Method and system for determining rock mass basic quality classification in real time based on drilling parameters

Abstract

The present application relates to the method and system for determining the basic quality classification of rock mass in real time based on drilling parameters; the method comprises the following steps: taking the core in the drilling process, collecting the drilling parameters for preprocessing, and calculating the key index of rock breaking; obtaining the effective value of the rock mass integrity coefficient Kv of the whole hole; obtaining the effective value of the saturated uniaxial compressive strength Rc of the rock with different breaking degrees according to the obtained key index of rock breaking; establishing the Rc-Kv pre-judgment model for judging the rock mass with different breaking degrees; obtaining the Rc and Kv values of the rock mass in the hole in real time according to the Rc-Kv pre-judgment model, and calculating the basic quality index of the rock mass. After the pre-judgment model is established, the basic quality classification of the rock mass in the hole is obtained through the mobile control terminal, which is simple to operate, improves the work efficiency, saves time and labor, and the method is not affected by the breaking degree of the rock mass, the pre-judgment result is accurate, and the application range is wide.

Description

Technical Field

[0001] This invention relates to the field of engineering geological drilling, and in particular to a method and system for determining the basic quality classification of rock mass in real time based on drilling parameters. Background Technology

[0002] Drilling remains one of the most important methods in engineering exploration, and with the development of artificial intelligence technology, monitoring while drilling using advanced sensors has become a research hotspot both domestically and internationally. One important research direction in monitoring while drilling is establishing the relationship between acquired drilling parameters and the mechanical parameters of rock and soil. Patent (ZL201811083588.6) discloses a method for rapidly testing the blockiness index of rock mass without core sampling using monitoring while drilling technology. By studying the quantitative relationship between drill bit advance speed, drill pipe rotation speed, drilling rig hydraulic oil pressure level, and the blockiness index of the rock mass in the core, the goal of quickly and accurately obtaining the blockiness index of the rock mass in the core can be achieved during drilling. Patent (202010612710.5) discloses a method for determining rock strength parameters based on a deep convolutional neural network. First, it collects the unconfined compressive strength (UCS), cohesion, and internal friction angle of various common rocks as the basic data for a database; it uses DPMA to collect data during drilling operations; then it uses the obtained data to train a CNN model; finally, it predicts the cohesion, internal friction angle, and UCS of intact rock as the final output of rock strength parameters. Patent (202010042715.9) discloses a method for predicting rock strength parameters based on monitoring while drilling technology, and provides a technique for predicting the uniaxial compressive strength, cohesion and internal friction angle of rocks based on monitoring while drilling technology.

[0003] Existing research mainly relies on parameters (oil pressure, displacement, rotational speed) directly measured by sensors to establish a learning sample library. However, the selected parameters are incomplete, leading to unreasonable applications. The predictive models established based on this are easily affected by working conditions during drilling in fractured rock masses, resulting in large data dispersion. This limits the application conditions of the predictive models and may lead to inaccurate judgments of rock mass strength and integrity parameters when applied in fractured or extremely fractured rock masses in engineering sites. Summary of the Invention

[0004] To address the aforementioned issues, one approach is to develop a method for real-time determination of basic rock mass quality classification based on drilling parameters, comprising the following steps:

[0005] Establish an Rc-Kv prediction model that can distinguish rock masses with different degrees of fragmentation;

[0006] Drilling is carried out on the rock mass, key rock-breaking indicators are collected during the drilling process, and then the Rc and Kv values ​​of the rock mass in the borehole are obtained in real time based on the obtained Rc-Kv prediction model, and the basic quality indicators of the rock mass are calculated.

[0007] The method for establishing the Rc-Kv prediction model includes:

[0008] S100: During the drilling process, core samples are taken and drilling parameters are collected. The drilling parameters are preprocessed to calculate key rock-breaking indicators.

[0009] S200: Obtain the effective value of rock mass integrity coefficient Kv at different depths of the borehole;

[0010] S300: Based on the obtained key rock breaking indicators, the effective value of the saturated uniaxial compressive strength Rc of rocks with different degrees of fragmentation is obtained.

[0011] S400: Based on the key rock-breaking indicators, the effective value of Rc, and the effective value of Kv, establish an Rc-Kv prediction model. Specifically, the Rc-Kv prediction model is a functional relationship between the effective values ​​of Rc and Kv and the key rock-breaking indicators.

[0012] Further, step S100 includes the following steps:

[0013] S101: Obtain drilling parameters, including oil pressure, drilling pressure, spindle speed, spindle torque, drilling displacement, and drilling time;

[0014] S102: Input drilling information on the mobile control terminal;

[0015] S103: Based on the obtained drilling parameters and borehole information, the key rock-breaking indicators are calculated, and the specific distribution patterns of the key rock-breaking indicators with time and borehole depth are obtained in real time.

[0016] Furthermore, the borehole information includes the borehole outer diameter, the core outer diameter, and the drill bit diameter.

[0017] Furthermore, obtaining the effective values ​​of the rock mass integrity coefficient Kv at different borehole depths in step S200 includes:

[0018] Camera work was carried out inside the borehole to obtain the volume joint number Jv of the rock mass at different depths of the borehole; by looking up the table, the effective value of the whole borehole rock mass integrity coefficient Kv was obtained.

