Method for monitoring abrasion of drill bit in real time in drilling process

Abstract

The invention provides a method for monitoring drill bit abrasion in real time in the drilling process, and belongs to the technical field of oil and gas drilling. The method comprises the steps that a database is built through logging engineering data and drill bit use data, and drilling parameters with the maximum correlation with the drill bit abrasion degree are screened; the database is subjected to noise reduction processing, the drill bit abrasion process is graded through a clustering analysis algorithm, and a drill bit abrasion process label is constructed; establishing a mapping relation between drilling parameters and wear labels through a machine learning method, and establishing a drill wear monitoring model; on the basis of the drill bit wear monitoring model, real-time data of drilling parameters are input, the drill bit wear model is trained and updated through the drill bit wear process label, and real-time updating of the parameters of the drill bit wear monitoring model in the drilling process is achieved. According to the method, the wear state of the underground drill bit can be quantitatively represented, so that tripping decision making on site is assisted, the drilling cost is reduced, the drilling quality and the drilling efficiency are improved, and the drilling safety and controllability are enhanced.

Description

Technical Field

[0001] The invention belongs to the technical field of oil and gas drilling, and in particular relates to a method for real-time monitoring of drill bit wear during drilling. Background Art

[0002] In the era of modern industrial revolution, drill bit wear monitoring technology has become an important research topic in the field of drilling engineering. Due to the increasing consumption of non-renewable resources such as oil and natural gas, efficient and accurate drill bit wear monitoring technology is of great significance for improving drilling efficiency, reducing drilling costs and ensuring energy security.

[0003] Drill bit wear directly affects the mechanical drilling speed during the drilling process, thereby affecting the progress and cost of the entire drilling project. Traditional drill bit wear monitoring methods are restricted by many factors, such as the complexity of working conditions, the difficulty in obtaining logging parameters, and the high cost of human intervention, which seriously affects drilling efficiency.

[0004] To solve this problem, the research on drill bit wear monitoring technology has become a hot topic at home and abroad. With the rapid development of big data technology and the rapid growth of data scale, the application of machine learning methods has great potential in the field of drill bit wear monitoring. Compared with the method based on physical model, machine learning technology can significantly improve the accuracy and real-time performance of drill bit wear monitoring, thereby improving drilling efficiency and reducing drilling costs.

[0005] The application of intelligent algorithms in drill bit wear monitoring technology can predict drill bit wear more accurately and quickly, providing strong support for the drilling process. This will help shorten the drilling cycle, reduce drilling costs, improve drilling quality and efficiency, and enhance the safety and controllability of drilling. In summary, drill bit wear monitoring technology is of great significance for promoting the sustainable development of the drilling industry, reducing corporate drilling costs, and reducing drilling risks.

[0006] The Chinese invention patent application document CN107560542A discloses a drill bit wear monitoring method, including the following steps: S1, setting multiple monitoring points within the drill bit revolution life interval; S2, stopping the use of the drill bit when the revolution of the monitoring point is reached; S3, performing point cloud scanning on the drill bit at the monitoring point through a three-dimensional laser scanner to collect the three-dimensional point cloud data of the drill bit; S4, establishing a three-dimensional model of the drill bit at the monitoring point through modeling software according to the three-dimensional point cloud data of the drill bit; S5, comparing the three-dimensional model of the drill bit at the monitoring point with the three-dimensional model of the standard drill bit; S6, judging whether the wear of the drill bit at the monitoring point is within the allowable error range, when the wear of the drill bit exceeds the allowable error range, the drill bit needs to be ground or scrapped; S7, after the drill bit is ground, S3 to S6 need to be executed again. This drill bit wear monitoring method has high measurement accuracy and can find some small defects, thereby improving the drilling quality of the drill bit. However, this method is limited by the complexity of the working conditions and the difficulty in obtaining logging parameters. Summary of the invention

[0007] In view of the problems existing in the above-mentioned prior art, the present invention provides a method for real-time monitoring of drill bit wear during drilling. The technical problem to be solved by the present invention is how to efficiently and accurately monitor the degree of drill bit wear.

