Dynamic control method and device for petroleum drilling trajectory

Through machine learning technology, the well angle and azimuth of the drill bit position are predicted in real time, and the problems of high cost of rotary guided directional drilling and low efficiency of curved screw directional drilling in the existing technology are solved, and the precise control and drilling of the wellbore trajectory are achieved. The increase in the inlet speed reduces drill bit damage and cost.

CN118933708APending Publication Date: 2024-11-12INTERCONTINENTAL STRAIT ENERGY TECH CO LTD +1
View PDF 0 Cites 6 Cited by

Patent Information

Application Number
CN202411362573.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-27
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

In the existing dynamic control technology of oil drilling trajectory, the cost of rotary guided directional drilling is high, the speed is slow, and the risk is high. The oblique angle measurement of the curved screw directional drilling well is inaccurate, the efficiency is low, and the drill bit is easy to be damaged, and there is a lack of cost reduction and efficiency improvement. plan.

Method used

Using machine learning method, the well angle and azimuth of the drill bit position are predicted in real time based on dynamic trajectory control data. Through data acquisition, state determination, data preprocessing, model training and trajectory control parameter recommendation, the precise control of the wellbore trajectory is achieved and the improvement of the Drilling speed and reduce drill bit damage.

Benefits of technology

Without the use of rotary guides, precise control of the wellbore trajectory is achieved, drilling speed is improved, cost is reduced, drill bit damage is reduced, and drilling efficiency and safety is improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118933708A_ABST
    Figure CN118933708A_ABST
Patent Text Reader

Abstract

The invention discloses a petroleum drilling trajectory dynamic control method and device. The method comprises the steps that dynamic trajectory control data at the same depth are collected; the track state parameter change rate of the current drilling state is calculated, and a prediction model corresponding to the current drilling state is obtained through training; predicting the trajectory state parameter change rate of the new drilling well section of the current drill bit by using the prediction model, and calculating the trajectory state parameter of the position where the current drill bit is located according to the predicted value of the trajectory state parameter change rate; by taking the trajectory state parameter of the position where the current drill bit is located as a benchmark, the target trajectory state parameter as a target function and the maximum value of the trajectory state parameter change rate corresponding to the unit footage of the well section as a constraint condition, performing dynamic solution by using a prediction model corresponding to the current drilling state to obtain at least one group of recommended trajectory control parameters; and calculating the drilling distance reaching the target trajectory state parameter so as to carry out drilling trajectory control. According to the invention, accurate control of the well track can be realized.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of oil drilling, and in particular to a method and device for dynamically controlling an oil drilling trajectory. Background Art

[0002] This section is intended to provide a background or context to the embodiments of the invention recited in the claims. No admission is made that the description herein is prior art by inclusion in this section.

[0003] There are two methods in the existing oil drilling trajectory dynamic control technology:

[0004] The first type: rotary directional drilling

[0005] Directional drilling is performed using rotary steerable equipment. Instructions are given on the ground through pump pressure pulses of circulating mud or through drilling tool rotation and drilling speed changes, and control, measurement and liquid pulse feedback are performed at the bottom of the well through rotary steerable equipment.

[0006] The disadvantages of rotary directional drilling are:

[0007] (a) High cost

[0008] At present, the comprehensive cost of using rotary guide in a well in a certain oil field is 170,000 yuan / day. The drilling cycle of the deflection section of a well is generally one week, and rotary guide may also be used in the horizontal section. The average cost of using rotary guide in a well is more than 1 million yuan.

[0009] (b) Slow drilling

[0010] Due to the limitation of rotary guide equipment, the pressure and drilling speed applied from the wellhead are limited, thus affecting the high-speed drilling of the drill bit. For example, for a φ215.9mm wellbore, it is generally impossible to apply 150kN to 200kN drilling pressure to the rotary guide, but some formations with poor drillability must use high drilling pressure and high speed to be effective.

[0011] (c) High risk

[0012] For wellbores that may experience serious downhole leakage or well sections that are prone to collapse in the broken zone, complications such as drill sticking may occur, or even accidents such as buried rotary guide instruments may occur, resulting in serious losses.

[0013] (d) Limited conditions of use

[0014] For example, the high temperature environment at the bottom of the well may cause the electronic components of the rotary guide to fail and become unusable.

[0015] The second type: bent screw directional drilling

[0016] The proportion of horizontal wells in oilfield development is increasing. At present, most of the horizontal well drilling processes use the "bent screw + MWD" drilling tool combination for trajectory control. The static well inclination method is usually used for trajectory control. The general process is as follows:

[0017] (a) On-site static inclination measurement: The static well inclination angle is the well inclination angle data measured when the drilling tool is in a static state. During the on-site top drive drilling process, after drilling each column (3 single columns, a total of about 28.5m), the drill bit is lifted off the bottom of the well to measure the static well inclination angle and azimuth. The measurement process is to stop the pump for 5 minutes, then start the pump to the circulating displacement, and wait for the bottom hole measurement data to be transmitted to the ground, which is expected to take about 10 to 15 minutes. It usually takes more than 30 minutes to complete a static inclination measurement on site.

[0018] (b) Directional engineers use static measurement data to analyze the drilling trajectory calculation software to predict the trend of the bottom hole trajectory change. According to the designed profile, they select the subsequent drilling method (sliding directional drilling or compound drilling) and drilling parameters.

[0019] (c) Static measurement of well inclination is affected by factors such as mud medium, mud pump efficiency, and driller operation. Sometimes the inclination measurement may not be successful, and it is necessary to measure again according to the above steps.

[0020] The current disadvantages of bent screw directional drilling are:

[0021] (1) Inaccurate measurement of well inclination

[0022] Since the measuring point of the drilling tool combination of "PDC drill bit + bent screw + resistivity + MWD +..." is generally 18m-22m away from the drill bit position, the measured data cannot immediately reflect the actual situation of the drill bit position, and the measures taken are delayed, affecting the precise control of the trajectory.

[0023] (2) Large interval between well inclination angle measurements

[0024] Generally, measurement is performed after a column is completed. However, the length of columns on land and offshore drilling rigs ranges from 19 to 30 meters. The distance between each measurement point is large, and problems are discovered too late. Although MWD can send the current well inclination and azimuth in real time, this dynamic information is generally not effectively used.

[0025] (3) Low efficiency of directional sliding drilling

[0026] Due to the above reasons, the bottom hole trajectory change trend predicted by the directional engineer is sometimes inaccurate. When it is found that the deviation between the trajectory and the designed trajectory is too large, it is necessary to correct the deviation urgently, and then abandon the composite drilling and switch to directional sliding drilling. As a result, during construction, it is often necessary to increase the directional inclination for a while and then reduce the directional inclination for a while, and the wellbore is tortuous and not smooth. Frequent directional sliding drilling is due to the slow speed of the machine due to the failure of the drill bit to rotate, and there is a risk of sticking. In order to eliminate the pressure and prevent the drill from getting stuck, the drill bit needs to be moved frequently, and the drilling efficiency is low. At the same time, during directional sliding drilling, the tool face will swing greatly and become unstable due to the fluctuation of the screw torque. The tool face is difficult to control, resulting in low directional efficiency.

[0027] (4) Directional sliding drilling can easily damage the drill bit

[0028] During directional sliding drilling, due to the large static friction, in order to push the drill pipe forward, a large drilling pressure is sometimes applied to overcome the static friction. Once the drill pipe starts to slide, the dynamic friction decreases rapidly, and the drill bit may collide violently with the formation, resulting in abnormal damage to the drill bit. When the drill bit is damaged, the drilling speed will be affected and may further lead to unnecessary tripping. Each trip requires all the drilling tools in the well to be taken out, and the drill bit must be replaced and re-inserted, which will waste more than 30-50 hours of effective working time depending on the depth of the well.

[0029] (5) Subjective factors have a greater impact

[0030] The actual work is based on the requirements of the geological guidance engineer, and the degree of accuracy should be achieved within a certain number of meters. The work effect is related to the directional engineer.

[0031] Therefore, there is currently a lack of a solution to reduce drilling costs while improving the efficiency of current construction methods. Summary of the invention

[0032] In the first aspect, an embodiment of the present invention provides a method for dynamically controlling a petroleum drilling trajectory, which can predict the trajectory state parameter change rate (well inclination, azimuth, etc.) of the current drill bit position in real time without using a rotary guide but a bent screw drill, thereby achieving precise control of the wellbore trajectory, improving the drilling speed, reducing drill bit damage, and thus reducing costs and increasing efficiency. The method includes:

[0033] Collect dynamic trajectory control data at the same depth;

[0034] Determine the current drilling status based on dynamic trajectory control data;

[0035] According to the dynamic trajectory control data, the trajectory state parameter change rate of the current drilling state is calculated;

[0036] Taking the characteristic values ​​in the dynamic trajectory control data as input and the trajectory state parameter change rate of the current drilling state as output, training is performed using machine learning to obtain a prediction model corresponding to the current drilling state;

[0037] According to the dynamic trajectory control data sequence and trajectory state parameter change rate sequence of the current section, a prediction model is used to predict the trajectory state parameter change rate of the new well section drilled by the current drill bit, and according to the predicted trajectory state parameter change rate prediction value, the trajectory state parameter at the current drill bit position is calculated;

[0038] Taking the trajectory state parameters at the current drill bit position as the benchmark, the target trajectory state parameters as the objective function, and the maximum value of the trajectory state parameter change rate corresponding to the unit footage of the well section as the constraint condition, the prediction model corresponding to the current drilling state is used for dynamic solution to obtain at least one set of recommended trajectory control parameters and their corresponding trajectory state parameter change rates, and calculate the drilling distance to reach the target trajectory state parameters;

[0039] The drilling trajectory is controlled according to the trajectory control parameters and the drilling distance.

[0040] In the second aspect, the embodiment of the present invention also provides a petroleum drilling trajectory dynamic control device, which can predict the well inclination angle, azimuth and other parameters of the current drill bit position in real time without using a rotary guide but a bent screw drill, so as to achieve precise control of the wellbore trajectory, increase the drilling speed, reduce the damage to the drill bit, thereby reducing costs and increasing efficiency. The device includes:

[0041] Data acquisition module, used to collect dynamic trajectory control data at the same depth;

[0042] A drilling state determination module is used to determine the current drilling state according to the dynamic trajectory control data;

[0043] A data preprocessing module is used to calculate the trajectory state parameter change rate of the current drilling state according to the dynamic trajectory control data;

[0044] A model training module is used to take the characteristic values ​​in the dynamic trajectory control data as input and the trajectory state parameter change rate of the current drilling state as output, and to perform training using a machine learning method to obtain a prediction model corresponding to the current drilling state;

[0045] The drill bit inclination angle prediction module is used to predict the trajectory state parameter change rate of the new well section drilled by the current drill bit based on the dynamic trajectory control data sequence and trajectory state parameter change rate sequence of the current section, and calculate the trajectory state parameter at the current drill bit location based on the predicted trajectory state parameter change rate prediction value;

[0046] The trajectory control parameter recommendation module is used to take the trajectory state parameters of the current drill bit position as the benchmark, the target trajectory state parameters as the objective function, and the maximum value of the trajectory state parameter change rate corresponding to the unit footage of the well section as the constraint condition, and use the prediction model corresponding to the current drilling state to perform dynamic solution, obtain at least one set of recommended trajectory control parameters and their corresponding trajectory state parameter change rate, and calculate the drilling distance to reach the target trajectory state parameters;

[0047] The trajectory control module is used to control the drilling trajectory according to the trajectory control parameters and the drilling distance.

