Self-adaptive drilling speed regulation method and system based on machine learning

Through the adaptive drilling speed regulation method based on machine learning, the rock hardness prediction model and optimal result library are used to adjust the drilling speed and force in real time, solving the drilling pressure and speed adjustment problems in the existing technology and improving the downhole anchor Efficiency and safety of rod support.

CN119002257AInactive Publication Date: 2024-11-22TAIYUAN INST OF CHINA COAL TECH & ENG GROUP +1

Patent Information

Application Number
CN202411004714.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2024-07-17
Filing Date
2024-07-25
Publication Date
2024-11-22
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing adaptive drilling technology is difficult to accurately adjust the drilling pressure and speed during the downhole anchor support process, resulting in a high probability of drilling rod damage when the rock hardness changes, affecting efficiency and safety.

Method used

Adaptive drilling speed regulation method based on machine learning is adopted to establish a rock hardness prediction model by obtaining historical data, generating an optimal rotation and feed flow database, adjusting the drilling speed and velocity in real time, and adjusting the drilling parameters in real time in combination with sensor data.

Benefits of technology

It improves drilling efficiency, reduces the probability of drilling pipe damage, avoids the complexity of manual adjustment and excessive evaluation formulas, and comprehensively considers the matching of rotation speed and feed speed.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119002257A_ABST
    Figure CN119002257A_ABST
Patent Text Reader

Abstract

The invention provides a self-adaptive drilling speed regulation method based on machine learning. The method comprises the following steps: acquiring historical data; establishing a rock hardness prediction model, taking the historical data as a training set to train the rock hardness prediction model, and generating an optimal result library at the same time; the drilling machine drills in real time, the current rotation parameter data and the feeding parameter data are input into the rock hardness prediction model, and the current rock hardness is output; retrieving to obtain the current optimal rotating flow and the optimal feeding flow matched with the current rotating flow and the optimal feeding flow; according to the invention, the rotation speed and the feeding speed of the drilling machine are input into the controller, and the controller controls the execution oil cylinder to drill the rock with the current hardness at the optimal rotation speed and the feeding speed, so that the use of excessive evaluation formulas is avoided, the matching of the rotation speed and the feeding speed of the drilling machine is comprehensively considered, and the drilling efficiency of the drilling machine is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of adaptive drilling technology, and in particular to an adaptive drilling speed regulation method and system based on machine learning. Background Art

[0002] During the process of underground anchor support, the drilling pressure during support is also constantly changing due to the nonlinear change of rock hardness. Existing anchor support drilling rigs are all hydraulically driven, and there is a coupling relationship between their drilling pressure and drilling speed. When encountering a working condition with low rock hardness, the drilling pressure is low. If the drilling speed is also low at this time, it will affect the support efficiency; when encountering a working condition with high rock hardness, the drilling pressure is very high. If the drilling speed is also very high at this time, it will damage the drill rod or drill bit, and even seriously threaten the safety of the operator. At the same time, under normal circumstances, during the drilling process, the rotation pressure and feed pressure represent the cutting force and propulsion force respectively. There is also a coupling relationship between the cutting force and the propulsion force, that is, the appropriate cutting force and propulsion force will improve the drilling efficiency.

[0003] Under current working conditions, the control of drilling pressure is manually controlled, that is, the speed regulating valve is manually adjusted according to the drilling pressure to adjust the drilling speed. For those who are not familiar with the equipment, the maximum rotation speed and maximum feed speed are directly used regardless of the hardness of the rock, and the probability of damage to the drill pipe will be as high as 90%.

[0004] At present, some adaptive drilling methods have the following two problems: 1) Some adaptive drilling methods first evaluate the current drilling status and then adjust the current rotation speed. For evaluating drilling status such as drill sticking probability, slag discharge smoothness, anchoring stability, etc., the evaluation formula has too many hyperparameters and cannot accurately construct a mathematical model, so it is still in the theoretical stage. 2) Some adaptive drilling methods do not participate in the adjustment of rotation speed, but only match the drilling power by adjusting the drilling speed. Summary of the invention

[0005] In order to solve the above technical problems or at least partially solve the above technical problems, the present disclosure provides an adaptive drilling speed control method and system based on machine learning.

