calibration flow gain based on the standpipe pressure, the outlet flow rate and the inlet flow rate of the drilling fluid.
[0085] In the embodiment of the present application, the pressure-calibrated flow gain (FPG) is mainly affected by the density of the drilling fluid and the formation pressure. When a well leak occurs, the FPG will tend to decrease, indicating that the outflow is less than the inflow.
[0086] In the embodiment of the present application, the standpipe pressure, the outlet flow rate and the inlet flow rate of the drilling fluid may be input into a second calculation formula to calculate the pressure calibration flow gain, wherein the second calculation formula includes:
[0087]
[0088] Where FPG is the pressure calibration flow gain, SPP is the standpipe pressure, FlowIn is the drilling fluid inlet flow, FlowOut is the drilling fluid outlet flow, and P ref is the standard reference pressure constant, c r is the compression ratio of drilling fluid.
[0089] Step S23, based on the d index, the moving average dispersion of the d index is obtained by subtracting the long and short sliding time windows.
[0090] In the embodiment of the present application, the moving average of the d index can be determined; the moving average dispersion is calculated using a third calculation formula based on the long and short time windows and the moving average, wherein the third calculation formula includes:
[0091] DMA t =MA t ,w1-MA t ,w2(w1 <w2);
[0092] Among them, MA t It is the moving average value at a specific time in the time series, w1 is a short time window, and w2 is a long time window; DMA t is the moving average dispersion;
[0093] Step S24, calculating the dispersion of the moving slope average value based on the dispersion of the moving average value.
[0094] In the embodiment of the present application, the dispersion of the moving slope average value is calculated using a fourth calculation formula based on the dispersion of the moving average value, wherein the fourth calculation formula includes:
[0095] DMSA t =MSA t ,w1-MSA t ,w2(w1 <w2);
[0096] Among them, DMSA t is the dispersion of the moving slope average in a specific time window, MSA t It is the moving slope average of the moving average dispersion in a specific time window.
[0097] The method provided in the embodiment of the present application constructs a characteristic parameter mixing matrix through the d index, pressure-calibrated flow gain, moving average discreteness and moving slope average discreteness, and inputs the characteristic parameter mixing matrix into a pre-established machine learning model to identify well leakage in the target well, which can further enhance the measurement data and thus improve the accuracy of identification.
[0098] In some embodiments, before step S1, the method further includes:
[0099] Step S101, obtaining initial sample measurement data of a sample well based on a time series during the drilling process.
[0100] In an embodiment of the present application, the electronic device can be connected to the Internet for communication, and obtain initial sample measurement data of the sample well based on a time series during the drilling process from the Internet.
[0101] Step S102: determining sample measurement data from the initial sample measurement data based on principal component analysis or Pearson correlation analysis.
[0102] In the embodiment of the present application, measurement data with a greater correlation with well leakage can be selected to determine sample measurement data.
[0103] Step S103: calculating sample characteristic parameters based on the sample measurement data.
[0104] In the embodiment of the present application, the sample characteristic parameters may include: d index, pressure calibration flow gain, dispersion of moving average value and dispersion of moving slope average value.
[0105] In the embodiment of the present application, the calculation method of the exponent, the pressure-calibrated flow gain, the discreteness of the moving average value and the discreteness of the moving slope average value refers to the above method and will not be repeated here.
[0106] Step S104: determining a time series-based sample characteristic parameter mixing matrix based on the sample characteristic parameters and the sample measurement data.
[0107] In the embodiment of the present application, the sample characteristic parameters and the sample measurement data may be combined based on the time points in the time series to obtain a sample characteristic parameter mixing matrix based on the time series.