[0008] In order to solve the above technical problems, the present invention provides a method for real-time monitoring of drill bit wear during drilling, comprising the following steps:

[0009] Step S1: using logging engineering data and drill bit usage data, constructing a database to screen the drilling parameters with the greatest correlation with the degree of drill bit wear;

[0010] Step S2: performing noise reduction processing on the database, classifying the drill wear process through a cluster analysis algorithm, and constructing a drill wear process label;

[0011] Step S3: Establishing a mapping relationship between drilling parameters and drill bit wear process labels through a machine learning method, and establishing a drill bit wear monitoring model;

[0012] Step S4: Based on the drill bit wear monitoring model, the real-time data of drilling parameters is input, and the drill bit wear monitoring model is updated by training using the drill bit wear process label, so as to realize the real-time update of the drill bit wear monitoring model parameters during the drilling process.

[0013] Furthermore, in step S1, the logging engineering data and drill bit usage data are merged into the same data table, and the data of each tripping and drilling are classified according to the type of drill bit, and the data of the same drill bit model are integrated to complete the construction of the database.

[0014] Furthermore, in step S1, the data table includes well depth, drilling time, drilling pressure, rotation speed, torque, drilling speed, drill bit diameter, drill bit model, drill bit wear level and well entry newness.

[0015] Furthermore, in step S1, a filtering algorithm is used to clean up abnormal values ​​of the logging engineering data and the drill bit usage data.

[0016] Furthermore, in step S1, the missing data in the logging engineering data and the drill bit usage data are supplemented by an interpolation method.

[0017] Furthermore, in step S2, a filtering algorithm is used to perform noise reduction on the data.

[0018] Furthermore, in step S2, the wear process is classified using the K-means clustering algorithm, and the data points are grouped according to the similarity between the data points, so as to classify the process of drill bit wear in the well and achieve quantitative characterization of the drill bit wear process data.

[0019] Furthermore, in step S3, the machine learning method includes a GRU neural network model.

[0020] Furthermore, in step S3, the machine learning method includes a GRU neural network model, an XGBoost model and a support vector machine model.

[0021] Furthermore, in step S4, the drilling parameters include well depth, drilling pressure, rotation speed, mechanical drilling speed, torque, displacement, mechanical specific energy and drill bit model.

[0022] The beneficial effects of the method for real-time monitoring of drill bit wear during drilling of the present invention include at least: establishing a drill bit wear monitoring model through an intelligent algorithm, which can quantitatively characterize the wear state of the downhole drill bit, thereby assisting on-site tripping and drilling decisions, reducing drilling costs, improving drilling quality and drilling efficiency, enhancing drilling safety and controllability, and making the drilling process more efficient and safe.

[0023] The present invention provides a method for real-time monitoring of drill bit wear during drilling. The method combines an artificial intelligence deep learning model to process real-time data obtained during drilling and establishes a mapping relationship between drilling parameters and drill bit wear levels. The wear degree of the bottom hole drill bit is quantitatively characterized, which can help construction personnel determine the timing of drilling, thereby assisting in shortening the drilling cycle, reducing drilling costs, improving drilling quality and efficiency, enhancing the safety and controllability of drilling, and making the drilling process more efficient and safe. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1The present invention is a flow chart of a method for real-time monitoring of drill bit wear during drilling.