[0019] Furthermore, the step S300 of obtaining the effective value of the saturated uniaxial compressive strength Rc of rocks with different degrees of fragmentation includes:

[0020] S301: Select a reference rock mass, conduct a compressive strength test, and obtain the effective value of the saturated uniaxial compressive strength Rc of the reference rock mass through statistical calculation;

[0021] S302: Establish a learning sample library of key rock breaking indicators and effective Rc values ​​to obtain an Rc prediction model;

[0022] S303: Using the prediction model obtained in S302, the effective value of Rc is obtained by predicting the key rock breaking indicators of non-benchmark rock mass segments, and the effective value of Rc of rock masses with different degrees of fracturing is obtained.

[0023] Further, in step S301, the reference rock mass is a complete or relatively complete rock core. The saturated uniaxial compressive strength Rc value of the rock is obtained by indoor uniaxial compressive strength test of the reference rock mass, and the effective value of the saturated uniaxial compressive strength Rc of the rock is obtained by statistical method.

[0024] Furthermore, step S400 includes establishing a learning sample library of key rock-breaking indicators, effective values ​​of Rc and Kv based on the acquired rock-breaking key indicators, effective values ​​of Rc and Kv, using artificial intelligence algorithms, to obtain the Rc-Kv prediction model.

[0025] On the other hand, the present invention also provides a system for determining the basic quality classification of rock mass in real time based on drilling parameters, the system comprising:

[0026] A monitoring-while-drilling system is used to acquire drilling parameters during the drilling process;

[0027] A data acquisition instrument, which is connected to the drilling monitoring system, is used to collect and store drilling parameters;

[0028] The image acquisition module is used to obtain the effective value of the rock mass integrity coefficient Kv at different depths of the borehole;

[0029] A mobile control terminal, which is connected to the data acquisition instrument for data reception and data processing, includes a data processing module.

[0030] The data processing module is used to preprocess the drilling parameters, calculate the key rock breaking indicators, obtain the effective value of the saturated uniaxial compressive strength Rc of rocks with different rock breaking degrees based on the obtained key rock breaking indicators, and finally establish an Rc-Kv prediction model based on the key rock breaking indicators, the effective value of Rc and the effective value of Kv.

[0031] Furthermore, the drilling monitoring system includes a feed pressure sensor, a return oil pressure sensor, a torque sensor, a vertical shaft proximity switch, a drive shaft inductive switch, and a displacement sensor, all connected to the data acquisition instrument. The feed pressure sensor and the return oil pressure sensor are mounted on the feed cylinder of the drilling rig, the torque sensor is mounted on the drive shaft of the drilling rig's gearbox, the vertical shaft proximity switch is mounted on the drilling rig's chuck, the drive shaft inductive switch is mounted on the drilling rig's gearbox, and the displacement sensor is mounted on the drilling rig's chuck to acquire drilling displacement data.

[0032] Furthermore, the data acquisition device includes a wireless communication transmission module for transmitting data to the mobile control terminal in real time.

[0033] This invention provides a method and system for real-time determination of basic rock mass quality grading based on drilling parameters. Various sensors are installed on the drilling rig to acquire drilling parameters during drilling operations, monitoring these parameters. Wireless communication technology is used to process the data in real-time on a mobile control terminal, resulting in a system capable of in-situ determination of basic rock mass quality grading. Specifically, the data processing method for drilling parameters in this invention refers to selecting key rock-breaking indicators based on the principles of rock fracture mechanics and the law of conservation of energy. The derivation and calculation process of these key rock-breaking indicators includes major drilling parameters such as drilling rate, spindle speed, drilling pressure, and spindle torque, while also considering the influence of drill bit diameter on the drilling parameters. Based on artificial intelligence algorithms and grounded in data indicators of the benchmark rock mass within the borehole, a learning sample library is rationally constructed to establish a predictive and discriminant model between the key rock-breaking indicators and key parameters for basic rock mass quality grading (i.e., the effective value of the saturated uniaxial compressive strength Rc and the effective value of the rock mass integrity coefficient Kv). This model is applicable to determining the basic quality grading indicators for rock masses with different degrees of fracture.

[0034] By employing the above technical solutions, this invention has the following advantages compared to existing technologies:

[0035] 1) The method and system for determining the basic quality classification of rock mass in real time based on drilling parameters provided by this invention have little impact on the normal drilling of the drilling rig on site during the construction process. After establishing a prediction model, the basic quality classification of the rock mass in the hole is obtained through a mobile control terminal. The operation is simple, reduces the amount of manual judgment, improves work efficiency, has low professional requirements for technical personnel, saves time and effort, and the method used is not affected by the degree of rock mass fragmentation. The prediction results are accurate and have a wide range of applications.

[0036] 2) The method and system for determining the basic quality classification of rock mass in real time based on drilling parameters provided by this invention are based on the law of conservation of energy and the key indicators for rock breaking proposed by the rock breaking mechanism. The selected parameters are comprehensive, scientific and reasonable, which improves the accuracy of the prediction model. Moreover, the real-time acquisition of the basic quality classification of rock mass solves the problem of the lag in obtaining the basic quality classification of rock mass by conducting various tests. It can quickly and directly provide reference suggestions for engineering construction on site. Attached Figure Description

[0037] Figure 1 This is a flowchart illustrating the method for determining the basic quality classification of rock mass in real time based on drilling parameters according to the present invention.

[0038] Figure 2 This is a flowchart illustrating the method for determining the basic quality classification of rock mass in real time based on drilling parameters according to the present invention.

[0039] Figure 3 This is a schematic diagram of the system for determining the basic quality classification of rock mass in real time based on drilling parameters according to the present invention.

[0040] Figure 4 This is a diagram of the BP network model structure in the method for real-time determination of basic rock mass quality classification based on drilling parameters in this invention.