[0006] According to a first aspect, an adaptive drilling speed control method based on machine learning comprises the following steps: obtaining historical data, including but not limited to historical rock hardness data, historical rotation parameter data, historical feed parameter data and historical drilling parameter data; establishing a rock hardness prediction model, using the above historical data as a training set to train the rock hardness prediction model, and obtaining the optimal rotation flow rate and optimal feed flow rate corresponding to each rock hardness based on experiments to generate an optimal result library; the drilling rig drills in real time, the execution cylinder controls the rotation and feed action of the drilling rig, and obtains the current rotation parameter data and feed parameter data through sensors; the rotation parameter data and feed parameter data are input into the rock hardness prediction model, and the current rock hardness is output; the current rock hardness is input into the optimal result library, and the current optimal rotation flow rate and optimal feed flow rate that match it are retrieved; the current optimal rotation flow rate and optimal feed flow rate are input into the controller, and the controller controls the execution cylinder to drill the rock at the optimal rotation speed and feed speed.

[0007] Optionally, the optimal rotational flow rate and the optimal feed flow rate corresponding to each type of rock hardness are obtained based on experiments, and an optimal result library is generated, specifically including: selecting preset target data as experimental parameters, conducting experiments according to the selected experimental parameters, so as to obtain the optimal rotational flow rate and the optimal feed flow rate corresponding to each type of rock hardness, and generating an optimal result library, wherein the preset target data include but are not limited to the probability of damage to the drill pipe and drill bit, and the drilling time.

[0008] Optionally, after the current optimal rotational flow rate and optimal feed flow rate are input into the controller, and the controller controls the execution cylinder to drill the rock at the optimal rotational speed and feed speed, it also includes: obtaining new rotation parameter data and feed parameter data through the sensor, and inputting the new rotation parameter data and feed parameter data into the rock hardness prediction model to obtain new rock hardness; when the new rock hardness is the same as the old rock hardness, the rotation speed and feed speed are maintained constant, and the old rock hardness is the rock hardness output by the rock hardness prediction model at the previous moment; when the new rock hardness is different from the old rock hardness, the optimal result library is re-entered to select a new optimal rotational flow rate and optimal feed flow rate, and then the controller controls the execution cylinder to drill according to the new optimal rotational flow rate and optimal feed flow rate.

[0009] According to the second aspect, an adaptive drilling speed control system based on machine learning is characterized in that it includes: a machine learning model, an optimal result library, a controller and an execution cylinder; the machine learning model is a rock hardness prediction model, which is used for parameter adjustment, parameter selection and training the rock hardness prediction model with historical data as a training set, and the trained rock hardness prediction model can output the current rock hardness according to the current rotation parameter data and feed parameter data, and the historical data includes but is not limited to historical rock hardness data, historical rotation parameter data, historical feed parameter data and historical drilling parameter data; the optimal result library is an input optimal rotation flow rate and optimal feed flow rate corresponding to each type of rock hardness; the controller is used to control the execution cylinder to drill the rock of the current hardness at the optimal rotation speed and feed speed; the execution cylinder is the actuator required for the rotation function and feed function of the drilling rig.

[0010] Optionally, it also includes: a hydraulic valve and a sensor; the hydraulic valve is used to control the rotation function and feeding function of the drilling rig; the sensor is used to sense the current drilling data.

[0011] Optionally, the hydraulic valve has a proportional characteristic for proportionally adjusting the flow rate of the pressure oil according to the instruction of the controller.

[0012] Optionally, the sensor includes at least a displacement sensor, an angle sensor, a pressure sensor and a flow sensor; the displacement sensor is used to sense the drilling depth, the angle sensor is used to sense the drilling angle, the pressure sensor is used to sense the rotation pressure and feed pressure, and the flow sensor is used to sense the rotation flow and feed flow.