[0108] Exemplarily, the sample parameters in the sample feature parameter mixing matrix may include:
[0109] "RPM","BLOCKPOS","WOB","SPP","CUP","TORQUE","SPM1","SPM2","SPM3","TSPM","TPIT","GAINLOSSTOT","MTI","MTO","MCI","MCO","M WI","MWO","MFI","FLOWOUT","GAS","SPP_AVE","SPP_AVE_SHIFT5","TPIT_AVE","TPIT_AVE_SHIFT5","FLOWOUT_AVE","FLOWOUT_AVE_SHIFT 5","DPG","FPG","MA1","MA2","DMA","MA_DPG","MK_DPG","MKS_DPG","MKL_DPG","DMSA","PRED_DPG","PRED_FPG","AREA_DPG","AREA_FPG".
[0110] Step S105, adding a leakage label to the sample characteristic parameter mixing matrix to obtain training data.
[0111] In the embodiment of the present application, the sample characteristic parameter mixing matrix can be manually labeled with the leakage interval segments, thereby adding the leakage labels to obtain the training data.
[0112] Step S106: training the initial model based on the training data to obtain the machine learning model.
[0113] In an embodiment of the present application, an initial model can be trained using training data to obtain a machine learning model.
[0114] In an embodiment of the present application, the initial model is a gradient boosting tree model.
[0115] In the embodiment of the present application, considering the efficiency of the algorithm, we used the gradient boosting tree model (XGBoost), which has the fastest learning and training speed and the highest accuracy, as the main method for machine learning models.
[0116] In the embodiment of the present application, XGBoost is an optimized distributed gradient boosting library designed to be efficient, flexible and portable. It implements machine learning algorithms under the Gradient Boosting framework. XGBoost provides parallel tree boosting (also known as GBDT, GBM) that can quickly and accurately solve many data science problems. The same code runs on major distributed environments (Hadoop, SGE, MPI) and can solve problems beyond billions of examples.
[0117] From the algorithmic point of view, XGBoost is an improvement on the gradient boosting algorithm. It uses Newton's method to solve the extreme value of the loss function, expands the loss function Taylor to the second order, and adds a regularization term to the loss function. The objective function during training consists of two parts, the first part is the gradient boosting algorithm loss, and the second part is the regularization term.
[0118] A method for identifying well leakage provided in an embodiment of the present application constructs the characteristics of the changing trends of four time series, namely DPG, FPG, DMA and DMSA, from the perspective of characteristic parameter construction, and constructs the characteristic parameters of the samples through four key trend characteristics, thereby realizing machine learning based on the dynamic time series of the well site. The machine model obtained based on the training can effectively improve the accuracy of drilling well leakage prediction.
[0119] The prediction results of using feature parameters for leak prediction and not using feature parameters for leak prediction provided in the embodiments of the present application, in the case of a smaller sample, the method provided in the embodiments of the present application can effectively improve the accuracy of drilling leak prediction, and the final average prediction accuracy can reach 67% based on XGBoost training and prediction, which is higher than the accuracy without using features. In addition, the machine learning method with a smaller sample has a better prediction effect. The introduction of new feature parameters of XGBoost, especially the introduction of feature parameters, brings about an improvement in prediction effect.
[0120] Based on the foregoing embodiments, the embodiments of the present application provide a device for identifying well leakage, and the modules included in the device, as well as the units included in each module, can be implemented by a processor in a computer device; of course, they can also be implemented by a specific logic circuit; in the implementation process, the processor can be a central processing unit (CPU, Central Processing Unit), a microprocessor (MPU, Microprocessor Unit), a digital signal processor (DSP, Digital Signal Processing) or a field programmable gate array (FPGA, Field Programmable Gate Array), etc.
[0121] The present application provides a device for identifying a lost circulation well, the lost circulation well identifying device comprising:
[0122] An acquisition module, used to acquire measurement data of a target well based on a time series during the drilling process;
[0123] A calculation module, used for calculating characteristic parameters based on the measurement data, wherein the characteristic parameters are used for characterizing the change trend of the measurement data based on the time series;
[0124] A construction module, used for constructing a time series-based characteristic parameter mixing matrix based on the characteristic parameters and the measurement data;
[0125] The identification module is used to input the characteristic parameter mixing matrix into a pre-established machine learning model to identify the well leakage of the target well.