[0041] Explanation of reference numerals in the attached figures:

[0042] 1-Drilling monitoring system; 11-Feed pressure sensor; 12-Return oil pressure sensor; 13-Torque sensor; 14-Vertical shaft proximity switch; 15-Drive shaft inductive switch; 16-Displacement sensor; 17-First magnet; 18-Second magnet; 19-Reflector;

[0043] 2-Data Acquisition Instrument;

[0044] 3-Mobile control terminal;

[0045] 4-Drilling rig; 41-Drilling tools; 42-Diesel engine; 43-Mud pump; 44-Feed cylinder; 45-Gearbox; 46-Drive shaft; 47-Chuck; 48-Vertical shaft;

[0046] 5-Cable. Detailed Implementation

[0047] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. In the accompanying drawings, the dimensions and relative dimensions of certain parts may be enlarged for clarity.

[0048] In the description of this invention, unless otherwise explicitly specified and limited, the terms "connection" and "linked" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal connection of two elements or the interaction between two elements. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0049] In the description of this invention, terms such as "upper," "lower," "left," "right," "front," and "rear," etc., refer to the orientation or positional relationship shown in the accompanying drawings. They are used only for ease of description and simplification of operation, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0050] Furthermore, in the description of this invention, the terms "first" and "second" are used only for descriptive purposes.

[0051] This invention provides a method and system for real-time determination of basic rock mass quality classification based on drilling parameters, wherein the method includes the following steps:

[0052] Establish an Rc-Kv prediction model that can distinguish rock masses with different degrees of fragmentation;

[0053] Drilling is carried out on the rock mass, key rock-breaking indicators are collected during the drilling process, and then the Rc and Kv values ​​of the rock mass in the borehole are obtained in real time based on the obtained Rc-Kv prediction model, and the basic quality indicators of the rock mass are calculated.

[0054] The method for establishing the Rc-Kv prediction model includes:

[0055] S100: During drilling, core samples are taken and drilling parameters are collected. These parameters are preprocessed to calculate key rock-breaking indices, including the impact energy β consumed per unit volume of rock broken. F Rotational energy β consumed per unit volume of rock breaking T And the drilling rate V;

[0056] Specifically, drilling and core sampling are carried out in the borehole on site, and monitoring is conducted while drilling. This includes setting up multiple sensors on the drilling rig to acquire drilling parameters, such as oil pressure, drilling pressure, spindle speed, spindle torque, drilling displacement, and drilling time. The borehole information is also recorded on the mobile control terminal. The drilling parameters are acquired and collected in real time, and key rock-breaking indicators are calculated based on the data.

[0057] S200: Obtain the effective value of rock mass integrity coefficient Kv at different depths of the borehole;

[0058] Specifically, image acquisition can be performed inside the borehole, i.e., using video recording, to obtain the volumetric joint number Jv of the rock mass at different depths of the borehole. According to the specifications, the effective value of the whole-hole rock mass integrity coefficient Kv can be obtained by referring to the table.

[0059] S300: Based on the obtained key rock-breaking indicators, the effective value of the saturated uniaxial compressive strength Rc of rocks with different degrees of fragmentation is obtained; specifically, it includes the following steps:

[0060] S301: Select a complete or relatively complete section of rock core from the borehole as the reference rock mass, conduct indoor uniaxial compressive strength tests to obtain the saturated uniaxial compressive strength Rc value of the rock, and obtain the effective value of the saturated uniaxial compressive strength Rc of the rock through statistical methods;

[0061] S302: Based on artificial intelligence algorithms, such as support vector machines and artificial neural networks, establish key rock-breaking indicators (β). F β T By using a learning sample library of β and V and the effective values ​​of Rc, an Rc prediction model is obtained. The functional relationship of the model can be expressed as: Rc = F(β) F ,β T ,V);

[0062] S303: Using the predictive model obtained in S302, the key rock-breaking indicators (β) obtained from the non-benchmark rock mass segment are... F β T Using V as input, the effective value of Rc for non-benchmark rock mass segments is predicted and obtained, thereby obtaining the effective value of Rc for rock masses with different degrees of fracturing.

[0063] S400: Based on key rock-breaking indicators, effective values ​​of Rc and Kv, an artificial intelligence algorithm is used for deep learning to establish an Rc-Kv prediction model that can distinguish rock masses with different degrees of fracturing. The functional relationship of the Rc-Kv prediction model can be expressed as: G(Rc, Kv)=F(β) F ,β T ,V).

[0064] As per the instruction manual Figure 1 The diagram shows a flowchart of a method for determining the basic quality classification of rock mass in real time based on drilling parameters.

[0065] The following is a description using specific examples.

[0066] Example 1

[0067] As per the instruction manual Figure 3As shown, this invention provides a system for real-time determination of basic rock mass quality grading based on drilling parameters, comprising: a drilling monitoring system 1 for acquiring drilling parameters during drilling; a data acquisition instrument 2 connected to the drilling monitoring system 1 for acquiring and storing the drilling parameters; an image acquisition module for acquiring the effective values ​​of the rock mass integrity coefficient Kv at different depths of the borehole; and a mobile control terminal 3 connected to the data acquisition instrument 2 for data reception and processing, the mobile control terminal 3 including a data processing module; the data processing module is used to preprocess the drilling parameters, calculate key rock breaking indicators, then obtain the effective values ​​of the saturated uniaxial compressive strength Rc of rocks with different degrees of rock breaking based on the acquired key rock breaking indicators, and finally establish an Rc-Kv prediction model based on the key rock breaking indicators, the effective values ​​of Rc and Kv.