[0013] Compared with the prior art, the technical solution provided by the embodiments of the present disclosure has the following advantages:

[0014] The present invention provides an adaptive drilling speed regulation method based on machine learning, comprising the following steps: acquiring historical data, the historical data including but not limited to historical rock hardness data, historical rotation parameter data, historical feed parameter data and historical drilling parameter data; establishing a rock hardness prediction model, using the above historical data as a training set to train the rock hardness prediction model, and acquiring the optimal rotation flow and optimal feed flow corresponding to each rock hardness based on experiments, and generating an optimal result library; when a drilling rig is currently drilling, an execution cylinder controls the rotation and feed actions of the drilling rig, and currently acquires rotation parameter data and feed parameter data through a sensor; inputs the rotation parameter data and feed parameter data into the rock hardness prediction model, and outputs the current rock hardness; inputs the current rock hardness into the optimal result library, and retrieves the current optimal rotation flow and optimal feed flow that match it; inputs the current optimal rotation flow and optimal feed flow into a controller, and the controller controls the execution cylinder to drill the rock of the current hardness at the optimal rotation speed and feed speed. The present invention obtains the optimal result library of drilling rig through simple experimental observation, inputs the current collected data into the rock hardness prediction model to predict the current rock hardness, and then selects the optimal rotation flow rate and the optimal feed flow rate in the optimal result library according to the current rock hardness, thereby avoiding the use of too many evaluation formulas for the drilling rig state when establishing the mathematical model, and comprehensively considering the matching of the drilling rig rotation speed and the feed speed, thereby improving the drilling efficiency of the drilling rig. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0017] Figure 1 A specific example flow chart of an adaptive drilling speed control method based on machine learning according to an embodiment of the present disclosure;

[0018] Figure 2 This is an example diagram of an adaptive drilling speed control system based on machine learning according to an embodiment of the present disclosure;

[0019] Figure 3 A specific example structural diagram of a computer device according to an embodiment of the present disclosure.

[0020] Among them, 1-machine learning model, 2-historical database, 3-optimal result library, 4-current optimal result, 5-controller, 6-hydraulic valve, 7-executor cylinder, 8-sensor, 9-current sensor data, 10-current rock hardness. DETAILED DESCRIPTION

[0021] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0022] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.

[0023] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0024] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0025] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0026] Figure 1 A specific example flow chart of an adaptive drilling speed control method based on machine learning is shown.

[0027] like Figure 1 As shown, the adaptive drilling speed control method based on machine learning includes the following steps:

[0028] S100, acquiring historical data, where the historical data includes but is not limited to historical rock hardness data, historical rotation parameter data, historical feed parameter data, and historical drilling parameter data.

[0029] In this embodiment, the rotational pressure and feed pressure of different rock hardnesses, different drilling depths, different drilling angles, and different input rotational flow rates and feed flow rates are collected through actual working scenarios, and then the rock hardness is used as a unit. For each rock hardness, different flow parameters are changed, and a large number of experiments are carried out to obtain historical data and establish a historical database 2, that is, each rock of a certain hardness has several groups of historical drilling parameter data and historical feed parameter data.

[0030] In this embodiment, historical data includes but is not limited to drilling depth, drilling rig angle, rotation pressure, feed pressure, rotation flow, and feed flow. Input parameters such as temperature, humidity, drill pipe diameter, and vibration may also be added based on the constructed machine learning model 1. The present invention is not limited to this.

[0031] S200, establishing a rock hardness prediction model, using the above historical data as a training set to train the rock hardness prediction model, and obtaining the optimal rotation flow rate and optimal feed flow rate corresponding to each type of rock hardness based on experiments, and generating an optimal result library 3.

[0032] In this embodiment, the rock hardness prediction model is a machine learning model 1, which is mainly a classification algorithm. Learning and training are performed through a training set constructed with historical data to build a rock hardness prediction model.

[0033] In this embodiment, the learning and training methods used to build the rock hardness prediction model include but are not limited to Gaussian process classification algorithm, decision tree, neural network, and random forest.

[0034] Optionally, based on the experiment, the optimal rotational flow rate and the optimal feed flow rate corresponding to each type of rock hardness are obtained to generate the optimal result library 3, which specifically includes: selecting preset target data as experimental parameters, conducting experiments according to the selected experimental parameters, so as to obtain the optimal rotational flow rate and the optimal feed flow rate corresponding to each type of rock hardness, and generate the optimal result library 3, and the preset target data include but are not limited to the probability of damage to the drill pipe and drill bit, and the drilling time.