[0126] In some embodiments, the lost circulation identification device is further used for:
[0127] Acquire initial sample measurement data based on time series during the drilling process of the sample well;
[0128] Determine sample measurement data from initial sample measurement data based on principal component analysis or Pearson correlation analysis;
[0129] Calculating sample characteristic parameters based on the sample measurement data;
[0130] Determine a time series-based sample characteristic parameter mixing matrix based on the sample characteristic parameters and the sample measurement data;
[0131] Adding a leakage label to the sample characteristic parameter mixing matrix to obtain training data;
[0132] The machine learning model is obtained by training the initial model based on the training data.
[0133] In some embodiments, the initial model is a gradient boosting tree model, and determining a time series-based sample feature parameter mixing matrix based on the sample feature parameters and the sample measurement data includes:
[0134] The sample characteristic parameters and the sample measurement data are combined based on the time points in the time series to obtain a sample characteristic parameter mixing matrix based on the time series.
[0135] In some embodiments, in some embodiments, the measurement data includes: mechanical drilling speed of the drill bit, the rotation speed of the drill bit, the drill bit diameter, the drilling pressure, the inlet flow rate and outlet flow rate of the drilling fluid, and the standpipe pressure; the characteristic parameters include: d index, pressure calibration flow gain, moving average dispersion and moving slope average dispersion; the characteristic parameters are calculated based on the measurement data, including:
[0136] The d index is calculated based on the mechanical drilling speed, the rotation speed, the diameter and the drilling pressure of the drill bit, where the d index is a value describing the relationship between the drilling speed and the engineering parameters;
[0137] Calculate the pressure calibration flow gain based on the standpipe pressure, the outlet flow rate and the inlet flow rate of the drilling fluid;
[0138] Based on the d index, the moving average dispersion of the d index is obtained by subtracting the long and short sliding time windows;
[0139] The dispersion of the moving slope average is calculated based on the moving average dispersion.
[0140] In some embodiments, the calculating of the d index based on the mechanical drilling speed, the rotation speed, the diameter and the bit pressure of the drill bit includes:
[0141] The machine speed, the drill bit speed, the drill bit diameter, and the drilling pressure are input into a first calculation formula to calculate the d index, wherein the first calculation formula includes:
[0142]
[0143] Among them, ROP is the mechanical drilling speed, N is the drill bit speed, WOB is the drilling pressure, D bit is the drill bit diameter.
[0144] In some embodiments, the step of calculating the pressure calibration flow gain based on the standpipe pressure, the outlet flow rate and the inlet flow rate of the drilling fluid comprises:
[0145] The standpipe pressure, the outlet flow rate and the inlet flow rate of the drilling fluid are input into a second calculation formula to calculate the pressure calibration flow gain, wherein the second calculation formula includes:
[0146]
[0147] Where FPG is the pressure calibration flow gain, SPP is the standpipe pressure, FlowIn is the drilling fluid inlet flow, FlowOut is the drilling fluid outlet flow, and P ref is the standard reference pressure constant, c r is the compression ratio of drilling fluid.
[0148] In some embodiments, the step of obtaining the moving average dispersion of the d index by smoothing and subtracting a long window and a short window based on the d index includes:
[0149] Determine a moving average of the d index; calculate the moving average dispersion using a third calculation formula based on the long and short time windows and the moving average, wherein the third calculation formula includes:
[0150] DMA t =MA t ,w1-MA t ,w2(w1 <w2);
[0151] Among them, MA t It is the moving average value at a specific time in the time series, w1 is a short time window, and w2 is a long time window; DMA t is the moving average dispersion;
[0152] The step of calculating the dispersion of the moving slope average value based on the moving average dispersion comprises:
[0153] The dispersion of the moving slope average value is calculated using a fourth calculation formula based on the dispersion of the moving average value, wherein the fourth calculation formula includes:
[0154] DMSA t =MSA t ,w1-MSA t ,w2(w1 <w2);
[0155] Among them, DMSA t is the dispersion of the moving slope average in a specific time window, MSA t It is the moving slope average of the moving average dispersion in a specific time window.