[0068] Specifically, the monitoring while drilling system is installed on the drilling rig 4. The drilling rig 4 includes a diesel engine 42, a mud pump 43, a feed cylinder 44, a gearbox 45, a drive shaft 46, and a chuck 47. The drive shaft 46 is connected to the gearbox 45. The chuck 47 is equipped with a vertical shaft 48 and a hydraulic transmission mechanism. The lower end of the vertical shaft 48 is equipped with a drill bit 41. The monitoring while drilling system 1 is installed on the drilling rig 4 and is used to acquire the parameters while drilling. Specifically, the monitoring while drilling system 1 includes a feed pressure sensor 11, a return oil pressure sensor 12, a torque sensor 13, a vertical shaft proximity switch 14, a drive shaft inductive switch 15, and a displacement sensor 16, all of which are connected to the data acquisition instrument 2. Each sensor is connected to the data acquisition instrument 2 via a cable 5. The feed pressure sensor 11 and the return oil pressure sensor 12 are installed on the feed cylinder 44 of the drilling rig 4, respectively, in the upper and lower chambers of the feed cylinder 44, and can measure the oil pipe pressures P1 and P2 passing through the upper and lower chambers, respectively; the torque sensor 13 is installed on the gearbox drive shaft 46 at the front end of the vertical shaft drive of the drilling rig 4, and is used to measure the output torque T1 of the gearbox 45; the vertical shaft proximity switch 14 is installed on the chuck 47 of the drilling rig 4, and can be bolted or glued to the chuck 47; a first magnet 17, which acts on the vertical shaft proximity switch 14, is installed on the vertical shaft 48 of the drilling rig, and is a strong magnet, which is glued to the vertical shaft 48 and uses electromagnetic... Based on the induction principle, when the vertical shaft proximity switch 14 approaches the first magnet 17, it receives a signal and measures the number N of pulses generated when the vertical shaft 48 rotates. The transmission shaft inductive switch 15 is installed on the gearbox 45 of the drilling rig. A second magnet 18, which interacts with the transmission shaft inductive switch 15, is installed on the transmission shaft 46 of the drilling rig. The second magnet 18 is a strong magnet and is attached to the transmission shaft. Using the principle of electromagnetic induction, it measures the number N1 of pulses generated when the transmission shaft rotates. The displacement sensor 16 is preferably a laser displacement sensor, which is mounted on the side of the chuck 47 through a bracket. It projects a laser vertically downward onto the reflector 19 on the transmission shaft 46. During drilling, as the hydraulic pressure drives the chuck to move up and down, displacement data d is collected and recorded every second.

[0069] All sensors in the drilling monitoring system are common industrial products. Sensors with corresponding ranges are selected according to different drilling rig models. All types of sensors used meet the "three-proof" product standards and are suitable for complex conditions at engineering survey sites.

[0070] In an optimized implementation, the data acquisition device 2 is preferably a multi-channel acquisition box device with its own charging power supply, which can collect and store time series signal data such as P1, P2, T1, N, N1, and d in real time. Preferably, the data acquisition device includes a wireless communication transmission module for transmitting data to a mobile control terminal for data processing in real time.

[0071] In an optimized implementation, the mobile control terminal 3 includes a tablet computer or mobile phone with specific data processing software. According to the method described in this invention, the mobile control terminal is used to receive data from the data acquisition instrument and perform data processing in real time, providing judgment results and conclusions, so that engineers can determine the basic quality classification of the rock mass.

[0072] Example 2

[0073] This invention also provides a method for determining the basic quality classification of rock mass in real time based on drilling parameters. This method refers to a data processing method and working method based on drilling parameters, specifically including the following data calculation methods and working principles:

[0074] (1) Vertical shaft rotational speed n:

[0075]

[0076] In the formula: N – the number of pulses of the vertical shaft proximity switch, one pulse corresponds to one rotation of the vertical shaft;

[0077] t—rotation time of vertical shaft, in minutes;

[0078] (2) Vertical shaft torque T:

[0079]

[0080] Where: η—transmission efficiency of the drilling rig rotation system, taken as an empirical value, not greater than 1;

[0081] n—vertical shaft rotation speed, calculated using formula (1);

[0082] n1—Drilling rig drive shaft speed, determined by formula... Calculate, where N1 is the number of pulses generated when the drive shaft rotates;

[0083] T1 – The output torque of the transmission, which is directly measured by torque sensor 13.

[0084] (3) Drilling pressure F:

[0085] F = G0 + P1 × S1 - P2 × S2 Formula (3)

[0086] In the formula, G0 is the weight of the drilling tool, which is obtained by weighing it directly on-site according to the drilling conditions and multiplying it by the acceleration due to gravity, in N.

[0087] P1 – Pressure in the upper chamber of the feed cylinder, directly measured by the feed pressure sensor 11, unit: Pa;

[0088] S1 – The area of ​​the upper chamber of the feed cylinder, a fixed value related to the cylinder structure, which can be found in the equipment manual, in meters (m²). 2 ;

[0089] P2 – Pressure in the lower chamber of the oil inlet cylinder, directly measured by the return oil pressure sensor 12, unit: Pa;

[0090] S2 – The area of ​​the lower chamber of the feed cylinder, a fixed value related to the cylinder structure, which can be found in the equipment manual, in meters (m²). 2 .