[0048] In a third aspect, an embodiment of the present invention further provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned oil drilling trajectory dynamic control method when executing the computer program.

[0049] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned oil drilling trajectory dynamic control method is implemented.

[0050] In a fifth aspect, an embodiment of the present invention further provides a computer program product, wherein the computer program product includes a computer program, and when the computer program is executed by a processor, the above-mentioned oil drilling trajectory dynamic control method is implemented.

[0051] In an embodiment of the present invention, dynamic trajectory control data at the same depth are collected; the current drilling state is determined according to the dynamic trajectory control data; the trajectory state parameter change rate of the current drilling state is calculated according to the dynamic trajectory control data; the characteristic value in the dynamic trajectory control data is used as input, the trajectory state parameter change rate of the current drilling state is used as output, and training is performed in a machine learning manner to obtain a prediction model corresponding to the current drilling state; according to the dynamic trajectory control data sequence and the trajectory state parameter change rate sequence of the current section, the prediction model is used to predict the trajectory state parameter change rate of the new drilling section of the current drill bit, and the trajectory state parameter of the current drill bit position is calculated according to the predicted trajectory state parameter change rate prediction value; the trajectory state parameter of the current drill bit position is used as a reference, the target trajectory state parameter is used as an objective function, and the maximum value of the trajectory state parameter change rate corresponding to the unit footage of the section is used as a constraint condition, and the prediction model corresponding to the current drilling state is used to perform dynamic solution to obtain at least one set of recommended trajectory control parameters and their corresponding trajectory state parameter change rates, and the drilling distance to reach the target trajectory state parameters is calculated; the drilling trajectory control is performed according to the trajectory control parameters and the drilling distance. Compared with the solutions of rotary guide directional drilling and bent screw directional drilling in the prior art, the embodiment of the present invention uses the characteristic values ​​in the dynamic trajectory control data corresponding to the current drilling state as input, and uses the trajectory state parameter change rate of the current drilling state as output, and trains the corresponding prediction model, which can be used to predict the trajectory state parameters of the current drill bit position and recommend trajectory control parameters, so as to calculate the accurate drilling distance, thereby effectively controlling the drilling trajectory, improving the drilling speed, reducing drill bit damage, and thus reducing costs and increasing efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0053] Figure 1 Flow chart of a method for dynamic control of oil drilling trajectory in an embodiment of the present invention;

[0054] Figure 2 The effect diagram of implementing two kinds of screw drilling tools to achieve full-section drilling in the present invention;

[0055] Figure 3 This is a drilling chart of the three-opening horizontal section of a well in the embodiment of the invention;

[0056] Figure 4 A schematic diagram of the application of the method proposed in an embodiment of the present invention in a certain well;

[0057] Figure 5 This is a chart showing the dynamic well inclination angle, drilling pressure and rotation speed changes at 4760m-4900m in a well in an embodiment of the present invention;

[0058] Figure 6 Schematic diagram of a dynamic control device for oil drilling trajectory in an embodiment of the present invention;

[0059] Figure 7 Schematic diagram of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION

[0060] To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings. Here, the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0061] First, the terms involved in the embodiments of the present invention are explained.

[0062] Wellbore trajectory: In oil drilling, according to different geological and engineering needs (for example, to ensure that the wellbore moves along the structural undulations of the oil and gas layer to achieve higher production), it is often necessary to manually adjust the wellbore trajectory, commonly known as "orientation". During the drilling process, it is necessary to adjust the wellbore inclination and azimuth of the wellbore extension. The change in wellbore inclination per unit drilling distance is called the wellbore inclination change rate, and the same is true for the azimuth change rate. The adjustment method is called "inclination", including increasing inclination (increasing wellbore inclination), decreasing inclination (reducing wellbore inclination), and stabilizing inclination (keeping the current wellbore inclination unchanged). In order to achieve inclination, two methods are generally used: rotary guide equipment (rotary guide) or bent screw drilling tools.

[0063] Well section: When a horizontal well is drilled, the wellbore trajectory is generally vertical, and this section is called the vertical well section; when the fast drilling reaches the target layer, the wellbore trajectory will gradually flatten and remain parallel to the target layer. The well section with increased inclination angle from the vertical well section to the horizontal direction is called the inclination section; after entering the target layer, the wellbore trajectory remains in the target layer, and this section is called the horizontal section (with an inclination angle of about 90°);

[0064] Rotary guide: Rotary guide is a precious precision instrument, easy to damage (because of a large number of mechanical and electronic components), and has harsh operating conditions (such as applied pressure, bottom hole temperature, etc.). After it goes down to the bottom of the well and sends instructions from the ground, the rotary guide tool can actively control the direction of the drill bit to achieve accurate drilling of the wellbore trajectory to the expected azimuth and well inclination.

[0065] Screw Drilling Tools: Screw Drilling Tools are downhole power drilling tools that use the drilling fluid circulating in the wellbore as a driving force to convert the liquid pressure energy into mechanical energy. While the drill pipe rotates, it can provide additional rotational force for the drill bit. It is simple to use, cheap, stable in performance, and highly efficient, but the control of the wellbore trajectory is not as precise as the rotary guide.

[0066] Bent screw drill: The front section of the bent screw drill has a certain bend angle (such as 1°, 1.25°, 1.5°, 1.75°, etc.), which can be used to adjust the well inclination and azimuth through directional drilling. The use of bent screw drills includes two methods: composite drilling and directional sliding drilling. Composite drilling refers to the simultaneous rotation of the drill pipe and the screw drill, which is faster; directional sliding drilling refers to the rotation of only the screw rotor and the drill pipe does not rotate, and the well inclination and azimuth are controlled by the bend angle of the screw drill. Generally, directional sliding drilling is used when the well inclination and azimuth need to be adjusted rapidly and significantly. In directional sliding drilling, the elbow of the screw is aligned with a certain direction (commonly known as the swing tool face), which is called the tool face, and can drill in the specified direction more effectively. In the process of composite drilling, the bent screw drill sends the current well inclination and azimuth to the ground through the MWD (wireless inclinometer while drilling); in directional sliding drilling, the tool face data will also be sent dynamically, and the tool face will swing at this time.

[0067] Inclination measurement: The above-mentioned inclination angle measured during the drilling process is called dynamic inclination measurement. Because the equipment is in continuous motion during the measurement process, there will be errors in dynamic inclination measurement. The drilling tool stops rotating and the drilling fluid circulates, and the downhole drilling tool stops for measurement, which is called measuring the static inclination angle. The distance of static inclination measurement varies according to different situations. At present, when there is a top drive at land and offshore drilling sites, the spacing is generally the drill pipe column, and the length is about 28.5m (generally, a single drill pipe is about 9.5m); for onshore drilling rigs without top drives, the spacing is generally a single drill pipe, and the length is about 9.5m. Static inclination angle is more accurate, but because of intermittent downtime (it usually takes more than 30 minutes to complete a static inclination measurement on site), it will affect production efficiency. The above data are generally measured by a dedicated measuring short section-wireless measurement while drilling, namely MWD. ​​MWD is located above the screw drill tool, behind the non-magnetic drill pipe or non-magnetic drill collar, away from the drill bit, generally reaching a distance of 18-22m.

[0068] The inventor believes that the deflection law under the same parameters will be different in different well areas, different stratum lithology, different bent screw combinations (including the position, number, diameter, and type of stabilizers), and with the wear of the drill bit (PDC) and the stabilizer. If it is particularly difficult to establish a physical model, involving geomechanics, rock mechanics, drilling tool mechanics, etc., and the amount of calculation is large, the calculation time of conventional machines may be measured in hours.

[0069] From the basic logic, in the same drilling trip, the drill bit combination has not changed, and the geological conditions, drilling fluid properties, and wellbore trajectory generally change little, so these influencing factors can be ignored. The actual inclination situation is highly correlated with the current construction parameters. In compound drilling, the inclination angle and azimuth change rate also change under different drilling pressures and rotation speeds; in sliding drilling, the inclination angle, azimuth change rate and tool face swing amplitude are different under different drilling pressures. Among them, since the screw drives the drill bit to rotate, the displacement is also a related parameter, but it is weakly correlated.

[0070] The principle of the embodiment of the present invention is to use an artificial intelligence method to perform continuous machine learning based on dynamic trajectory control data according to the drilling footage depth (such as 1 meter) or time interval (such as 55 seconds) for composite drilling and sliding drilling conditions when using a bent screw. The well inclination and azimuth of the current drill bit position are predicted in real time based on the learned AI model and current construction parameters, and correction and error analysis are performed in combination with the static well inclination data collected subsequently, so as to achieve more accurate empirical trajectory closed-loop control under all working conditions of the whole well.

[0071] Figure 1 The flowchart of the oil drilling trajectory dynamic control method in the embodiment of the present invention includes:

[0072] Step 101, collecting dynamic trajectory control data at the same depth;

[0073] Step 102, determining the current drilling state according to the dynamic trajectory control data;

[0074] Step 103, calculating the trajectory state parameter change rate of the current drilling state according to the dynamic trajectory control data;

[0075] Step 104, taking the characteristic value in the dynamic trajectory control data as input and the trajectory state parameter change rate of the current drilling state as output, training is performed using a machine learning method to obtain a prediction model corresponding to the current drilling state;

[0076] Step 105, using a prediction model to predict the trajectory state parameter change rate of the new well section drilled by the current drill bit according to the dynamic trajectory control data sequence and the trajectory state parameter change rate sequence of the current section, and calculating the trajectory state parameter at the current drill bit location according to the predicted trajectory state parameter change rate prediction value;

[0077] Step 106, taking the trajectory state parameters at the current drill bit position as a reference, the target trajectory state parameters as an objective function, and the maximum value of the trajectory state parameter change rate corresponding to the unit footage of the well section as a constraint condition, using the prediction model corresponding to the current drilling state to perform dynamic solution, obtain at least one set of recommended trajectory control parameters and their corresponding trajectory state parameter change rates, and calculate the drilling distance to reach the target trajectory state parameters;

[0078] Step 107, performing drilling trajectory control according to the trajectory control parameters and the drilling distance.

[0079] In the embodiment of the present invention, compared with the solutions of rotary guide directional drilling and bent screw directional drilling in the prior art, the embodiment of the present invention uses the characteristic values ​​in the dynamic trajectory control data corresponding to the current drilling state as input, and uses the trajectory state parameter change rate of the current drilling state as output, and trains the corresponding prediction model, which can be used to predict the trajectory state parameters of the current drill bit position and recommend trajectory control parameters, so as to calculate the accurate drilling distance, thereby effectively controlling the drilling trajectory, and improving the drilling speed and reducing drill bit damage, thereby reducing costs and increasing efficiency.

[0080] The above steps 101 to 107 are an iterative process, that is, the above process is executed each time the dynamic trajectory control data at the same depth is collected. Therefore, step 107 is fed back to step 101.