[0025] Figure 2 is a schematic diagram of a GRU neural network model of a method for real-time monitoring of drill bit wear during drilling in the present invention; the figure embodies the GRU network calculation principle and process, and calculates the gate values ​​of the update gate and the reset gate, which are z t and r t , the calculation method is to use x t With h t-1 The concatenation is linearly transformed and then activated by sigmoid. The gate value is then reset to act on h t-1 It represents how much information from the previous time step can be used. Then use this reset h t-1 Perform basic RNN calculations, that is, with x t The splicing is linearly changed, and after tanh activation, a new h is obtained t The gate value of the last updated gate will act on the new h t , and the 1-gate value will act on h t-1 Then add the two results together to get the final hidden state output h t ,This process means that the update gate has the ability to retain the previous result. When the gate value approaches 1, the output is the new h t , and when the gate value approaches 0, the output is h of the previous time step t-1 ; where z t and r t They are the gate values ​​of the update gate and the reset gate, σ is the sigmoid function, tanh is the tanh function, and x t is the input of the current value, h t-1 is the output at the previous moment, h t is the output at the current moment, is the output after tanh activation.

[0026] Figure 3 It is a schematic diagram of the clustering and grading process of a method for real-time monitoring of drill bit wear during drilling in the present invention.

[0027] Figure 4 It is a schematic diagram of the monitoring process output of a method for real-time monitoring of drill bit wear during drilling in the present invention. DETAILED DESCRIPTION

[0028] In order to make the purpose, 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 in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0030] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0031] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be directly connected or indirectly connected through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0032] In order to better understand the purpose, structure and function of the present invention, a method for real-time monitoring of drill bit wear during drilling of the present invention is further described in detail below with reference to the accompanying drawings.

[0033] Embodiment 1:

[0034] Figure 1 The present invention shows a method for real-time monitoring of drill bit wear during drilling, comprising the following steps:

[0035] Step S1: using logging engineering data and drill bit usage data, constructing a database to screen the drilling parameters with the greatest correlation with the degree of drill bit wear;

[0036] Step S2: performing noise reduction processing on the database, classifying the drill wear process through a cluster analysis algorithm, and constructing a drill wear process label;

[0037] Step S3: Establishing a mapping relationship between drilling parameters and drill bit wear process labels through a machine learning method, and establishing a drill bit wear monitoring model;

[0038] Step S4: Based on the drill bit wear monitoring model, the real-time data of drilling parameters is input, and the drill bit wear monitoring model is updated by training using the drill bit wear process label, so as to realize the real-time update of the drill bit wear monitoring model parameters during the drilling process.

[0039] Embodiment 2:

[0040] Figure 1 The present invention shows a method for real-time monitoring of drill bit wear during drilling, comprising the following steps:

[0041] Step S1: using logging engineering data and drill bit usage data, constructing a database to screen the drilling parameters with the greatest correlation with the degree of drill bit wear;

[0042] Step S2: performing noise reduction processing on the database, classifying the drill wear process through a cluster analysis algorithm, and constructing a drill wear process label;

[0043] Step S3: Establishing a mapping relationship between drilling parameters and drill bit wear process labels through a machine learning method, and establishing a drill bit wear monitoring model;

[0044] Step S4: Based on the drill bit wear monitoring model, the real-time data of drilling parameters is input, and the drill bit wear monitoring model is updated by training using the drill bit wear process label, so as to realize the real-time update of the drill bit wear monitoring model parameters during the drilling process.

[0045] The difference between this embodiment and the first embodiment is that:

[0046] In the step S1, the logging engineering data and the drill bit usage data are merged into the same data table, and the data of each tripping and drilling are classified according to the type of drill bit, and the data of the same drill bit model are integrated to complete the construction of the database.

[0047] In step S1, the data table includes parameters such as well depth, drilling time, drilling pressure, rotation speed, torque, drilling speed, drill bit diameter, drill bit model, drill bit wear level, and well entry freshness.

[0048] In step S1, a filtering algorithm is used to clean up abnormal values ​​of logging engineering data and drill bit usage data.

[0049] In the step S1, the missing data in the logging engineering data and the drill bit usage data are supplemented by using the interpolation method.

[0050] In step S2, a filtering algorithm is used to perform noise reduction on the data.