[0091] (4) Hole depth L and drilling rate V:

[0092] During the drilling process, the chuck reciprocates continuously in the up-down direction. Only the vertical displacement generated when the chuck is engaged with the vertical shaft and moves downward is the effective drilling displacement. When the chuck is released and the vertical shaft moves upward, it is only an auxiliary operation during the drilling process.

[0093] According to the operating principle of the monitoring while drilling system, if the displacement sensor collects one data point per second, then the m displacement data points recorded in time t are d1, d2, d3...d m When the displacement difference Δd = d m -d m-1 When the displacement difference Δd is greater than 0, the chuck is judged to be moving upward; when the displacement difference Δd = d m -d m-1 When Δd < 0, the chuck is determined to be moving downwards. Therefore, for Δd = d m -d m-1 The sum of the absolute values ​​of the values ​​less than 0 is the cumulative advance in time t1, which is the hole depth L1.

[0094]

[0095] Therefore, the drilling rate during the drilling process from hole depth L1 to hole depth L2 during the time interval Δt from time t1 to time t2 is:

[0096]

[0097] (5) Key indicators for rock breaking

[0098] According to the principles of rock fracture mechanics and the law of conservation of energy, under specific lithological conditions, rock drillability is the rock's ability to resist drill bit damage. Whether it is impact or rotary rock breaking, the interaction between the drill bit face and sides and the rock mass is ultimately manifested in the form of energy consumption. This is because the rock fracture process is related to the input energy, and the energy input to the rock mass during drilling mainly consists of two parts:

[0099] The work E done by effective drill pressure as the drill bit moves during drilling. F and the work E generated by the rotation of the drilling rig T .

[0100] The key rock-breaking indicators proposed in this invention include: the impact energy β consumed per unit volume of rock-breaking material. F Rotational energy β consumed per unit volume of rock breaking T And the drilling rate V. Based on the integrity of the rock mass, rock cores are classified as intact, relatively intact, relatively broken, broken, and extremely broken. Whether it is a relatively intact to intact rock core or a relatively broken to extremely broken rock core, the rock breaking process is completed gradually per unit volume and is related to the input energy. That is, the rock breaking process under different basic rock mass quality classifications is necessarily related to the impact energy and rotational energy per unit volume of rock breaking.

[0101] Key parameters for basic rock mass quality grading include the saturated uniaxial compressive strength Rc and the rock mass integrity coefficient Kv. Furthermore, it can be inferred that under specific lithological conditions, during drilling of intact to relatively intact rock masses, the value of the key parameter Rc for basic rock mass quality grading can be accurately obtained through laboratory experiments, such as uniaxial compressive strength tests, and its value is necessarily related to the key rock-breaking index (β). F β T There is a relationship between Rc and V. When drilling in fractured to extremely fractured rock masses, the Rc value is significantly different from that of intact rock masses. Moreover, the Rc value is difficult to obtain accurately directly through uniaxial compressive strength tests. The relationship between the Rc values ​​of fractured and intact rock masses is difficult to describe quantitatively through a formula. However, it is certain that when drilling in fractured rock masses, the key rock-breaking indicators will respond and change to some extent. These key rock-breaking indicators can be collected and acquired in real time through a drilling monitoring system, and the Rc value of rock masses under different fracture states can be predicted through these key rock-breaking indicators.

[0102] The calculation method and derivation process of the key rock-breaking indicators during the time interval Δt from time t1 to time t2 are as follows:

[0103]

[0104]

[0105] In formulas (6) and (7), V b The volume of the rock mass broken by the drill bit;

[0106]

[0107] In formula (8), D2 is the borehole outer diameter and D1 is the core outer diameter. During destructive drilling, D1 = 0. Their specific values ​​can be directly measured on site.

[0108] Substituting formula (8) into formulas (6) and (7), we get:

[0109]

[0110]

[0111] The key rock-breaking index β can be obtained using formulas 9 and 10. F and β T .

[0112] When calculating and obtaining key rock-breaking indicators using formulas 8, 9, and 10, the drill bit diameter parameter needs to be input.

[0113] (6) Key parameters for basic quality classification of rock mass

[0114] The key parameters for basic rock mass quality classification mainly include determining the saturated uniaxial compressive strength Rc and the rock mass integrity coefficient Kv. Traditional methods primarily involve uniaxial compressive strength tests, borehole wave velocity tests, and rock block wave velocity tests. When the degree of rock mass fragmentation makes it impossible to measure the Rc value, the point load strength index I is used instead. S(50) The traditional method of determining the basic rock mass quality grade (BQ) is time-consuming and labor-intensive, and requires highly specialized personnel. Obtaining compressive strength and rock mass wave velocity values ​​through laboratory experiments inevitably involves a certain lag. Therefore, this traditional method is not suitable for widespread application during extensive drilling operations. The calculated compressive strength values ​​often exhibit large numerical dispersion and mismatch with the actual rock mass compressive strength, leading to misjudgments or errors in the basic rock mass quality classification. Furthermore, manual operation may introduce errors, thus requiring highly specialized personnel.

[0115] The Rc values ​​in the learning samples described in this method are obtained through indoor uniaxial compressive strength tests on relatively complete and intact benchmark rock masses, and their effective Rc values ​​are obtained using statistical methods. Preferably, the effective Rc values ​​are statistical standard values ​​of more than 6 values. It is required that the statistical data samples for solving both parameters both need to exceed 6, and the coefficient of variation should be less than or equal to 0.3. The statistical calculation method is as follows:

[0116] average value: Where n is the number of statistical data items, n≥6 Formula (11)

[0117] Standard value:

[0118] Coefficient of variation:

[0119] Correction factor:

[0120] The effective value of Rc, a key parameter for basic rock mass quality classification, is as follows:

[0121] in For the statistical average, formula (15)

[0122] It should be noted that when the coefficient of variation is greater than 0.3, data with large differences should be removed according to the standard deviation multiple until the coefficient of variation is less than or equal to 0.3.