[0081] Each step is described in detail below.

[0082] In step 101, dynamic trajectory control data at the same depth is collected.

[0083] In one embodiment, the dynamic trajectory control data includes drilling data collected from the wellhead and data collected from a wireless inclinometer while drilling (MWD);

[0084] When the current drilling state is the composite drilling state, the drilling data includes drilling pressure, rotation speed, well depth and displacement, the data collected by the wireless drilling inclinometer includes well inclination angle and azimuth angle, the trajectory state parameter change rate includes the deflection capacity and the azimuth angle change rate, the trajectory state parameters include well inclination angle and azimuth angle, and the trajectory control parameters include drilling pressure, rotation speed and displacement;

[0085] When the current drilling state is the directional sliding drilling state, the drilling data includes drilling pressure, well depth, and displacement; the data collected by the wireless downhole inclinometer includes well inclination angle, tool face, and azimuth angle; the trajectory state parameter change rate includes the deflection capacity, azimuth angle change rate, and tool face swing amplitude; the trajectory state parameters include well inclination angle, azimuth angle, and tool face swing amplitude; and the trajectory control parameters include drilling pressure and displacement;

[0086] The method further comprises:

[0087] The drilling data collected from the wellhead and the data collected from the wireless while-drilling inclinometer are aligned at the same depth as a set of data, and the same depth is used as a depth offset value.

[0088] Specifically, the data collected from the wellhead can be collected using a drilling parameter instrument or a comprehensive logging instrument.

[0089] It should be emphasized that drilling data includes drilling pressure, rotation speed, well depth and displacement, as well as other drilling data, which can be determined according to needs. In addition to the well inclination and azimuth, the data collected by the wireless downhole inclinometer can also include other data, which is also determined according to needs. Since the measurement point position to the current MWD is a certain distance L from the drill bit, the collected data does not represent the current drill bit situation, so the data collected from the wellhead and the data collected from the MWD need to be aligned at the same depth and used as a set of data. The physical meaning of this data is the well inclination, azimuth and tool face data measured by the MWD at the current measurement point, as well as the drilling pressure, rotation speed and displacement (optional) when drilling to this position in history. In all subsequent processing logic, this depth offset value and the corresponding time offset must be taken into account.

[0090] Data can be collected directly from instruments or related systems, or entered manually.

[0091] In step 102, the current drilling state is determined according to the dynamic trajectory control data;

[0092] In one embodiment, determining the current drilling state according to the dynamic trajectory control data includes:

[0093] If there is drilling pressure and the well depth changes, it is determined that the current state is drilling;

[0094] If the rotation speed is collected, it is determined that the drilling state is a compound drilling state; if the rotation speed is not collected, it is determined that the drilling state is a directional sliding drilling state.

[0095] If there is no drilling pressure and the well depth does not change, it is determined that the drilling state is not in progress.

[0096] Specifically, when judging whether the current state is drilling, the sitting card and the hanging weight can also be considered.

[0097] The above-mentioned drilling identification can be performed by using a state machine or other rule-based judgment methods, or by training an artificial intelligence model for intelligent identification. Determine whether the drilling state is a composite drilling state or a directional sliding drilling state, and enter the corresponding composite drilling condition processing sub-process and directional sliding drilling condition processing sub-process respectively. These two sub-processes correspond.

[0098] In step 103, the trajectory state parameter change rate of the current drilling state is calculated according to the dynamic trajectory control data;

[0099] In one embodiment, the trajectory state parameter change rate of the current drilling state is calculated according to the dynamic trajectory control data, including:

[0100] Determine the difference between the current dynamic trajectory control data and the last dynamic trajectory control data corresponding to the current drilling state;

[0101] When the difference in the situation exceeds the expected change, the current dynamic trajectory control data is regarded as the first piece of the next section of dynamic trajectory control data corresponding to the current drilling state;

[0102] When the situation difference does not exceed a preset range, the rate of change of the trajectory state parameter of the current drilling state is calculated.

[0103] Specifically, the embodiments of the present invention respectively introduce the data preprocessing corresponding to the composite drilling condition processing sub-process and the directional sliding drilling condition processing sub-process.

[0104] In one embodiment, determining the difference between the current dynamic trajectory control data and the last dynamic trajectory control data corresponding to the current drilling state includes:

[0105] If the difference between the depth of the current dynamic trajectory control data and the depth of the last dynamic trajectory control data corresponding to the current drilling state is greater than the first preset value D max , or the difference between the acquisition time of the current dynamic trajectory control data and the acquisition time of the last dynamic trajectory control data corresponding to the current drilling state is greater than the second preset value T max , or there is data of different drilling states between the current dynamic trajectory control data and the last dynamic trajectory control data corresponding to the current drilling state, and it is determined that the difference between the current dynamic trajectory control data and the last dynamic trajectory control data corresponding to the current drilling state exceeds the expected change.

[0106] (1) Composite drilling condition processing sub-process:

[0107] If the difference between the depth of the current dynamic trajectory control data and the depth of the last composite drilling data corresponding to the current drilling state is greater than the first preset value D max , or the difference between the acquisition time of the current dynamic trajectory control data and the acquisition time of the last composite drilling data corresponding to the current drilling state is greater than the second preset value T max, or there is data of different drilling states between the current dynamic trajectory control data and the last composite drilling data corresponding to the current drilling state, and it is determined that the difference between the current dynamic trajectory control data and the last composite drilling data corresponding to the current drilling state exceeds the expected change.

[0108] Under the composite drilling condition, the dynamic trajectory control data is composite drilling data. It is determined that the difference between the current dynamic trajectory control data and the last dynamic trajectory control data corresponding to the current drilling state exceeds the expected change, which means that the implicit characteristics have changed greatly. The current point is regarded as the first data of the next sample data. At this time, wait for the next data.

[0109] (2) Directional sliding drilling condition processing sub-process:

[0110] If the difference between the depth of the current dynamic trajectory control data and the depth of the last directional sliding drilling data corresponding to the current drilling state is greater than the first preset value D max , or the difference between the acquisition time of the current dynamic trajectory control data and the acquisition time of the last directional sliding drilling data corresponding to the current drilling state is greater than the second preset value T max , or there is data of different drilling states between the current dynamic trajectory control data and the last directional sliding drilling data corresponding to the current drilling state, and it is determined that the difference between the current dynamic trajectory control data and the last directional sliding drilling data corresponding to the current drilling state exceeds the expected change.

[0111] The overall steps of this sub-process are consistent with the above-mentioned composite drilling condition processing sub-process. The main difference is that the wellhead speed is not collected because there is no wellhead speed at this time, but the tool face and azimuth are collected; in addition, it is necessary to increase the prediction of the tool face swing amplitude, incorporate the tool face swing amplitude into the objective function to control the tool face swing amplitude, and correct the tool face cumulative error accordingly.

[0112] In one embodiment, calculating the change rate of the trajectory state parameter of the current drilling state includes:

[0113] The deflection capacity is calculated using the following formula:

[0114] ΔS i =(S i -S i-1 ) / (D i -D i-1 )

[0115] Among them, ΔS i is the deflection capability of dynamic trajectory control data, S i is the well inclination angle in the current dynamic trajectory control data, S i-1 is the well inclination angle in the previous dynamic trajectory control data, Di is the well depth in the current dynamic trajectory control data, D i-1 The well depth in the control data for the last dynamic trajectory;

[0116] The azimuth change rate is calculated using the following formula:

[0117] ΔA i =(A i -A i-1 ) / (D i -D i-1 )

[0118] ΔA i is the azimuth change rate of the current dynamic trajectory control data, A i is the azimuth in the current dynamic trajectory control data, A i-1 Control the azimuth in the data for the previous dynamic trajectory;

[0119] The tool face swing amplitude is calculated using the following formula:

[0120]

[0121] ΔW i is the tool face swing amplitude of the current dynamic trajectory control data, W i is the tool face in the current dynamic trajectory control data, W i-1 Control the toolface in the data for the last dynamic trajectory.

[0122] In the embodiment of the present invention, the characteristic value in the dynamic trajectory control data corresponding to the current drilling state is required to be input, and the trajectory state parameter change rate of the current drilling state is required to be output to train the prediction model corresponding to different drilling states.

[0123] In one embodiment, the characteristic value in the dynamic trajectory control data is input, and the trajectory state parameter change rate of the current drilling state is output, and training is performed using a machine learning method to obtain a prediction model corresponding to the current drilling state, including:

[0124] Each time a new set of data is obtained, or according to the preset interval, the prediction model is trained according to the following steps:

[0125] When the current drilling state is a composite drilling state, starting from the second dynamic trajectory control data in the current segment of dynamic trajectory control data, according to the start and end time of the previous dynamic trajectory control data and the current dynamic trajectory control data during drilling, the average values ​​of drilling pressure, rotation speed and displacement are taken as characteristic values; the deflection capacity and azimuth change rate are taken as the trajectory state parameter change rates, and training is performed using a machine learning method to obtain a prediction model corresponding to the composite drilling state;

[0126] When the current drilling state is a directional sliding drilling state, starting from the second dynamic trajectory control data in the current section of the dynamic trajectory control data, according to the start and end time of the previous dynamic trajectory control data and the current dynamic trajectory control data during drilling, the average values ​​of the drilling pressure, the average values ​​of the displacement and the average values ​​of the tool face are taken as characteristic values; the deflection ability, the azimuth change rate and the tool face swing amplitude are taken as the trajectory state parameter change rates, and training is performed using a machine learning method to obtain a prediction model corresponding to the directional sliding drilling state.

[0127] The molecular process is described in detail below.

[0128] (1) Composite drilling condition processing sub-process:

[0129] Starting from the second dynamic trajectory control data in the current segment of dynamic trajectory control data, that is, for each adjacent dynamic trajectory control data in each segment (excluding the first one), according to the start and end time of the previous dynamic trajectory control data and the current dynamic trajectory control data during drilling, the average value of the drilling pressure WOB is taken. i , the average value of the speed RPM i , the average value of displacement MFI i as a feature value.

[0130] Using machine learning, a prediction model M corresponding to the composite drilling state with multiple inputs and outputs is trained based on the above sequence. m .

[0131] The input can be adjusted according to actual needs, such as adding the slope depth of the measuring point. The basic input combination is as follows:

[0132] {WOB i , RPM i , MFI i}=》{ΔS i , ΔA i}

[0133] (2) Directional sliding drilling condition processing sub-process:

[0134] Starting from the second dynamic trajectory control data in the current segment of dynamic trajectory control data, that is, for each adjacent dynamic trajectory control data in each segment (excluding the first one), according to the start and end time of the previous dynamic trajectory control data and the current dynamic trajectory control data during drilling, the average value of the drilling pressure WOB is taken. i , the average value of displacement MFI i (optional) and the average value W of the tool surface i as a feature value.