[0035] In this embodiment, each type of rock hardness needs to match an optimal rotation flow rate and an optimal feed flow rate, and the optimal selection criterion is to select the drill pipe and drill bit damage probability, drilling time, etc. observed in the experimental process as preset target data, and generate the optimal result library 3 based on the selection results.

[0036] S300 , the drilling rig drills in real time, the oil cylinder 7 controls the rotation and feeding action of the drilling rig, and obtains the current rotation parameter data and feeding parameter data through the sensor 8 .

[0037] In this embodiment, when the drilling rig just starts drilling, the execution cylinder 7 controls the drilling rig to perform the initial rotation and feed action. At the same time, the sensor 8 currently obtains the data when the drilling rig is just started. The currently obtained data includes the current rotation pressure, the current rotation flow, the current feed pressure, the current feed flow, and may also include the current drilling depth and the current drilling angle.

[0038] In this embodiment, when performing the initial rotation and feeding actions, the rotation parameter data and the feeding parameter data of the execution cylinder 7 can find matching values ​​in several groups of historical drilling parameter data and historical feeding parameter data.

[0039] S400, inputting the rotation parameter data and the feed parameter data into the rock hardness prediction model, and outputting the current rock hardness 10.

[0040] In this embodiment, the rotation parameter data, feed parameter data and data in the historical data are classified and matched through classification operations in the rock prediction model to obtain the corresponding rock hardness in the historical data 2. The rock hardness found is the current rock hardness 10 predicted by the rock hardness prediction model.

[0041] S500, input the current rock hardness 10 into the optimal result library 3, and retrieve the current optimal rotation flow rate and optimal feed flow rate that match it.

[0042] In this embodiment, the current rock hardness 10 is input into the optimal result library 3 for retrieval and matching to obtain the current optimal rotation flow rate and optimal feed flow rate, that is, the rotation flow rate and feed flow rate during the initial rotation and feed actions evolve into the current optimal rotation flow rate and optimal feed flow rate.

[0043] S600, input the current optimal rotation flow rate and the optimal feed flow rate into the controller 5, and the controller 5 controls the execution cylinder 7 to drill the rock at the optimal rotation speed and feed speed.

[0044] Optionally, it also includes: obtaining new rotation parameter data and feed parameter data through the sensor 8, and inputting the new rotation parameter data and feed parameter data into the rock hardness prediction model to obtain new rock hardness; when the new rock hardness is the same as the old rock hardness, maintaining a constant rotation speed and feed speed, and the old rock hardness is the rock hardness output by the rock hardness prediction model at the previous moment; when the new rock hardness is different from the old rock hardness, re-entering the optimal result library 3 to select a new optimal rotation flow and optimal feed flow, and then the controller 5 controls the execution cylinder 7 to perform drilling according to the new optimal rotation flow and optimal feed flow.

[0045] In this embodiment, the drilling action of the drilling rig is a continuous process. During the drilling process, the hardness of the rock changes nonlinearly. Therefore, it is necessary to continuously judge the change of the rock hardness, compare the data collected at the current moment with the data at the previous moment, so as to judge whether the rock hardness has changed, and continuously adjust the rotation flow rate and feed flow rate of the drilling rig according to the change of rock hardness, so as to achieve the optimal evolution of the whole drilling process.

[0046] The disclosed embodiment acquires historical data, including but not limited to historical rock hardness data, historical rotation parameter data, historical feed parameter data and historical drilling parameter data; establishes a rock hardness prediction model, uses the above historical data as a training set to train the rock hardness prediction model, and acquires the optimal rotation flow and optimal feed flow corresponding to each rock hardness based on experiments, and generates an optimal result library 3; when the drilling rig is currently drilling, the execution cylinder 7 controls the rotation and feed action of the drilling rig, and currently acquires the rotation parameter data and feed parameter data through the sensor 8; inputs the rotation parameter data and feed parameter data into the rock hardness prediction model, and outputs the current rock hardness 10; inputs the current rock hardness 10 into the optimal result library 3, and retrieves the current optimal rotation flow and optimal feed flow that match it; inputs the current optimal rotation flow and optimal feed flow into the controller 5, and the controller 5 controls the execution cylinder 7 to drill the rock of the current hardness at the optimal rotation speed and feed speed. The present invention obtains the optimal result library 3 of the drilling rig through simple experimental observation, inputs the current collected data into the rock hardness prediction model to predict the current rock hardness 10, and then selects the optimal rotation flow rate and the optimal feed flow rate in the optimal result library 3 according to the current rock hardness 10, thereby avoiding the use of too many evaluation formulas for the drilling rig state when establishing the mathematical model, and comprehensively considering the matching of the drilling rig rotation speed and the feed speed, thereby improving the drilling efficiency of the drilling rig.