[0123] The effective Kv value in the learning sample described in this method is obtained by automatically identifying and capturing the number of joints and fractures in the rock mass throughout the borehole using borehole camera technology, obtaining the volumetric joint number Jv of the rock mass, and then looking up the Kv value in the table according to the specifications.

[0124] Therefore, the BQ value can be calculated according to the standard formula:

[0125] BQ = 100 + 3Rc + 250Kv (Formula 16)

[0126] When Rc > 90Kv + 30, Rc = 90Kv + 30 and Kv should be substituted into the calculation of BQ value;

[0127] When Kv > 0.04Rc + 0.4, Kv = 0.04Rc + 0.4 and Rc should be substituted into the calculation of the BQ value.

[0128] The on-site operation steps of the method of this invention are as follows (see attached instruction manual). Figure 2 As shown, the details are as follows:

[0129] 1) Real-time drilling and core sampling are conducted in the borehole. Drilling monitoring is performed using a monitoring system to obtain drilling parameters such as P1, P2, T, and n. Borehole information, such as D2 and D1, is entered into the mobile control terminal. Key rock-breaking indicators (β) are calculated according to formulas 9 and 10. F β T (V), and obtain its specific distribution law with time and hole depth in real time;

[0130] 2) Conduct borehole photography to obtain the rock mass volume joint number Jv at different depths of the borehole, and obtain the distribution of the effective value of Kv for the whole borehole by referring to the table;

[0131] 3) Select a complete or relatively complete section of rock core from the borehole as the reference rock mass, conduct indoor uniaxial compressive strength tests to obtain the saturated uniaxial compressive strength Rc value of the rock, and use formula 11-15 to statistically obtain the effective value of the saturated uniaxial compressive strength Rc of the rock.

[0132] 4) Based on artificial intelligence algorithms, such as support vector machines and artificial neural networks, establish key rock-breaking indicators (β). F β T By using a learning sample library of β and V and the effective values ​​of Rc, an Rc prediction model is obtained. The functional relationship of the model can be expressed as: Rc = F(β) F ,β T ,V);

[0133] 5) Using the prediction model obtained in step 4, the key rock-breaking indicators (β) obtained from the non-benchmark rock mass segment are... F β T Using V as input, the effective value of Rc for non-benchmark rock mass segments is predicted and obtained, thereby obtaining the effective value of Rc for rock masses with different degrees of fracturing.

[0134] 6) Based on key rock-breaking indicators, effective values ​​of Rc and Kv, an artificial intelligence algorithm is used for deep learning to establish an Rc-Kv prediction model that can distinguish rock masses with different degrees of fracturing. The functional relationship of the Rc-Kv prediction model can be expressed as: G(Rc, Kv)=F(β F ,β T ,V);

[0135] 7) Drill other boreholes with the same lithology. Based on the prediction model obtained in step 6, obtain the Rc and Kv values ​​of the rock mass in the borehole in real time, and calculate the basic quality index of the rock mass according to formula 16.

[0136] In the same borehole, the lithology of the reference section and the non-reference section is basically the same, only the degree of fracturing is different. Therefore, the method proposed in this invention, based on the laws of energy conservation and rock breaking mechanism, after establishing the relationship between the key rock breaking parameters and the Rc value, the prediction model can also be adapted to the Rc discrimination of the non-reference section.

[0137] At this time, because of the input quantity (β) of the non-reference segment F β T The values ​​of Rc (V) have been obtained during drilling. After being input into the model, the Rc value can be predicted. Provided the learning samples are representative, and the accuracy of the Rc prediction model established based on the benchmark core section is guaranteed, the direct prediction results are reliable.

[0138] To better illustrate the method described in this invention, an example is given: For instance, in the case of Carboniferous limestone formations, in order to obtain the basic quality classification index (BQ) value of the drilled hard rock formation in real time, an Rc prediction model and an Rc-Kv prediction model applicable to limestone formations in this region can be established based on the method described.

[0139] For the core sample from the benchmark rock mass section inside the borehole, indoor compressive strength tests were conducted. After statistical analysis, the effective value of Rc was obtained. The statistical process of one of the effective Rc values ​​of 34.08 MPa used in the learning sample library of the Rc prediction model in the example is shown in Table 1:

[0140] Table 1. Statistical Calculation of Rc Effective Values

[0141] Sample number Saturated compressive strength value Rc (MPa) name 1 35.90 limestone 2 43.70 limestone 3 45.70 limestone 4 46.60 limestone 5 25.80 limestone 6 35.30 limestone 7 46.00 limestone Number of samples 7.00 average value 39.86 Standard deviation 7.81 coefficient of variation 0.20 Correction coefficient 0.86 Standard value 34.08

[0142] In this example, the data processing method used is the widely applied BP neural network, with the key rock-breaking index (β) as the primary parameter. F β T The relationship between V and Rc is non-linear. The model consists of an input layer, an output layer, and hidden layers. Each node layer includes a certain number of neurons. Nodes within the same node layer are not interconnected. The model structure is shown in the appendix of the instruction manual. Figure 4 As shown.