[0135] Using machine learning, a multi-input and multi-output prediction model M corresponding to the directional sliding drilling state is trained based on the above sequence. m The input can be adjusted according to actual needs, such as adding the slope depth of the measuring point. The basic input combination is as follows:

[0136] {WOB i , MFI i ,W i}=》{ΔS i , ΔA i ,ΔW i}

[0137] In the above embodiments, the machine learning algorithm does not need to be limited to a specific algorithm and can be replaced according to the development of IT technology. Considering the time series characteristics of the well inclination data and the fact that the neighboring data is closer to the characteristics of the current data than the earlier data, data mining or artificial intelligence models such as long short-term memory network (LSTM), ordinary differential equations (ODE), Transformer, etc. can be used, or machine learning algorithms combined with the above models, such as the convolutional neural network "bidirectional long short-term memory network" (CNN-BiLSTM-Attention) with an attention mechanism, so that the closer the data is, the more weight it has.

[0138] Each time a new set of data is obtained, the model needs to be retrained. Considering the performance of the on-site computer, an interval training method can also be used.

[0139] In one embodiment, the method further includes:

[0140] Based on the construction data of historical wells, pre-trained models corresponding to different drilling states are trained;

[0141] According to the pre-trained models corresponding to different drilling states, incremental training is performed on the prediction models corresponding to different drilling states.

[0142] The embodiment of the present invention needs to predict the rate of change of trajectory state parameters, and predict the well inclination angle and azimuth angle. The embodiment of the present invention needs to predict the two sub-processes separately, and for each sub-process, there are two states. In this busy embodiment, there are three ways to predict. One is to predict the state of the current drill bit position; the second is to list different trajectory control parameters and predict the rate of change of trajectory state parameters at the current drill bit position to provide a reference for adjusting the construction parameters in the next step; the third is to give the deflection requirements of the section to be drilled, such as the expected change value of the well inclination angle within a certain distance, and the constraint value of the rate of change of trajectory state parameters, and automatically recommend the rate of change of trajectory state parameters at different depths in the next step.

[0143] Each state is divided into simple working conditions and complex working conditions.

[0144] A) Simple working condition

[0145] (1) Composite drilling condition processing sub-process:

[0146] When the current drilling state is the composite drilling state, if the change range of drilling pressure, rotation speed and displacement of the wellbore between the depth of the current measurement point and the depth of the drill bit position is less than the preset change range ΔWOB max , ΔRPM max , ΔMFI max , take the average value of drilling pressure, speed and displacement; the average value of drilling pressure WOB n , average speed RPM n and the mean value of the displacement MFI n Input into the prediction model corresponding to the composite drilling state, predict the deflection capacity and azimuth change rate of the new well section drilled by the current drill bit, and obtain the deflection capacity prediction value and azimuth change rate prediction value;

[0147] (2) Directional sliding drilling condition processing sub-process:

[0148] When the current drilling state is directional sliding drilling state, if the change range of drilling pressure, displacement and tool face between the depth of the current measuring point and the depth of the drill bit is less than the preset change range ΔWOB max , ΔMFI max , ΔW max , take the average value of drilling pressure, displacement and tool face; the average value of drilling pressure WOB n , the average value of the displacement MFI n and the average value W of the tool face n The prediction model corresponding to the directional sliding drilling state is input to predict the deflection capacity, azimuth change rate and tool face swing amplitude of the new drilling section of the current drill bit, and the predicted values ​​of the deflection capacity, azimuth change rate and tool face swing amplitude are obtained.

[0149] In one embodiment, the trajectory state parameter at the current drill bit position is calculated according to the predicted value of the trajectory state parameter change rate, including:

[0150] Use the following formula to calculate the well inclination angle at the current drill bit location:

[0151] S=S i +(L×ΔS n )

[0152] Among them, S is the well inclination angle of the current drill bit position, S i is the well inclination angle at the current measuring point, ΔS n is the predicted value of the deflection capacity of the newly drilled well section, and L is the distance between the current drill bit and the current measuring point;

[0153] Use the following formula to calculate the azimuth of the current drill position:

[0154] A=A i +(L×ΔA n )

[0155] Among them, A is the azimuth of the current drill position, A i is the azimuth of the current measuring point, L is the distance between the drill head and the measuring point, ΔA n It is the predicted value of the azimuth change rate of the newly drilled well section;

[0156] The following formula is used to calculate the tool face swing amplitude at the current drill bit position:

[0157] ΔW=ΔW n

[0158] Among them, ΔW is the azimuth of the current drill bit position, ΔW n It is the predicted value of tool face swing amplitude.

[0159] B) Complex working conditions

[0160] (1) Composite drilling condition processing sub-process:

[0161] When the current drilling state is the composite drilling state, if the change range of any one of the drilling pressure, rotation speed and displacement of the wellbore drilling between the depth of the current measurement point and the depth of the drill bit is not less than the preset change range ΔWOB max , ΔRPM max , ΔMFI max , and the segment length of the dynamic trajectory control data is greater than the minimum segment SEG min The dynamic trajectory control data is segmented to form multiple data segments [SEG1, SEG2, …, SEG n], take the average value for each data segment; for each data segment, input the average value of the drilling pressure, the average value of the rotation speed and the average value of the displacement of the data segment into the corresponding prediction model, predict the deflection capacity and azimuth change rate of each data segment, and obtain the deflection capacity prediction value and azimuth change rate prediction value of each data segment [ΔS1, ΔA1], [ΔS2, ΔA2], … [ΔSn, ΔAn];

[0162] (2) Directional sliding drilling condition processing sub-process:

[0163] When the current drilling state is directional sliding drilling state, if the change range of any item of drilling pressure, displacement and tool face in the wellbore drilling between the depth of the current measuring point and the depth of the drill bit position is not less than the preset change range ΔWOB max , ΔMFI max , ΔW max , and the segment length of the dynamic trajectory control data is greater than the minimum segment SEG min The dynamic trajectory control data is segmented to form multiple data segments [SEG1, SEG2, …, SEG n ], take the average value for each data segment; for each data segment, input the average drilling pressure, displacement and tool face of the data segment into the corresponding prediction model, predict the deflection capacity, azimuth change rate and tool face swing amplitude of each data segment, and obtain the deflection capacity prediction value, azimuth change rate prediction value and tool face swing amplitude prediction value of each data segment [ΔS1, ΔA1, ΔW1], [ΔS2, ΔA2, ΔW2], … [ΔS n , ΔA n , ΔW n ].

[0164] In one embodiment, the trajectory state parameter at the current drill bit position is calculated according to the predicted value of the trajectory state parameter change rate, including:

[0165] Use the following formula to calculate the well inclination angle at the current drill bit location:

[0166]

[0167] Among them, S is the well inclination angle of the current drill bit position, S k is the well inclination angle corresponding to the kth data segment of the current measurement point, L k is the segment length corresponding to the kth data segment, ΔS k is the predicted value of the deflection capacity of the newly drilled well section corresponding to the kth data segment, and n is the number of data segments;

[0168] Use the following formula to calculate the azimuth of the current drill position:

[0169]

[0170] Among them, A is the azimuth of the current drill position, A k is the azimuth corresponding to the kth data segment of the current measurement point, ΔA k is the predicted value of the azimuth change rate of the new drilling section corresponding to the kth data segment;

[0171] The following formula is used to calculate the tool face swing amplitude at the current drill bit position:

[0172] ΔW=ΔW n

[0173] Among them, ΔW is the azimuth of the current drill bit position, ΔW n It is the predicted value of the tool face swing amplitude corresponding to the last data segment.

[0174] In step 106, the trajectory state parameters at the current drill bit position are used as a reference, the target trajectory state parameters are used as an objective function, and the maximum value of the trajectory state parameter change rate corresponding to the unit footage of the well section is used as a constraint condition. The prediction model corresponding to the current drilling state is used to perform dynamic solution to obtain at least one set of recommended trajectory control parameters and their corresponding trajectory state parameter change rates, and the drilling distance to reach the target trajectory state parameters is calculated;

[0175] (1) Composite drilling condition processing sub-process:

[0176] Based on the well inclination S and azimuth A at the current drill bit location as the benchmark, the target well inclination and azimuth as the objective function, and the maximum well inclination change rate and azimuth change rate per unit footage as the constraints, the trained prediction model is used for dynamic solution to obtain one or more recommended drilling pressure, rotation speed, displacement combinations, and their corresponding deflection capacity ΔS and azimuth change rate ΔA, and the drilling distance to reach the target well inclination and target azimuth is calculated.

[0177] (2) Directional sliding drilling condition processing sub-process:

[0178] Based on the well inclination S and azimuth A at the current drill bit location as the benchmark, the target well inclination and azimuth as the objective function, and the maximum well inclination change rate and azimuth change rate per unit footage and the maximum swing amplitude of the tool face as the constraints, the trained prediction model is used for dynamic solution to obtain one or more recommended drilling pressure and displacement combinations, as well as their corresponding deflection capacity ΔS and azimuth change rate ΔA, and tool face swing amplitude W, and the drilling distance to reach the target well inclination and azimuth is calculated.

[0179] The solution process in the above process can use conventional parameter iteration, particle swarm iteration (iterative particle swarm algorithm) and other methods to traverse the potential parameter space.

[0180] The recommended drilling pressure and displacement parameters can serve as real-time reminders to operators and can be further connected to the drilling rig control system to achieve real-time automatic closed-loop control.

[0181] In addition to the above prediction scheme, the embodiment of the present invention also proposes other schemes.

[0182] (1) Using the classic formula for prediction

[0183] Modeling the drilling tool mechanics to obtain the drilling tool mechanics model;

[0184] At least one set of recommended trajectory control parameters and their corresponding trajectory state parameter change rates are calculated according to the drilling tool mechanics model.

[0185] Among them, the drilling tool mechanics are modeled and solved, including the calculation and analysis of the guiding force and guiding capacity of the bent screw BHA, the study of the influence of key parameters on the guiding force of the BHA drill bit, the inversion calculation of the balanced lateral force of the formation, and the design of BHA structural parameters for different configurations of sliding guidance and composite drilling capabilities.

[0186] This method is highly interpretable, but it requires the collection of a large amount of drilling tools and geological information, requires a large amount of calculation, is not closely integrated with measured data, and the calculated results may differ significantly from the actual situation.

[0187] The main difference is that this method does not use artificial intelligence methods.

[0188] (2) Grouping and calculating the measured data

[0189] The dynamic trajectory control data are grouped according to the depth intervals;

[0190] Analyze the trajectory control rules of different groups;

[0191] According to the analyzed trajectory control law, at least one set of recommended trajectory control parameters and their corresponding trajectory state parameter change rates are predicted.

[0192] This method requires the collected data to maintain a fixed spacing and to group the data in a fixed range. It is a relative quantitative method. When the data spacing is irregular or discontinuous, additional special processing is required. In addition, it is not necessarily reasonable for humans to preset the grouping range of rotation speed and drilling pressure, which may coarsen the potential parameter relationship and turn the continuous space into a discrete space.

[0193] The main difference is that this patent does not need to group parameters and does not require equal spacing, and the computer intelligently finds the rules.

[0194] (3) Iterative solution based on artificial intelligence model

[0195] After depth migration of the dynamic trajectory control data collected at the same depth, artificial intelligence models corresponding to different drilling states are trained respectively;

[0196] Based on the measurement point and all previous dynamic trajectory control data as a sample sequence, the artificial intelligence model is used to iteratively solve the problem in the following steps until the trajectory state parameters of the current position of the drill bit are solved:

[0197] Taking the average value of the measured dynamic trajectory control data at the next predicted depth as input, the trajectory state parameters at the next depth are predicted;

[0198] The predicted values ​​of trajectory state parameters and the measured trajectory state parameters are put into a new set of data at the end of the sample sequence.