[0047] like Figure 2 As shown, the present disclosure also provides an adaptive drilling speed control system based on machine learning, including: a machine learning model 1, an optimal result library 3, a controller 5 and an execution cylinder 7; the machine learning model 1 is a rock hardness prediction model, which is used for parameter adjustment, parameter selection and training the rock hardness prediction model with historical data as a training set, and the trained rock hardness prediction model can output the current rock hardness 10 according to the current rotation parameter data and feed parameter data, and the historical data includes but is not limited to historical rock hardness data, historical rotation parameter data, historical feed parameter data and historical drilling parameter data; the optimal result library 3 is the input optimal rotation flow rate and optimal feed flow rate corresponding to each rock hardness; the controller 5 is used to control the execution cylinder 7 to drill the rock of the current hardness at the optimal rotation speed and feed speed; the execution cylinder 7 is the actuator required for the rotation function and feed function of the drilling rig.

[0048] In this embodiment, it also includes: a hydraulic valve 6 and a sensor 8; the hydraulic valve 6 is used to control the rotation function and the feeding function of the drilling rig; the sensor 8 is used to sense the current drilling data.

[0049] In this embodiment, the hydraulic valve 6 has a proportional characteristic, and is used to proportionally adjust the flow rate of the pressure oil according to the instruction of the controller 5 .

[0050] In this embodiment, the sensor 8 includes at least a displacement sensor, an angle sensor, a pressure sensor and a flow sensor; the displacement sensor is used to sense the drilling depth, the angle sensor is used to sense the drilling angle, the pressure sensor is used to sense the rotation pressure and feed pressure, and the flow sensor is used to sense the rotation flow and feed flow.

[0051] In this embodiment, historical data is stored in the historical database 2, and the machine learning model retrieves historical data from the historical database 2 for model training. At the same time, the rotation parameter data and feed parameter data currently obtained by the sensor 8 are transmitted to the machine learning model 1. The current rock hardness 10 output by the machine learning model 1 is transmitted to the optimal result library 3. The current optimal rotation flow and optimal feed flow retrieved from the optimal result library 3 are transmitted to the controller 5. The controller 5 controls the hydraulic valve 6 to output pressure oil according to the current optimal rotation flow and optimal feed flow. The pressure oil flows into the execution cylinder 7 to adjust the drilling rig rotation speed and drilling rig speed.

[0052] In this embodiment, firstly, taking rock hardness as a unit, for each type of rock hardness, different rotational flow rates and feed parameters are changed, a large number of experiments are performed, historical data are obtained, and a historical database 2 is established. Historical data are retrieved from the historical database 2, and a training set is constructed with the historical data as a combination of rotational flow rate, feed flow rate and rock hardness. Then, the training set is transmitted to the rock hardness prediction model, and the rock hardness prediction model is trained by continuously adjusting parameters and selecting models to obtain a trained rock hardness prediction model. At the same time, the optimal rotational flow rate and optimal feed flow rate matched by each type of hardness of rock are combined according to the historical data to generate an optimal result library 3. Then, the current rotational parameter data and feed parameter data are obtained through the sensor 8, and the rotational parameter data and feed parameter data are input into the rock hardness prediction model, and the current rock hardness 10 is output. The current rock hardness 10 is input into the optimal result library 3, and the current optimal rotational flow rate and optimal feed flow rate matching it are retrieved, and the current optimal rotational flow rate and optimal feed flow rate are transmitted to the controller 5.