[0143] Based on drilling monitoring of the Carboniferous limestone strata, and through repeated testing, the number of hidden nodes in the neural network was determined to be 8. The tangent sigmoid function was selected as the activation function, and an adaptive learning rate momentum gradient descent backpropagation algorithm was employed to improve the network's training efficiency. First, a certain amount of measured or experimental data is provided to the network as samples. Through learning and training on this data, the network establishes the mapping relationship between these influencing factors, thereby enabling the network to have predictive capabilities. Taking the calculation of the Rc value as an example, 20 sets of data obtained from monitoring were used as training samples, as shown in Table 2 below:

[0144] Table 2 Training Samples for Rc Prediction Model

[0145]

[0146]

[0147] A BP neural network model for predicting Rc values ​​in Carboniferous limestone was established using deep learning. This model was imported into the mobile control terminal software and applied during drilling monitoring to predict Rc values ​​in non-benchmark rock masses. Here, non-benchmark rock masses refer to core samples other than intact or relatively intact sections. Some of the prediction results are shown in Table 3 below.

[0148] Table 3 Results of Rc Effective Value Judgment in Non-Benchmark Rock Mass Sections

[0149] Serial Number <![CDATA[β F ]]> <![CDATA[β T ]]> V Rc 1 0.647971 1.743898 0.023 32.67 2 0.669707 1.857532 0.015 29.99 3 0.497628 3.604615 0.014 52.57 4 0.472269 4.216573 0.012 39.53 5 0.67333 2.527394 0.023 55.22 6 0.588196 2.41407 0.022 54.87 7 0.508496 1.987238 0.024 38.56 8 0.570082 2.725981 0.02 46.79

[0150] Furthermore, by utilizing a BP neural network and relying on key rock-breaking indicators (β) from multiple boreholes... F β T Based on the effective values ​​of Rc, Kv, and V, an Rc-Kv prediction model can be established. The 40 training samples are shown in Table 4 below:

[0151] Table 4 Training Samples for the Rc-Kv Prediction Model

[0152]

[0153]

[0154] When drilling is carried out in other boreholes within the work area, the basic quality classification of the limestone strata in the borehole can be predicted in real time during the in-situ monitoring process, as shown in Table 5 below:

[0155] Table 5 Real-time Judgment Results of Basic Quality Classification (BQ) of Rock Mass

[0156] Serial Number Starting depth Termination depth Rc value (MPa) Kv BQ Basic classification of rock mass quality 1 30.06 30.47 22.56 0.32 247.68 Ⅴ 2 30.47 30.59 35.37 0.35 293.61 Ⅳ 3 30.59 30.84 63.08 0.72 469.24 Ⅱ 4 30.84 30.95 55.78 0.69 439.84 Ⅲ 5 30.95 31.14 49.55 0.67 416.15 Ⅲ 6 31.14 31.3 36.87 0.4 310.61 Ⅳ 7 31.3 31.48 49.74 0.53 381.72 Ⅲ 8 31.48 31.63 34.81 0.54 339.43 Ⅳ 9 31.63 31.92 40.63 0.38 316.89 Ⅳ 10 31.92 32.03 32.18 0.56 336.54 Ⅳ

[0157] The above examples only illustrate the basic quality classification of limestone strata. The discrimination method and system of this invention are not limited to any particular lithology. Prediction models for key parameters of basic quality classification of other lithologies can be implemented using the above method.

[0158] According to the method and system for determining the basic quality grade of rock mass in real time based on drilling parameters provided by the present invention, the basic quality grade of the rock mass being drilled in the borehole can be confirmed in real time and in situ in a mobile control terminal, which is convenient for on-site implementation and provides data guidance for real-time drilling.

[0159] In summary, the method and system for real-time determination of basic quality grading of borehole rock mass based on drilling parameters provided by this invention involves setting up a drilling monitoring system on the drilling rig to acquire drilling parameters during the drilling process, calculating key rock-breaking indicators based on these parameters, performing video recording inside the borehole to acquire the volume joint number Jv of the rock mass at different depths, obtaining the effective value of Kv for the entire borehole by looking up a table, conducting indoor uniaxial compressive strength tests on a benchmark rock mass, and obtaining the effective value of Rc through statistical calculations, establishing a learning sample library of key rock-breaking indicators and effective Rc values ​​based on artificial intelligence algorithms to obtain an Rc prediction model; importing key rock-breaking indicators acquired from non-benchmark rock mass segments based on the Rc prediction model to predict their effective Rc values, thus obtaining effective Rc values ​​for different degrees of fragmentation; furthermore, establishing an Rc-Kv prediction model that can distinguish rock masses with different degrees of fragmentation, allowing for real-time in-situ acquisition of key parameters Rc and Kv values ​​for basic quality grading of borehole rock mass during real-time drilling, and calculating basic rock mass quality grading data. The method of this invention is based on the law of conservation of energy and the rock breaking mechanism, and proposes key rock breaking indicators. The selected parameters are comprehensive and scientifically reasonable, which improves the accuracy of the prediction model. By obtaining the basic quality classification of the rock mass in real time through the prediction model, the problem of the lag in obtaining the basic quality classification of the rock mass by conducting various tests is solved, and reference suggestions can be quickly and directly provided on site for engineering construction.

[0160] Those skilled in the art will understand that the present invention can be implemented in many other specific forms without departing from the spirit and scope of the invention. Although embodiments of the invention have been described, it should be understood that the invention is not limited to these embodiments, and those skilled in the art can make changes and modifications within the spirit and scope of the invention as defined in the appended claims.