[0199] The above prediction process can be expressed as:

[0200] Based on the actual measured points and all previous data as a sequence, the measured drilling pressure and rotation speed (average value, processing logic is the same as above) of the next predicted depth are used as input to predict the well inclination, azimuth and tool face of the next depth;

[0201] Then, the predicted well inclination angle, azimuth angle, tool face and measured drilling pressure and rotation speed are put into a new set of data at the end of the sample sequence;

[0202] Then, the measured drilling pressure and rotation speed at the next predicted depth are used as input to predict the next set of well inclination angles, azimuth angles, and tool faces. This cycle is repeated until the well inclination angles, azimuth angles, and tool faces at the current position of the drill bit are solved.

[0203] Generally speaking, the difference between the dynamic measurement of well inclination and the static measurement (one column is 28.5m) is plus or minus 0.1-0.2 degrees. After each static measurement point, it is necessary to analyze the error interval and reverse the error distribution.

[0204] (1) Composite drilling condition processing sub-process:

[0205] When the current drilling state is the composite drilling state, after each static measurement point, the difference between the dynamic well inclination angle and the static well inclination angle of the last static measurement point is calculated and used as the error interval of the well inclination angle, and the difference between the dynamic azimuth angle and the static azimuth angle of the last static measurement point is calculated and used as the error interval of the azimuth angle; according to the depths of the last two static measurement points, for the covered dynamic measurement points, the error interval of the well inclination angle and the azimuth angle are distributed to each dynamic measurement point according to the proportion of the depth between each adjacent point, and the value of the last dynamic measurement point is corrected to the value of the corresponding static measurement point; after the value of the dynamic measurement point is corrected, the corresponding prediction model is retrained;

[0206] (2) Directional sliding drilling condition processing sub-process:

[0207] When the current drilling state is the directional sliding drilling state, after each static measurement point, the difference between the dynamic well inclination and the static well inclination of the last static measurement point is calculated and used as the error interval of the well inclination, the difference between the dynamic azimuth and the static azimuth of the last static measurement point is calculated and used as the error interval of the azimuth, and the difference between the dynamic tool face and the static tool face of the last static measurement point is calculated and used as the tool face interval of the azimuth; according to the depths of the last two static measurement points, for the covered dynamic measurement points, the error interval of the well inclination, the error interval of the azimuth and the error interval of the tool face are allocated to each dynamic measurement point according to the proportion of the depth between each adjacent point, and the value of the last dynamic measurement point is corrected to the value of the corresponding static measurement point; after the value of the dynamic measurement point is corrected, the corresponding prediction model is retrained.

[0208] A specific embodiment is given below to illustrate the specific application of the method and device proposed in the embodiment of the present invention.

[0209] Figure 2 The present invention implements two effect diagrams of using screw drilling tools to achieve full-section drilling. When screw drilling tools are used instead of rotary guides, the well inclination angle, azimuth and other parameters of the current drill bit position can be predicted in real time to achieve precise control of the wellbore trajectory, increase the drilling speed, and reduce drill bit damage, thereby reducing costs and increasing efficiency.

[0210] Figure 3 This is a drilling chart of the horizontal section of a well in the third opening in the embodiment of the invention. Figure 4 This is a schematic diagram of the application of the method proposed in an embodiment of the present invention in a certain well. Figure 3 and Figure 4It is one of the cases that uses the dynamic trajectory control data of each meter, performs parameter grouping tests and predicts the control of the drilling section to be drilled, and uses the PDC drill bit + screw double support + MWD drilling tool combination to achieve continuous trajectory control of composite drilling. It has created a well with a horizontal section depth of 3962m-5291m (using DBS's GT55DRs PDC drill bit + 1.25° bent screw (φ200 support) + φ205 tail support + resistivity + non-magnetic suspension, etc.), a drilling footage of 1329m in one trip, and an average mechanical drilling speed of 12.7m / h. Among them, the continuous composite footage from the well depth of 4017m-5291m is 1274m, setting a record for the longest footage of horizontal section bent screw composite drilling in water-based mud in the block.

[0211] Figure 4 The green curve in the middle represents the top drive speed that changes with well depth, the blue dots represent the drilling pressure applied as the well depth changes, and the red curve represents the drilling time that changes with well depth. Figure 4 The two dot-line combined curves on the upper side represent the well inclination angle and azimuth angle respectively.

[0212] from Figure 4 It can be seen that by continuously recording the dynamic well inclination angle, corresponding drilling pressure, rotation speed and other data measured with the well depth during the composite drilling process, the data was initially grouped (drilling pressure 50-100kN, rotation speed 40-80rpm) for testing to master the control rules of the drilling tool assembly under different data;

[0213] The error analysis of the recorded dynamic well inclination angle was carried out through the static measurement point well inclination angle data. The error of the dynamic well inclination angle of the downhole combination is ±0.15.

[0214] According to the analysis and judgment of the dynamic well inclination angle change law of the above different dynamic trajectory control data, the trajectory of the drilling section is controlled and predicted; the predicted trajectory state parameters are compared and analyzed with the measured trajectory state parameters in the subsequent well section, and then the drilling trajectory control parameters (biting pressure, rotation speed) are continuously optimized and corrected, so that the control and prediction accuracy are further improved;

[0215] This cycle was repeated, and dynamic trajectory control data was used for continuous analysis, adjustment and continuous prediction, achieving the good result of continuous composite drilling of 1,274 m in the well.

[0216] Figure 5 This is a chart showing the dynamic well inclination angle, drilling pressure and rotation speed changes in a well from 4760m to 4900m in an embodiment of the present invention. Figure 5 It utilizes the dynamic trajectory control data of each meter, uses a PDC drill bit + screw double support + MWD drill tool combination, and uses drilling pressure and rotation speed to control and predict the drilling section to achieve composite drilling continuous trajectory control (4760m-4900m local section of a well). Among them, the light green area is the inclination increasing section, and the light orange area is the inclination decreasing section.

[0217] Figure 5 The green curve in the middle represents the top drive speed that changes with well depth, the blue dots represent the drilling pressure applied as the well depth changes, and the red curve represents the drilling time that changes with well depth. Figure 5 The upper dot-line combined curve represents the well inclination angle.

[0218] From the change diagram of well inclination per meter, we can see that:

[0219] In the 6m section from 4764m to 4770m (mid-to-late stage of using the drill bit and the centralizer), the drilling pressure was 120kN, the rotation speed was 60rpm, and the well inclination angle changed from a decreasing trend to an increasing trend and the increase in inclination was achieved at an increasing well inclination angle of 0.6° / 10m;

[0220] In the 20m section from 4812m to 4832m deep (drill bit and centralizer were worn out in the later period of use), drilling pressure was 60kN-75kN, rotation speed was 80rpm, and the well inclination angle changed from increasing to decreasing and decreased at a decreasing well inclination angle of 0.6° / 10m;

[0221] In the 14m section from 4886m to 4900m (the drill bit and the centralizer were seriously worn in the later period of use), the drilling pressure was 110-120kN, the rotation speed was 30rpm, and the well inclination angle changed from a decreasing trend to an increasing trend and still achieved an increasing well inclination angle of 0.6° / 10m;

[0222] In this way, we continuously analyze the dynamic well inclination, drilling pressure and rotation speed data, make timely adjustments to the drilling pressure and rotation speed, and continuously predict the trajectory, so as to achieve continuous control of the horizontal section trajectory within the designed target box (target oil layer).

[0223] The embodiment of the present invention further provides a petroleum drilling trajectory dynamic control device, the principle of which is similar to the petroleum drilling trajectory dynamic control method, which will not be described in detail here.

[0224] Figure 6 This is a schematic diagram of a petroleum drilling trajectory dynamic control device in an embodiment of the present invention, comprising:

[0225] The data acquisition module 601 is used to acquire dynamic trajectory control data at the same depth;

[0226] A drilling state determination module 602 is used to determine the current drilling state according to the dynamic trajectory control data;

[0227] The data preprocessing module 603 is used to calculate the trajectory state parameter change rate of the current drilling state according to the dynamic trajectory control data;

[0228] The model training module 604 is used to take the characteristic values ​​in the dynamic trajectory control data as input and the trajectory state parameter change rate of the current drilling state as output, and to perform training using a machine learning method to obtain a prediction model corresponding to the current drilling state;

[0229] The drill bit inclination angle prediction module 605 is used to predict the trajectory state parameter change rate of the new well section drilled by the current drill bit using the prediction model according to the dynamic trajectory control data sequence and the trajectory state parameter change rate sequence of the current section, and calculate the trajectory state parameter at the current drill bit position according to the predicted trajectory state parameter change rate prediction value;

[0230] The trajectory control parameter recommendation module 606 is used to take the trajectory state parameters of the current drill bit position as a reference, the target trajectory state parameters as an objective function, and the maximum value of the trajectory state parameter change rate corresponding to the unit footage of the well section as a constraint condition, and use the prediction model corresponding to the current drilling state to perform dynamic solution to obtain at least one set of recommended trajectory control parameters and their corresponding trajectory state parameter change rates, and calculate the drilling distance to reach the target trajectory state parameters;

[0231] The trajectory control module 607 is used to perform drilling trajectory control according to the trajectory control parameters and the drilling distance.

[0232] In one embodiment, the dynamic trajectory control data includes drilling data collected from the wellhead and data collected from a wireless while-drilling inclinometer;

[0233] When the current drilling state is the composite drilling state, the drilling data includes drilling pressure, rotation speed, well depth and displacement, the data collected by the wireless drilling inclinometer includes well inclination angle and azimuth angle, the trajectory state parameter change rate includes the deflection capacity and the azimuth angle change rate, the trajectory state parameters include well inclination angle and azimuth angle, and the trajectory control parameters include drilling pressure, rotation speed and displacement;

[0234] When the current drilling state is the directional sliding drilling state, the drilling data includes drilling pressure, well depth, and displacement; the data collected by the wireless downhole inclinometer includes well inclination angle, tool face, and azimuth angle; the trajectory state parameter change rate includes the deflection capacity, azimuth angle change rate, and tool face swing amplitude; the trajectory state parameters include well inclination angle, azimuth angle, and tool face swing amplitude; and the trajectory control parameters include drilling pressure and displacement;

[0235] The data acquisition module is used to:

[0236] The drilling data collected from the wellhead and the data collected from the wireless while-drilling inclinometer are aligned at the same depth as a set of data, and the same depth is used as a depth offset value.

[0237] In one embodiment, the drilling status determination module is used to:

[0238] If the well depth changes, determine the current drilling state;

[0239] If the rotation speed is collected, it is determined that the drilling state is a compound drilling state; if the rotation speed is not collected, it is determined that the drilling state is a directional sliding drilling state.

[0240] If the well depth does not change, it is determined that the well is not currently in the drilling state.