[0053] When the drilling rig starts drilling, the hydraulic valve 6 changes the valve core opening according to the current optimal rotation flow and optimal feed flow received from the controller 5, thereby adjusting the hydraulic oil flow, and then controls the execution cylinder 7 to complete the rotation and feed action of the drilling rig.

[0054] In this embodiment, the machine learning model 1 continuously receives the current drilling depth, current drilling rig angle, current rotational pressure, current feed pressure, current rotational flow rate and current feed flow rate sensed by the sensor 8, and performs classification operations. If the rock hardness at the next moment obtained by the machine learning model 1 is the same as the current rock hardness 10, the rotational flow rate and feed flow rate are maintained constant for drilling. If the rock hardness at the next moment changes compared with the current rock hardness 10, the optimal result library 3 is re-selected to select the optimal rotational flow rate and optimal feed flow rate at the next moment, and retransmit them to the controller 5.

[0055] In this embodiment, the actuator cylinder 7 may also be a hydraulic motor, an electric cylinder, an electric motor, a pneumatic cylinder, or a pneumatic motor, but the present invention is not limited thereto.

[0056] The embodiment of the present invention also provides a computer device, such as Figure 3 As shown, Figure 3 It is a structural diagram of a computer device provided by an optional embodiment of the present invention, and the computer device may include at least one processor 41, at least one communication interface 42, at least one communication bus 43 and at least one memory 44, wherein the communication interface 42 may include a display screen (Display) and a keyboard (Keyboard), and the optional communication interface 42 may also include a standard wired interface and a wireless interface. The memory 44 may be a high-speed RAM memory (Random Access Memory, volatile random access memory) or a non-volatile memory (non-volatile memory), such as at least one disk storage. The memory 44 may optionally be at least one storage device located away from the aforementioned processor 41. The memory 44 stores application programs, and the processor 41 calls the program code stored in the memory 44 to execute the steps of any of the above-mentioned method embodiments.

[0057] The communication bus 43 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The communication bus 43 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0058] Among them, the memory 44 may include a volatile memory (English: volatile memory), such as a random access memory (English: random-access memory, abbreviated: RAM); the memory may also include a non-volatile memory (English: non-volatile memory), such as a flash memory (English: flash memory), a hard disk drive (English: hard disk drive, abbreviated: HDD) or a solid-state drive (English: solid-state drive, abbreviated: SSD); the memory 44 may also include a combination of the above types of memory.

[0059] The processor 41 may be a central processing unit (CPU), a network processor (NP), or a combination of a CPU and a NP.

[0060] The processor 41 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The PLD may be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof.

[0061] Optionally, the memory 44 is also used to store program instructions. The processor 41 can call the program instructions to implement the present invention. Figure 1 An adaptive drilling speed control method based on machine learning is shown in the embodiment.

[0062] The embodiment of the present invention further provides a non-transitory computer storage medium, wherein the computer storage medium stores computer executable instructions, and the computer executable instructions can execute the method in any of the above method embodiments. The storage medium may be a disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), a flash memory, a hard disk drive (HDD) or a solid-state drive (SSD), etc.; the storage medium may also include a combination of the above types of memories.

[0063] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0064] The above description is only a specific embodiment of the present disclosure, so that those skilled in the art can understand or implement the present disclosure. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present disclosure. Therefore, the present disclosure will not be limited to the embodiments described herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An adaptive drilling speed control method based on machine learning, characterized in that: The steps include: Acquire historical data, the historical data including but not limited to historical rock hardness data, historical rotation parameter data, historical feed parameter data and historical drilling parameter data; Establish a rock hardness prediction model, use the above historical data as a training set to train the rock hardness prediction model, and obtain the optimal rotation flow rate and optimal feed flow rate corresponding to each type of rock hardness based on experiments to generate an optimal result library (3); The drilling rig drills in real time, and the oil cylinder (7) controls the rotation and feeding action of the drilling rig, and obtains the current rotation parameter data and feeding parameter data through the sensor (8); Inputting the rotation parameter data and the feed parameter data into the rock hardness prediction model, and outputting the current rock hardness; Input the current rock hardness into the optimal result library (3), and retrieve the current optimal rotation flow rate and optimal feed flow rate that match it; The current optimal rotation flow rate and the optimal feed flow rate are input into the controller (5), and the controller (5) controls the execution cylinder (7) to drill the rock at the optimal rotation speed and feed speed.