[0156] It should be noted that, in the embodiment of the present application, if the above-mentioned well leakage identification method is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the embodiment of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a U disk, a mobile hard disk, a read-only memory (ROM, ReadOnly Memory), a disk or an optical disk. In this way, the embodiment of the present application is not limited to any specific combination of hardware and software.
[0157] Accordingly, an embodiment of the present application provides a storage medium on which a computer program is stored, characterized in that when the computer program is executed by a processor, the steps in the well leakage identification method provided in the above embodiment are implemented.
[0158] An embodiment of the present application provides an electronic device; Figure 2 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application is shown in FIG. Figure 2As shown, the electronic device 400 includes: a processor 401, at least one communication bus 402, a user interface 403, at least one external communication interface 404, and a memory 405. The communication bus 402 is configured to realize the connection and communication between these components. The user interface 403 may include a display screen, and the external communication interface 404 may include a standard wired interface and a wireless interface. The processor 401 is configured to execute the program of the well leakage identification method stored in the memory to implement the steps in the well leakage identification method provided in the above embodiment.
[0159] It should be noted here that the description of the above storage medium and electronic device embodiments is similar to the description of the above method embodiments, and has similar beneficial effects as the method embodiments. For technical details not disclosed in the storage medium and device embodiments of this application, please refer to the description of the method embodiments of this application for understanding.
[0160] It should be understood that "one embodiment" or "an embodiment" mentioned throughout the specification means that specific features, structures or characteristics related to the embodiment are included in at least one embodiment of the present application. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner. It should be understood that in various embodiments of the present application, the size of the sequence number of the above-mentioned processes does not mean the order of execution, and the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application. The above-mentioned sequence numbers of the embodiments of the present application are only for description and do not represent the advantages and disadvantages of the embodiments.
[0161] It should be noted that, in this article, the terms "include", "comprises" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, object or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, object or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, object or device including the element.
[0162] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.
[0163] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.
[0164] In addition, all functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may be a separate unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.
[0165] A person of ordinary skill in the art can understand that: all or part of the steps of implementing the above-mentioned method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above-mentioned method embodiment; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM, Read Only Memory), disks or optical disks, etc. Various media that can store program codes.
[0166] Alternatively, if the above-mentioned integrated unit of the present application is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application can essentially or in other words, the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for a controller to execute all or part of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROMs, magnetic disks, or optical disks.
[0167] The above is only an implementation method of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
Claims
1. A method for identifying lost circulation, characterized in that: The method comprises: Obtaining time series measurement data of the target well during the drilling process; Calculating characteristic parameters based on the measurement data, wherein the characteristic parameters are used to characterize a change trend of the measurement data based on a time series; Constructing a time series-based characteristic parameter mixing matrix based on the characteristic parameters and the measurement data; The characteristic parameter mixing matrix is input into a pre-established machine learning model to identify lost circulation in the target well.
2. The method according to claim 1, characterized in that: The method further comprises: Acquire initial sample measurement data based on time series during the drilling process of the sample well; Determine sample measurement data from initial sample measurement data based on principal component analysis or Pearson correlation analysis; Calculating sample characteristic parameters based on the sample measurement data; Determine a time series-based sample characteristic parameter mixing matrix based on the sample characteristic parameters and the sample measurement data; Adding a leakage label to the sample characteristic parameter mixing matrix to obtain training data; The machine learning model is obtained by training the initial model based on the training data.