[0241] In one embodiment, the data preprocessing module is used to:

[0242] Determine the difference between the current dynamic trajectory control data and the last dynamic trajectory control data corresponding to the current drilling state;

[0243] When the difference in the situation exceeds the expected change, the current dynamic trajectory control data is regarded as the first piece of the next section of dynamic trajectory control data corresponding to the current drilling state;

[0244] When the situation difference does not exceed a preset range, the rate of change of the trajectory state parameter of the current drilling state is calculated.

[0245] In one embodiment, the data preprocessing module is used to:

[0246] If the difference between the depth of the current dynamic trajectory control data and the depth of the last dynamic trajectory control data corresponding to the current drilling state is greater than a first preset value, or the difference between the acquisition time of the current dynamic trajectory control data and the acquisition time of the last dynamic trajectory control data corresponding to the current drilling state is greater than a second preset value, or there is data of different drilling states between the current dynamic trajectory control data and the last dynamic trajectory control data corresponding to the current drilling state, it is determined that the difference between the situations of the current dynamic trajectory control data and the last dynamic trajectory control data corresponding to the current drilling state exceeds an expected change.

[0247] In one embodiment, the data preprocessing module is used to:

[0248] The deflection capacity is calculated using the following formula:

[0249] ΔS i =(S i -S i-1 ) / (D i -D i-1 )

[0250] Among them, ΔS i is the deflection capability of dynamic trajectory control data, S i is the well inclination angle in the current dynamic trajectory control data, S i-1 is the well inclination angle in the previous dynamic trajectory control data, D iis the well depth in the current dynamic trajectory control data, D i-1 The well depth in the control data for the last dynamic trajectory;

[0251] The azimuth change rate is calculated using the following formula:

[0252] ΔA i =(A i -A i-1 ) / (D i -D i-1 )

[0253] ΔA i is the azimuth change rate of the current dynamic trajectory control data, A i is the azimuth in the current dynamic trajectory control data, A i-1 Control the azimuth in the data for the previous dynamic trajectory;

[0254] The tool face swing amplitude is calculated using the following formula:

[0255]

[0256] ΔW i is the tool face swing amplitude of the current dynamic trajectory control data, W i is the tool face in the current dynamic trajectory control data, W i-1 Control the toolface in the data for the last dynamic trajectory.

[0257] In one embodiment, the model training module is used to:

[0258] Each time a new set of data is obtained, or according to the preset interval, the prediction model is trained according to the following steps:

[0259] When the current drilling state is a composite drilling state, starting from the second dynamic trajectory control data in the current segment of dynamic trajectory control data, according to the start and end time of the previous dynamic trajectory control data and the current dynamic trajectory control data during drilling, the average values ​​of drilling pressure, rotation speed and displacement are taken as characteristic values; the deflection capacity and azimuth change rate are taken as the trajectory state parameter change rates, and training is performed using a machine learning method to obtain a prediction model corresponding to the composite drilling state;

[0260] When the current drilling state is a directional sliding drilling state, starting from the second dynamic trajectory control data in the current section of the dynamic trajectory control data, according to the start and end time of the previous dynamic trajectory control data and the current dynamic trajectory control data during drilling, the average values ​​of the drilling pressure, the average values ​​of the displacement and the average values ​​of the tool face are taken as characteristic values; the deflection ability, the azimuth change rate and the tool face swing amplitude are taken as the trajectory state parameter change rates, and training is performed using a machine learning method to obtain a prediction model corresponding to the directional sliding drilling state.

[0261] In one embodiment, the drill bit inclination angle prediction model is specifically used for:

[0262] Based on the construction data of historical wells, pre-trained models corresponding to different drilling states are trained;

[0263] According to the pre-trained models corresponding to different drilling states, incremental training is performed on the prediction models corresponding to different drilling states.

[0264] In one embodiment, the drill bit inclination angle prediction model is specifically used for:

[0265] When the current drilling state is the composite drilling state, if the variation ranges of the drilling pressure, rotation speed and displacement of the wellbore drilling between the depth of the current measuring point and the depth of the drill bit position are all less than the preset variation range, the drilling pressure, rotation speed and displacement are averaged; the average values ​​of the drilling pressure, rotation speed and displacement are input into the prediction model corresponding to the composite drilling state, and the deflection capacity and azimuth change rate of the new drilling section of the current drill bit are predicted, and the deflection capacity prediction value and the azimuth change rate prediction value are obtained;

[0266] When the current drilling state is the directional sliding drilling state, if the change range of the drilling pressure, displacement and tool face of the wellbore drilling between the depth of the current measuring point and the depth of the drill bit position is less than the preset change range, the drilling pressure, displacement and tool face are averaged; the average values ​​of the drilling pressure, displacement and tool face are input into the prediction model corresponding to the directional sliding drilling state to predict the deflection capacity, azimuth change rate and tool face swing amplitude of the new drilling section of the current drill bit, and obtain the deflection capacity prediction value, azimuth change rate prediction value and tool face swing amplitude prediction value.

[0267] In one embodiment, the drill bit inclination angle prediction model is specifically used for:

[0268] Use the following formula to calculate the well inclination angle at the current drill bit location:

[0269] S=S i +(L×ΔS n )

[0270] Among them, S is the well inclination angle of the current drill bit position, Si is the well inclination angle at the current measuring point, ΔS n is the predicted value of the deflection capacity of the newly drilled well section, and L is the distance between the current drill bit and the current measuring point;

[0271] Use the following formula to calculate the azimuth of the current drill position:

[0272] A=A i +(L×ΔA n )

[0273] Among them, A is the azimuth of the current drill position, A i is the azimuth of the current measuring point, L is the distance between the drill head and the measuring point, ΔA n It is the predicted value of the azimuth change rate of the newly drilled well section;

[0274] The following formula is used to calculate the tool face swing amplitude at the current drill bit position:

[0275] ΔW=ΔW n

[0276] Among them, ΔW is the azimuth of the current drill bit position, ΔW n It is the predicted value of tool face swing amplitude.

[0277] In one embodiment, the drill bit inclination angle prediction model is specifically used for:

[0278] According to the dynamic trajectory control data sequence and trajectory state parameter change rate sequence of the current section, the prediction model is used to predict the trajectory state parameter change rate of the new well section drilled by the current drill bit, including:

[0279] When the current drilling state is a composite drilling state, if the change amplitude of any one of the drilling pressure, rotation speed and displacement of the wellbore drilling between the depth of the current measuring point and the depth of the drill bit position is not less than the preset change amplitude, and the segment length of the dynamic trajectory control data is greater than the minimum segment, the dynamic trajectory control data is segmented to form multiple data segments, and an average value is taken for each data segment; for each data segment, the average value of the drilling pressure, the average value of the rotation speed and the average value of the displacement of the data segment are input into the corresponding prediction model, and the deflection ability and azimuth change rate of each data segment are predicted, so as to obtain the deflection ability prediction value and azimuth change rate prediction value of each data segment;

[0280] When the current drilling state is the directional sliding drilling state, if the change amplitude of any one of the drilling pressure, displacement and tool face of the wellbore drilling between the depth of the current measuring point and the depth of the drill bit position is not less than the preset change amplitude, and the segment length of the dynamic trajectory control data is greater than the minimum segment, the dynamic trajectory control data is segmented to form multiple data segments, and an average value is taken for each data segment; for each data segment, the average value of the drilling pressure, the average value of the displacement and the average value of the tool face of the data segment are input into the corresponding prediction model to predict the deflection ability, azimuth change rate and tool face swing amplitude of each data segment, and obtain the deflection ability prediction value, azimuth change rate prediction value and tool face swing amplitude prediction value of each data segment.

[0281] In one embodiment, the drill bit inclination angle prediction model is specifically used for:

[0282] Use the following formula to calculate the well inclination angle at the current drill bit location:

[0283]

[0284] Among them, S is the well inclination angle of the current drill bit position, S k is the well inclination angle corresponding to the kth data segment of the current measurement point, L k is the segment length corresponding to the kth data segment, ΔS k is the predicted value of the deflection capacity of the newly drilled well section corresponding to the kth data segment, and n is the number of data segments;

[0285] Use the following formula to calculate the azimuth of the current drill position:

[0286]

[0287] Among them, A is the azimuth of the current drill position, A k is the azimuth corresponding to the kth data segment of the current measurement point, ΔA k is the predicted value of the azimuth change rate of the new drilling section corresponding to the kth data segment;

[0288] The following formula is used to calculate the tool face swing amplitude at the current drill bit position:

[0289] ΔW=ΔW n

[0290] Among them, ΔW is the azimuth of the current drill bit position, ΔW n It is the predicted value of the tool face swing amplitude corresponding to the last data segment.

[0291] In one embodiment, the device further comprises a formula prediction module, which is used to:

[0292] Modeling the drilling tool mechanics to obtain the drilling tool mechanics model;

[0293] At least one set of recommended trajectory control parameters and their corresponding trajectory state parameter change rates are calculated according to the drilling tool mechanics model.

[0294] In one embodiment, the device further comprises a group calculation module, which is used to:

[0295] The dynamic trajectory control data are grouped according to the depth intervals;

[0296] Analyze the trajectory control rules of different groups;

[0297] According to the analyzed trajectory control law, at least one set of recommended trajectory control parameters and their corresponding trajectory state parameter change rates are predicted.

[0298] In one embodiment, the device further comprises an artificial intelligence model solving module, which is used to:

[0299] After depth migration of the dynamic trajectory control data collected at the same depth, artificial intelligence models corresponding to different drilling states are trained respectively;

[0300] Based on the measurement point and all previous dynamic trajectory control data as a sample sequence, the artificial intelligence model is used to iteratively solve the problem in the following steps until the trajectory state parameters of the current position of the drill bit are solved:

[0301] Taking the average value of the measured dynamic trajectory control data at the next predicted depth as input, predict the trajectory state parameters at the next depth;

[0302] The predicted values ​​of trajectory state parameters and the measured trajectory state parameters are put into a new set of data at the end of the sample sequence.

[0303] In one embodiment, the device further comprises a correction module, configured to:

[0304] When the current drilling state is the composite drilling state, after each static measurement point, the difference between the dynamic well inclination angle and the static well inclination angle of the last static measurement point is calculated and used as the error interval of the well inclination angle, and the difference between the dynamic azimuth angle and the static azimuth angle of the last static measurement point is calculated and used as the error interval of the azimuth angle; according to the depths of the last two static measurement points, for the covered dynamic measurement points, the error interval of the well inclination angle and the azimuth angle are distributed to each dynamic measurement point according to the proportion of the depth between each adjacent point, and the value of the last dynamic measurement point is corrected to the value of the corresponding static measurement point; after the value of the dynamic measurement point is corrected, the corresponding prediction model is retrained;

[0305] When the current drilling state is the directional sliding drilling state, after each static measurement point, the difference between the dynamic well inclination and the static well inclination of the last static measurement point is calculated and used as the error interval of the well inclination, the difference between the dynamic azimuth and the static azimuth of the last static measurement point is calculated and used as the error interval of the azimuth, and the difference between the dynamic tool face and the static tool face of the last static measurement point is calculated and used as the tool face interval of the azimuth; according to the depths of the last two static measurement points, for the covered dynamic measurement points, the error interval of the well inclination, the error interval of the azimuth and the error interval of the tool face are allocated to each dynamic measurement point according to the proportion of the depth between each adjacent point, and the value of the last dynamic measurement point is corrected to the value of the corresponding static measurement point; after the value of the dynamic measurement point is corrected, the corresponding prediction model is retrained.