2. The method for adaptive drilling speed regulation based on machine learning according to claim 1 is characterized in that: The method of obtaining the optimal rotation flow rate and the optimal feed flow rate corresponding to each type of rock hardness based on the experiment and generating the optimal result library (3) specifically includes: Preset target data are selected as experimental parameters, and experiments are carried out according to the selected experimental parameters to obtain the optimal rotation flow rate and optimal feed flow rate corresponding to each type of rock hardness, and generate an optimal result library (3). The preset target data include but are not limited to the probability of damage to the drill pipe and drill bit, and the drilling time.

3. The method for adaptive drilling speed regulation based on machine learning according to claim 1, characterized in that: After the current optimal rotation flow rate and the optimal feed flow rate are input into the controller (5), and the controller (5) controls the execution cylinder (7) to drill at the optimal rotation speed and feed speed, the method further includes: Acquiring new rotation parameter data and feed parameter data through a sensor (8), and inputting the new rotation parameter data and feed parameter data into a rock hardness prediction model to obtain new rock hardness; When the new rock hardness is the same as the old rock hardness, the rotation speed and feed speed are kept constant, and the old rock hardness is the rock hardness output by the rock hardness prediction model at the last moment; When the new rock hardness is different from the old rock hardness, the optimal result library (3) is re-entered to select a new optimal rotation flow rate and an optimal feed flow rate, and then the controller (5) controls the execution cylinder (7) to perform drilling according to the new optimal rotation flow rate and the optimal feed flow rate.

4. An adaptive drilling speed control system based on machine learning, used to implement the adaptive drilling speed control method based on machine learning as described in any one of claims 1 to 3, characterized in that: include: Machine learning model (1), optimal result library (3), controller (5) and execution cylinder (7); The machine learning model (1) is a rock hardness prediction model, which is used for parameter adjustment, parameter selection, and training the rock hardness prediction model using historical data as a training set. The trained rock hardness prediction model can output the current rock hardness according to current rotation parameter data and feed parameter data, wherein the historical data includes but is not limited to historical rock hardness data, historical rotation parameter data, historical feed parameter data, and historical drilling parameter data; The optimal result library (3) is the optimal input rotation flow rate and optimal feed flow rate corresponding to each rock hardness; The controller (5) is used to control the execution cylinder (7) to drill the rock of current hardness at an optimal rotation speed and feed speed; The actuator cylinder (7) is an actuator required for the rotation function and the feeding function of the drilling machine.

5. The adaptive drilling speed control system based on machine learning according to claim 4 is characterized in that: Also includes: Hydraulic valve (6) and sensor (8); The hydraulic valve (6) is used to control the rotation function and feeding function of the drilling rig; The sensor (8) is used to sense current drilling data.

6. The adaptive drilling speed control system based on machine learning according to claim 5 is characterized in that: The hydraulic valve (6) has a proportional characteristic and is used to proportionally adjust the flow rate of the pressure oil according to the instruction of the controller (5).

7. The adaptive drilling speed control system based on machine learning according to claim 5, characterized in that: The sensor (8) comprises at least a displacement sensor, an angle sensor, a pressure sensor and a flow sensor; The displacement sensor is used to sense the drilling depth, the angle sensor is used to sense the drilling angle, the pressure sensor is used to sense the rotation pressure and feed pressure, and the flow sensor is used to sense the rotation flow and feed flow.

Citation Information

Patent Citations

  • Rotary-break rock breaking method based on drill bit self-adaptive resonance

    CN114776222A

  • Method of real-time drilling simulation

    US20070185696A1

  • Control method

    US20160245064A1

  • Methods and systems for improved drilling operations using real-time and historical drilling data

    WO2014066981A1

Cited By

  • Anchor rod drill carriage system and method capable of rapidly detecting hardness of rock mass

    CN119572136A

  • Anchor drilling vehicle system and method capable of quickly detecting rock mass hardness

    CN119572136B

  • Full-autonomous drilling control system and control method for coal mine drilling robot

    CN120007193A