3. The method according to claim 2, characterized in that The initial model is a gradient boosting tree model, and the determining of a sample feature parameter mixing matrix based on a time series based on the sample feature parameters and the sample measurement data includes: The sample characteristic parameters and the sample measurement data are combined based on the time points in the time series to obtain a sample characteristic parameter mixing matrix based on the time series.
4. The method according to claim 1, characterized in that The measurement data include: mechanical drilling speed of the drill bit, the rotation speed of the drill bit, the drill bit diameter, the drilling pressure, the inlet flow rate and outlet flow rate of the drilling fluid, and the standpipe pressure. The characteristic parameters include: d index, pressure calibration flow gain, moving average dispersion and moving slope average dispersion. The characteristic parameters calculated based on the measurement data include: The d index is calculated based on the mechanical drilling speed, the rotation speed, the diameter and the drilling pressure of the drill bit, where the d index is a value describing the relationship between the drilling speed and the engineering parameters; Calculate the pressure calibration flow gain based on the standpipe pressure, the outlet flow rate and the inlet flow rate of the drilling fluid; Based on the d index, the moving average dispersion of the d index is obtained by subtracting the long and short sliding time windows; The dispersion of the moving slope average is calculated based on the moving average dispersion.
5. The method according to claim 4, characterized in that The d index is calculated based on the mechanical drilling speed, the rotation speed, the diameter and the drilling pressure of the drill bit, including: The machine speed, the drill bit speed, the drill bit diameter, and the drilling pressure are input into a first calculation formula to calculate the d index, wherein the first calculation formula includes: Among them, ROP is the mechanical drilling speed, N is the drill bit speed, WOB is the drilling pressure, D bit is the drill bit diameter.
6. The method according to claim 4, characterized in that The method of calculating the pressure calibration flow gain based on the standpipe pressure, the outlet flow rate and the inlet flow rate of the drilling fluid comprises: The standpipe pressure, the outlet flow rate and the inlet flow rate of the drilling fluid are input into a second calculation formula to calculate the pressure calibration flow gain, wherein the second calculation formula includes: Where FPG is the pressure calibration flow gain, SPP is the standpipe pressure, FlowIn is the drilling fluid inlet flow, FlowOut is the drilling fluid outlet flow, and P ref is the standard reference pressure constant, c r is the compression ratio of drilling fluid.
7. The method according to claim 5, characterized in that The moving average dispersion of the d index obtained by subtracting the long and short sliding time windows based on the d index includes: Determine a moving average of the d index; calculate the moving average dispersion using a third calculation formula based on the long and short time windows and the moving average, wherein the third calculation formula includes: DMA t =MA t ,w1-MA t ,w2(w1 <w2); Among them, MA t It is the moving average value at a specific time in the time series, w1 is a short time window, and w2 is a long time window; DMA t is the moving average dispersion; The step of calculating the dispersion of the moving slope average value based on the moving average dispersion comprises: The dispersion of the moving slope average value is calculated using a fourth calculation formula based on the dispersion of the moving average value, wherein the fourth calculation formula includes: DMSA t =MSA t ,w1-MSA t ,w2(w1<w2); Among them, DMSA t is the dispersion of the moving slope average in a specific time window, MSA t It is the moving slope average of the moving average dispersion in a specific time window.
8. A device for identifying lost circulation, characterized in that: include: An acquisition module, used to acquire measurement data of a target well based on a time series during the drilling process; A calculation module, used for calculating characteristic parameters based on the measurement data, wherein the characteristic parameters are used for characterizing the change trend of the measurement data based on the time series; A construction module, used for constructing a time series-based characteristic parameter mixing matrix based on the characteristic parameters and the measurement data; The identification module is used to input the characteristic parameter mixing matrix into a pre-established machine learning model to identify the well leakage of the target well.
9. An electronic device, characterized in that: include: A memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the method according to any one of claims 1 to 7 is executed.
10. A storage medium, characterized in that: The computer program stored in the storage medium can be executed by one or more processors and can be used to implement the method as claimed in any one of claims 1 to 7.