[0051] In step S2, Figure 3As shown in the figure, the wear process is classified by combining the K-means clustering algorithm, and the data points are grouped according to the similarity between the data points, so as to classify the process of drill bit wear in the well and quantitatively characterize the drill bit wear process data. The K-means clustering algorithm is an unsupervised learning algorithm used to divide data points into K different clusters or categories. The basic idea of ​​this algorithm is to group data points according to the similarity between the data points. The wear process of the drill bit in the well can be classified to solve the problem that the drill bit wear process is difficult to quantitatively characterize.

[0052] In step S3, the machine learning method includes a GRU neural network model.

[0053] In step S3, the machine learning method includes a GRU neural network model, an XGBoost model and a support vector machine model.

[0054] like Figure 2 As shown in the figure, after comparison, it is found that the GRU neural network model performs best in the problem of time-series drill wear monitoring. Due to its unique design structure, the GRU neural network model is suitable for processing and predicting important events with very long intervals and delays in time series, and is suitable for use in drilling processes with time series and drastic data fluctuations. In addition, there is less drill wear data, and GRU performs better on small data training sets.

[0055] The final classification result is compared with the GRU network output to verify the accuracy of the model. In the actual operation, you only need to input the real-time value of the corresponding drilling parameter to get the degree of drill bit wear during the drilling process. The results are as follows: Figure 4 shown.

[0056] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.

Claims

1. A method for real-time monitoring of drill bit wear during drilling, characterized in that: The following steps are involved: Step S1: using logging engineering data and drill bit usage data, constructing a database to screen the drilling parameters with the greatest correlation with the degree of drill bit wear; Step S2: performing noise reduction processing on the database, classifying the drill wear process through a cluster analysis algorithm, and constructing a drill wear process label; Step S3: Establishing a mapping relationship between drilling parameters and drill bit wear process labels through a machine learning method, and establishing a drill bit wear monitoring model; Step S4: Based on the drill bit wear monitoring model, the real-time data of drilling parameters is input, and the drill bit wear monitoring model is updated by training using the drill bit wear process label, so as to realize the real-time update of the drill bit wear monitoring model parameters during the drilling process.

2. The method for real-time monitoring of drill bit wear during drilling according to claim 1, characterized in that: In the step S1, the logging engineering data and the drill bit usage data are merged into the same data table, and the data of each tripping and drilling are classified according to the type of drill bit, and the data of the same drill bit model are integrated to complete the construction of the database.

3. The method for real-time monitoring of drill bit wear during drilling according to claim 2, characterized in that: In step S1, the data table includes well depth, drilling time, drilling pressure, rotation speed, torque, drilling speed, drill bit diameter, drill bit model, drill bit wear level and well entry newness.

4. The method for real-time monitoring of drill bit wear during drilling according to claim 3, characterized in that: In step S1, a filtering algorithm is used to clean up abnormal values ​​of logging engineering data and drill bit usage data.

5. The method for real-time monitoring of drill bit wear during drilling according to claim 4, characterized in that: In the step S1, the missing data in the logging engineering data and the drill bit usage data are supplemented by using the interpolation method.

6. The method for real-time monitoring of drill bit wear during drilling according to claim 1, characterized in that: In step S2, a filtering algorithm is used to perform noise reduction on the data.

7. The method for real-time monitoring of drill bit wear during drilling according to claim 1, characterized in that: In step S2, the wear process is classified using the K-means clustering algorithm, and the data points are grouped according to the similarity between the data points, so as to classify the drill bit wear process in the well and achieve quantitative characterization of the drill bit wear process data.

8. The method for real-time monitoring of drill bit wear during drilling according to claim 1, characterized in that: In step S3, the machine learning method includes a GRU neural network model.

9. The method for real-time monitoring of drill bit wear during drilling according to claim 1, characterized in that: In step S3, the machine learning method includes a GRU neural network model, an XGBoost model and a support vector machine model.

10. The method for real-time monitoring of drill bit wear during drilling according to claim 1, characterized in that: In step S4, the drilling parameters include well depth, drilling pressure, rotation speed, mechanical drilling speed, torque, displacement, mechanical specific energy and drill bit model.

Images