Claims

1. A method for real-time determination of basic rock mass quality classification based on drilling parameters, characterized in that, Includes the following steps: Establish an Rc-Kv prediction model that can distinguish rock masses with different degrees of fragmentation; Drilling is conducted into the rock mass, and key rock-breaking indicators are collected during the drilling process. These key indicators include the impact energy consumed per unit volume of rock broken. Rotational energy consumed per unit volume of rock breaking and drilling rate V Then, based on the obtained Rc-Kv prediction model, the Rc and Kv values ​​of the rock mass inside the borehole are obtained in real time, and the basic quality classification of the rock mass is calculated. The method for establishing the Rc-Kv prediction model includes: S100: During the drilling process, core samples are taken and drilling parameters are collected. The drilling parameters are preprocessed to calculate key rock-breaking indicators. S200: Obtain the effective value of rock mass integrity coefficient Kv at different depths of the borehole; S300: Based on the obtained key rock breaking indicators, the effective value of the saturated uniaxial compressive strength Rc of rocks with different degrees of fragmentation is obtained. S400: Based on the key rock-breaking indicators, the effective value of Rc, and the effective value of Kv, an Rc-Kv prediction model is established. Specifically, the Rc-Kv prediction model represents the functional relationship between the effective values ​​of Rc and Kv and the key rock-breaking indicators. This functional relationship is expressed as follows: ; Step S100 includes the following steps: S101: Obtain drilling parameters, including oil pressure, drilling pressure, spindle speed, spindle torque, drilling displacement, and drilling time; S102: Input borehole information on the mobile control terminal; the borehole information includes borehole outer diameter, core outer diameter, and drill bit diameter; S103: Based on the obtained drilling parameters and borehole information, the key rock-breaking indicators are calculated, and the specific distribution patterns of the key rock-breaking indicators with time and borehole depth are obtained in real time.

2. The method for real-time determination of basic rock mass quality classification based on drilling parameters according to claim 1, characterized in that, The step S200 of obtaining the effective values ​​of the rock mass integrity coefficient Kv at different borehole depths includes: Camera work was carried out inside the borehole to obtain the volume joint number Jv of the rock mass at different depths of the borehole; by looking up the table, the effective value of the whole borehole rock mass integrity coefficient Kv was obtained.

3. The method for real-time determination of basic rock mass quality classification based on drilling parameters according to claim 1, characterized in that, The step S300, which involves obtaining the effective value of the saturated uniaxial compressive strength Rc of rocks with different degrees of fragmentation, includes: S301: Select a reference rock mass, conduct a compressive strength test, and obtain the effective value of the saturated uniaxial compressive strength Rc of the reference rock mass through statistical calculation; S302: Establish a learning sample library of key rock breaking indicators and effective Rc values ​​to obtain an Rc prediction model; S303: Using the prediction model obtained in S302, the effective value of Rc is obtained by predicting the key rock breaking indicators of non-benchmark rock mass segments, and the effective value of Rc of rock masses with different degrees of fracturing is obtained.

4. The method for real-time determination of basic rock mass quality classification based on drilling parameters according to claim 3, characterized in that, In step S301, the reference rock mass is a complete or relatively complete rock core. The saturated uniaxial compressive strength Rc value of the rock is obtained by indoor uniaxial compressive strength test, and the effective value of the saturated uniaxial compressive strength Rc of the rock is obtained by statistical method.

5. The method for real-time determination of basic rock mass quality classification based on drilling parameters according to claim 1, characterized in that, Step S400 includes establishing a learning sample library of key rock-breaking indicators, effective values ​​of Rc and Kv based on the acquired rock-breaking key indicators, effective values ​​of Rc and Kv, using artificial intelligence algorithms, and obtaining the Rc-Kv prediction model.

6. A system for implementing the method for real-time determination of basic rock mass quality classification based on drilling parameters as described in any one of claims 1-5, characterized in that, The system includes: A monitoring-while-drilling system is used to acquire drilling parameters during the drilling process; A data acquisition instrument, which is connected to the drilling monitoring system, is used to collect and store drilling parameters; The image acquisition module is used to obtain the effective value of the rock mass integrity coefficient Kv at different depths of the borehole; A mobile control terminal, which is connected to the data acquisition instrument for data reception and data processing, the mobile control terminal includes a data processing module; The data processing module is used to preprocess the drilling parameters, calculate the key rock breaking indicators, obtain the effective value of the saturated uniaxial compressive strength Rc of rocks with different rock breaking degrees based on the obtained key rock breaking indicators, and finally establish an Rc-Kv prediction model based on the key rock breaking indicators, the effective value of Rc and the effective value of Kv.

7. The system for real-time determination of basic rock mass quality classification based on drilling parameters according to claim 6, characterized in that, The drilling monitoring system includes a feed pressure sensor, a return oil pressure sensor, a torque sensor, a vertical shaft proximity switch, a drive shaft inductive switch, and a displacement sensor, all connected to the data acquisition instrument. The feed pressure sensor and the return oil pressure sensor are mounted on the feed cylinder of the drilling rig, the torque sensor is mounted on the drive shaft of the drilling rig's gearbox, the vertical shaft proximity switch is mounted on the drilling rig's chuck, the drive shaft inductive switch is mounted on the drilling rig's gearbox, and the displacement sensor is mounted on the drilling rig's chuck to acquire drilling displacement data.

8. The system for real-time determination of basic rock mass quality classification based on drilling parameters according to claim 6, characterized in that, The data acquisition device includes a wireless communication transmission module for transmitting data to the mobile control terminal in real time.

Images