[0306] In summary, the method and device proposed in the embodiments of the present invention have the following beneficial effects:

[0307] (1) Reduce or completely abandon the use of rotary guides in horizontal well sections or high-angle stable well sections to save costs

[0308] Without using rotary guide, it not only saves equipment costs, but also allows higher drilling pressure and rotation speed parameters to be used, greatly increasing the drilling speed, especially in formations with poor drillability. If used in the horizontal section, it can further shorten the horizontal well drilling cycle and greatly save costs.

[0309] (2) No additional data collection methods are required

[0310] Using a relatively simple and inexpensive device such as a bent screw, the existing data collected is used, and no additional data collection is required. No special group test is required, and the rules are automatically obtained without interfering with the construction. Using the data continuously measured and transmitted by the drilling measurement instrument, there is no need for additional special drilling stop measurement, and the production efficiency is high. This method can reduce the number and time of static well inclination angle measurements and improve drilling efficiency.

[0311] (3) Improve the composite time rate and reduce the proportion of sliding drilling

[0312] Advance judgment provides a basis for determining follow-up measures in advance, and drilling parameters can be adjusted more timely. It also takes into account the control of well inclination, azimuth and tool face, and the trajectory control is more precise. It avoids or greatly reduces the directional sliding drilling process, improves the single-trip drilling footage and speed, and basically solves the problem of difficult trajectory control of the bent screw drill bit assembly in long horizontal sections (especially water-based mud). When sliding drilling, the tool face is kept as stable as possible, and the given sliding drilling parameters are more accurate. The trip drilling footage and mechanical drilling speed under the same conditions in a typical block are expected to increase by about 10%.

[0313] (4) Reduce downhole complexity and drill bit damage, as well as the corresponding losses

[0314] By reducing directional sliding drilling, the occurrence of underground complex situations such as sticking and drilling jams can be effectively avoided, the damage to the drill bit can be reduced, the service life of the drilling bit can be extended, the drilling footage can be increased, and the cost of using the drill bit can be saved. Furthermore, the extra tripping and drilling operations caused by damage and premature aging of the drill bit can be reduced, further shortening the construction time.

[0315] (5) Keep the wellbore trajectory smooth

[0316] The wellbore formed by composite drilling is smoother, and the completion electrical logging and casing operation are smoother. Precise and efficient trajectory control is conducive to extending the service life of the drill bit, increasing the machine speed and drilling footage, and reducing drilling costs.

[0317] (6) Reduce human errors

[0318] Fully automatic learning, prediction and control, without the need for human interaction, can reduce the requirements for directional engineers and reduce errors caused by human negligence.

[0319] An embodiment of the present invention further provides a computer device, Figure 7 It is a schematic diagram of a computer device in an embodiment of the present invention. The computer device 700 includes a memory 710, a processor 720, and a computer program 730 stored in the memory 710 and executable on the processor 720. When the processor 720 executes the computer program 730, the above-mentioned oil drilling trajectory dynamic control method is implemented.

[0320] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned oil drilling trajectory dynamic control method is implemented.

[0321] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the above-mentioned oil drilling trajectory dynamic control method is implemented.

[0322] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0323] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0324] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0325] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0326] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for dynamic control of oil drilling trajectory, characterized in that: include: Collect dynamic trajectory control data at the same depth; Determine the current drilling status based on dynamic trajectory control data; According to the dynamic trajectory control data, the trajectory state parameter change rate of the current drilling state is calculated; Taking the characteristic values ​​in the dynamic trajectory control data as input and the trajectory state parameter change rate of the current drilling state as output, training is performed using machine learning to obtain a prediction model corresponding to the current drilling state; According to the dynamic trajectory control data sequence and trajectory state parameter change rate sequence of the current section, a prediction model is used to predict the trajectory state parameter change rate of the new well section drilled by the current drill bit, and according to the predicted trajectory state parameter change rate prediction value, the trajectory state parameter at the current drill bit position is calculated; Taking the trajectory state parameters at the current drill bit position as the benchmark, the target trajectory state parameters as the objective function, and the maximum value of the trajectory state parameter change rate corresponding to the unit footage of the well section as the constraint condition, the prediction model corresponding to the current drilling state is used for dynamic solution to obtain at least one set of recommended trajectory control parameters and their corresponding trajectory state parameter change rates, and calculate the drilling distance to reach the target trajectory state parameters; The drilling trajectory is controlled according to the trajectory control parameters and the drilling distance.

2. The method according to claim 1, characterized in that The dynamic trajectory control data includes drilling data collected from the wellhead and data collected from the wireless while-drilling inclinometer; When the current drilling state is the composite drilling state, the drilling data includes drilling pressure, rotation speed, well depth and displacement, the data collected by the wireless drilling inclinometer includes well inclination angle and azimuth angle, the trajectory state parameter change rate includes the deflection capacity and the azimuth angle change rate, the trajectory state parameters include well inclination angle and azimuth angle, and the trajectory control parameters include drilling pressure, rotation speed and displacement; When the current drilling state is the directional sliding drilling state, the drilling data includes drilling pressure, well depth, and displacement; the data collected by the wireless downhole inclinometer includes well inclination angle, tool face, and azimuth angle; the trajectory state parameter change rate includes the deflection capacity, azimuth angle change rate, and tool face swing amplitude; the trajectory state parameters include well inclination angle, azimuth angle, and tool face swing amplitude; and the trajectory control parameters include drilling pressure and displacement; The method further comprises: The drilling data collected from the wellhead and the data collected from the wireless while-drilling inclinometer are aligned at the same depth as a set of data, and the same depth is used as a depth offset value.

3. The method according to claim 2, characterized in that According to the dynamic trajectory control data, the current drilling status is determined, including: If there is drilling pressure and the well depth changes, it is determined that the current state is drilling; If the rotation speed is collected, it is determined that the drilling state is a composite drilling state; if the rotation speed is not collected, it is determined that the drilling state is a directional sliding drilling state; If there is no drilling pressure and the well depth does not change, it is determined that the drilling state is not in progress.

4. The method according to claim 2, characterized in that: According to the dynamic trajectory control data, the trajectory state parameter change rate of the current drilling state is calculated, including: Determine the difference between the current dynamic trajectory control data and the last dynamic trajectory control data corresponding to the current drilling state; When the difference in the situation exceeds the expected change, the current dynamic trajectory control data is regarded as the first piece of the next section of dynamic trajectory control data corresponding to the current drilling state; When the situation difference does not exceed a preset range, the rate of change of the trajectory state parameter of the current drilling state is calculated.

5. The method according to claim 2, characterized in that: Determine the difference between the current dynamic trajectory control data and the last dynamic trajectory control data corresponding to the current drilling state, including: If the difference between the depth of the current dynamic trajectory control data and the depth of the last dynamic trajectory control data corresponding to the current drilling state is greater than a first preset value, or the difference between the acquisition time of the current dynamic trajectory control data and the acquisition time of the last dynamic trajectory control data corresponding to the current drilling state is greater than a second preset value, or there is data of different drilling states between the current dynamic trajectory control data and the last dynamic trajectory control data corresponding to the current drilling state, it is determined that the difference between the situations of the current dynamic trajectory control data and the last dynamic trajectory control data corresponding to the current drilling state exceeds an expected change.

6. The method according to claim 2, characterized in that According to the dynamic trajectory control data, the trajectory state parameter change rate of the current drilling state is calculated, including: The deflection capacity is calculated using the following formula: ΔS i =(S i -S i-1 ) / (D i -D i-1 ) Among them, ΔS i is the deflection capability of dynamic trajectory control data, S i is the well inclination angle in the current dynamic trajectory control data, S i-1 is the well inclination angle in the previous dynamic trajectory control data, D i is the well depth in the current dynamic trajectory control data, D i-1 The well depth in the control data for the last dynamic trajectory; The azimuth change rate is calculated using the following formula: ΔA i =(A i -A i-1 ) / (D i -D i-1 ) ΔA i is the azimuth change rate of the current dynamic trajectory control data, A i is the azimuth in the current dynamic trajectory control data, A i-1 Control the azimuth in the data for the previous dynamic trajectory; The tool face swing amplitude is calculated using the following formula: ΔW i is the tool face swing amplitude of the current dynamic trajectory control data, W i is the tool face in the current dynamic trajectory control data, W i-1 Control the toolface in the data for the last dynamic trajectory.

7. The method according to claim 2, characterized in that Taking the characteristic values ​​in the dynamic trajectory control data as input and the trajectory state parameter change rate of the current drilling state as output, the prediction model corresponding to the current drilling state is obtained by training using machine learning, including: Each time a new set of data is obtained, or according to the preset interval, the prediction model is trained according to the following steps: When the current drilling state is a composite drilling state, starting from the second dynamic trajectory control data in the current segment of dynamic trajectory control data, according to the start and end time of the previous dynamic trajectory control data and the current dynamic trajectory control data during drilling, the average values ​​of drilling pressure, rotation speed and displacement are taken as characteristic values; the deflection capacity and azimuth change rate are taken as the trajectory state parameter change rates, and training is performed using a machine learning method to obtain a prediction model corresponding to the composite drilling state; When the current drilling state is a directional sliding drilling state, starting from the second dynamic trajectory control data in the current section of the dynamic trajectory control data, according to the start and end time of the previous dynamic trajectory control data and the current dynamic trajectory control data during drilling, the average values ​​of the drilling pressure, the average values ​​of the displacement and the average values ​​of the tool face are taken as characteristic values; the deflection ability, the azimuth change rate and the tool face swing amplitude are taken as the trajectory state parameter change rates, and training is performed using a machine learning method to obtain a prediction model corresponding to the directional sliding drilling state.

8. The method according to claim 7, characterized in that Also includes: Based on the construction data of historical wells, pre-trained models corresponding to different drilling states are trained; According to the pre-trained models corresponding to different drilling states, incremental training is performed on the prediction models corresponding to different drilling states.

9. The method according to claim 2, characterized in that: According to the dynamic trajectory control data sequence and trajectory state parameter change rate sequence of the current section, the prediction model is used to predict the trajectory state parameter change rate of the new well section drilled by the current drill bit, including: When the current drilling state is the composite drilling state, if the variation ranges of the drilling pressure, rotation speed and displacement of the wellbore drilling between the depth of the current measuring point and the depth of the drill bit position are all less than the preset variation range, the drilling pressure, rotation speed and displacement are averaged; the average values ​​of the drilling pressure, rotation speed and displacement are input into the prediction model corresponding to the composite drilling state, and the deflection capacity and azimuth change rate of the new drilling section of the current drill bit are predicted, and the deflection capacity prediction value and the azimuth change rate prediction value are obtained; When the current drilling state is the directional sliding drilling state, if the change ranges of the drilling pressure, displacement and tool face of the wellbore drilling between the depth of the current measuring point and the depth of the drill bit position are less than the preset change range, the drilling pressure, displacement and tool face are averaged; the average values ​​of the drilling pressure, displacement and tool face are input into the prediction model corresponding to the directional sliding drilling state to predict the deflection capacity, azimuth change rate and tool face swing amplitude of the new drilling section of the current drill bit, and obtain the deflection capacity prediction value, azimuth change rate prediction value and tool face swing amplitude prediction value.

10. The method according to claim 9, characterized in that According to the predicted value of the trajectory state parameter change rate, the trajectory state parameter at the current drill bit position is calculated, including: Use the following formula to calculate the well inclination angle at the current drill bit location: S=S i +(L×ΔS n ) Among them, S is the well inclination angle of the current drill bit position, S i is the well inclination angle at the current measuring point, ΔS n is the predicted value of the deflection capacity of the newly drilled well section, and L is the distance between the current drill bit and the current measuring point; Use the following formula to calculate the azimuth of the current drill position: A=A i +(L×ΔA n ) Among them, A is the azimuth of the current drill position, A i is the azimuth of the current measuring point, L is the distance between the drill head and the measuring point, ΔA n It is the predicted value of the azimuth change rate of the newly drilled well section; The following formula is used to calculate the tool face swing amplitude at the current drill bit position: ΔW=ΔW n Among them, ΔW is the azimuth of the current drill bit position, ΔW n It is the predicted value of tool face swing amplitude.

11. The method according to claim 2, characterized in that According to the dynamic trajectory control data sequence and trajectory state parameter change rate sequence of the current section, the prediction model is used to predict the trajectory state parameter change rate of the new well section drilled by the current drill bit, including: When the current drilling state is a composite drilling state, if the change amplitude of any one of the drilling pressure, rotation speed and displacement of the wellbore drilling between the depth of the current measuring point and the depth of the drill bit position is not less than the preset change amplitude, and the segment length of the dynamic trajectory control data is greater than the minimum segment, the dynamic trajectory control data is segmented to form multiple data segments, and an average value is taken for each data segment; for each data segment, the average value of the drilling pressure, the average value of the rotation speed and the average value of the displacement of the data segment are input into the corresponding prediction model, and the deflection ability and azimuth change rate of each data segment are predicted, so as to obtain the deflection ability prediction value and azimuth change rate prediction value of each data segment; When the current drilling state is the directional sliding drilling state, if the change amplitude of any one of the drilling pressure, displacement and tool face of the wellbore drilling between the depth of the current measuring point and the depth of the drill bit position is not less than the preset change amplitude, and the segment length of the dynamic trajectory control data is greater than the minimum segment, the dynamic trajectory control data is segmented to form multiple data segments, and an average value is taken for each data segment; for each data segment, the average value of the drilling pressure, the average value of the displacement and the average value of the tool face of the data segment are input into the corresponding prediction model to predict the deflection ability, azimuth change rate and tool face swing amplitude of each data segment, and obtain the deflection ability prediction value, azimuth change rate prediction value and tool face swing amplitude prediction value of each data segment.

12. The method according to claim 11, characterized in that According to the predicted value of the trajectory state parameter change rate, the trajectory state parameter at the current drill bit position is calculated, including: Use the following formula to calculate the well inclination angle at the current drill bit location: Among them, S is the well inclination angle of the current drill bit position, S k is the well inclination angle corresponding to the kth data segment of the current measurement point, L k is the segment length corresponding to the kth data segment, ΔS k is the predicted value of the deflection capacity of the newly drilled well section corresponding to the kth data segment, and n is the number of data segments; Use the following formula to calculate the azimuth of the current drill position: Among them, A is the azimuth of the current drill position, A k is the azimuth corresponding to the kth data segment of the current measurement point, ΔA k is the predicted value of the azimuth change rate of the new drilling section corresponding to the kth data segment; The following formula is used to calculate the tool face swing amplitude at the current drill bit position: ΔW=ΔW n Among them, ΔW is the azimuth of the current drill bit position, ΔW n It is the predicted value of the tool face swing amplitude corresponding to the last data segment.

13. The method according to claim 1, characterized in that Also includes: Modeling the drilling tool mechanics to obtain the drilling tool mechanics model; At least one set of recommended trajectory control parameters and their corresponding trajectory state parameter change rates are calculated according to the drilling tool mechanics model.

14. The method according to claim 1, characterized in that Also includes: The dynamic trajectory control data are grouped according to the depth intervals; Analyze the trajectory control rules of different groups; According to the analyzed trajectory control law, at least one set of recommended trajectory control parameters and their corresponding trajectory state parameter change rates are predicted.

15. The method according to claim 1, characterized in that Also includes: After depth migration of the dynamic trajectory control data collected at the same depth, artificial intelligence models corresponding to different drilling states are trained respectively; Based on the measurement point and all previous dynamic trajectory control data as a sample sequence, the artificial intelligence model is used to iteratively solve the problem in the following steps until the trajectory state parameters of the current position of the drill bit are solved: Taking the average value of the measured dynamic trajectory control data at the next predicted depth as input, the trajectory state parameters at the next depth are predicted; The predicted values ​​of trajectory state parameters and the measured trajectory state parameters are put into a new set of data at the end of the sample sequence.

16. The method according to claim 1, characterized in that Also includes: When the current drilling state is the composite drilling state, after each static measurement point, the difference between the dynamic well inclination angle and the static well inclination angle of the last static measurement point is calculated and used as the error interval of the well inclination angle, and the difference between the dynamic azimuth angle and the static azimuth angle of the last static measurement point is calculated and used as the error interval of the azimuth angle; According to the depths of the last two static measurement points, for the covered dynamic measurement points, the error interval of the well inclination angle and the azimuth angle are allocated to each dynamic measurement point according to the ratio of the depths between adjacent points, and the value of the last dynamic measurement point is corrected to the value of the corresponding static measurement point; after the value of the dynamic measurement point is corrected, the corresponding prediction model is retrained; When the current drilling state is the directional sliding drilling state, after each static measurement point, the difference between the dynamic well inclination angle and the static well inclination angle of the last static measurement point is calculated and used as the error interval of the well inclination angle, the difference between the dynamic azimuth angle and the static azimuth angle of the last static measurement point is calculated and used as the error interval of the azimuth angle, and the difference between the dynamic tool face and the static tool face of the last static measurement point is calculated and used as the tool face interval of the azimuth angle; According to the depths of the last two static measurement points, for the covered dynamic measurement points, the error intervals of the well inclination, the error intervals of the azimuth, and the error intervals of the tool face are allocated to each dynamic measurement point according to the ratio of the depths between adjacent points, and the value of the last dynamic measurement point is corrected to the value of the corresponding static measurement point; after the value of the dynamic measurement point is corrected, the corresponding prediction model is retrained.

17. A dynamic control device for oil drilling trajectory, characterized in that: include: Data acquisition module, used to collect dynamic trajectory control data at the same depth; A drilling state determination module is used to determine the current drilling state according to the dynamic trajectory control data; A data preprocessing module is used to calculate the trajectory state parameter change rate of the current drilling state according to the dynamic trajectory control data; A model training module is used to take the characteristic values ​​in the dynamic trajectory control data as input, take the trajectory state parameter change rate of the current drilling state as output, and use machine learning to perform training to obtain a prediction model corresponding to the current drilling state; The drill bit inclination angle prediction module is used to predict the trajectory state parameter change rate of the new well section drilled by the current drill bit based on the dynamic trajectory control data sequence and trajectory state parameter change rate sequence of the current section, and calculate the trajectory state parameter at the current drill bit location based on the predicted trajectory state parameter change rate prediction value; The trajectory control parameter recommendation module is used to take the trajectory state parameters of the current drill bit position as the benchmark, the target trajectory state parameters as the objective function, and the maximum value of the trajectory state parameter change rate corresponding to the unit footage of the well section as the constraint condition, and use the prediction model corresponding to the current drilling state to perform dynamic solution, obtain at least one set of recommended trajectory control parameters and their corresponding trajectory state parameter change rate, and calculate the drilling distance to reach the target trajectory state parameters; The trajectory control module is used to control the drilling trajectory according to the trajectory control parameters and the drilling distance.

18. The device according to claim 17, characterized in that Also includes a formula prediction module for: Modeling the drilling tool mechanics to obtain the drilling tool mechanics model; At least one set of recommended trajectory control parameters and their corresponding trajectory state parameter change rates are calculated according to the drilling tool mechanics model.

19. The device according to claim 17, characterized in that It also includes a group calculation module for: The dynamic trajectory control data are grouped according to the depth intervals; Analyze the trajectory control rules of different groups; According to the analyzed trajectory control law, at least one set of recommended trajectory control parameters and their corresponding trajectory state parameter change rates are predicted.

20. The device according to claim 17, characterized in that It also includes an artificial intelligence model solving module for: After depth migration of the dynamic trajectory control data collected at the same depth, artificial intelligence models corresponding to different drilling states are trained respectively; Based on the measurement point and all previous dynamic trajectory control data as a sample sequence, the artificial intelligence model is used to iteratively solve the problem in the following steps until the trajectory state parameters of the current position of the drill bit are solved: Taking the average value of the measured dynamic trajectory control data at the next predicted depth as input, the trajectory state parameters at the next depth are predicted; The predicted values ​​of trajectory state parameters and the measured trajectory state parameters are put into a new set of data at the end of the sample sequence.

21. The device according to claim 17, characterized in that Also included is a correction module for: When the current drilling state is the composite drilling state, after each static measurement point, the difference between the dynamic well inclination angle and the static well inclination angle of the last static measurement point is calculated and used as the error range of the well inclination angle; According to the depths of the last two static measurement points, for the covered dynamic measurement points, the error interval of the well inclination angle and the azimuth angle are allocated to each dynamic measurement point according to the ratio of the depths between adjacent points, and the value of the last dynamic measurement point is corrected to the value of the corresponding static measurement point; after the value of the dynamic measurement point is corrected, the corresponding prediction model is retrained; When the current drilling state is the directional sliding drilling state, after each static measurement point, the difference between the dynamic well inclination angle and the static well inclination angle of the last static measurement point is calculated and used as the error interval of the well inclination angle, the difference between the dynamic azimuth angle and the static azimuth angle of the last static measurement point is calculated and used as the error interval of the azimuth angle, and the difference between the dynamic tool face and the static tool face of the last static measurement point is calculated and used as the tool face interval of the azimuth angle; According to the depths of the last two static measurement points, for the covered dynamic measurement points, the error intervals of the well inclination, the error intervals of the azimuth, and the error intervals of the tool face are allocated to each dynamic measurement point according to the ratio of the depths between adjacent points, and the value of the last dynamic measurement point is corrected to the value of the corresponding static measurement point; after the value of the dynamic measurement point is corrected, the corresponding prediction model is retrained.

22. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 16 is implemented.

23. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 16 is implemented.

24. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 16 is implemented.

Citation Information

Cited By

  • Directional well measurement method and system based on cloud decoding technology

    CN119149875A

  • Method and device for designing real drilling deviation rectification borehole trajectory

    CN119337489A

  • Directional drilling clustering trajectory correction method based on acceleration sensing

    CN120026894A

  • Exploration drilling parameter control method and system

    CN120575835A

  • Omnibearing deviation prevention device for geological exploration and use method thereof

    CN121138720A