CN118933708A

CN118933708A - Oil drilling trajectory dynamic control method and device

Info

Publication number: CN118933708A
Application number: CN202411362573.9A
Authority: CN
Inventors: 张宗林; 龚翔旻; 侯明扬; 胡强; 范兆祥; 张棒; 王朋
Original assignee: Intercontinental Strait Energy Technology Beijing Co ltd; Zhouji Strait Energy Technology Co ltd
Current assignee: Intercontinental Strait Energy Technology Beijing Co ltd; Zhouji Strait Energy Technology Co ltd
Priority date: 2024-09-27
Filing date: 2024-09-27
Publication date: 2024-11-12

Abstract

The invention discloses a dynamic control method and a dynamic control device for petroleum drilling tracks, wherein the method comprises the following steps: collecting dynamic track control data at the same depth; calculating the track state parameter change rate of the current drilling state, and training to obtain a prediction model corresponding to the current drilling state; predicting the change rate of the track state parameter of the newly drilled well section of the current drill bit by using a prediction model, and calculating the track state parameter of the position of the current drill bit according to the predicted change rate prediction value of the track state parameter; and taking the track state parameter of the current drill bit as a reference, taking the target track state parameter as an objective function, taking the maximum value of the track state parameter change rate corresponding to the well section unit footage as a constraint condition, dynamically solving by using a prediction model corresponding to the current drilling state to obtain at least one group of recommended track control parameters, and calculating the drilling distance reaching the target track state parameter to control the drilling track. The invention can realize the accurate control of the well track.

Description

Dynamic control method and device for petroleum drilling track

Technical Field

The invention relates to the technical field of petroleum drilling, in particular to a dynamic control method and device for petroleum drilling tracks.

Background

This section is intended to provide a background or context to the embodiments of the invention that are recited in the claims. The description herein is not admitted to be prior art by inclusion in this section.

In the existing petroleum drilling track dynamic control technology, there are two modes:

first kind: rotary guiding directional drilling

Directional drilling is performed using a rotary steerable apparatus. The pump pressure pulse of the circulating mud or the rotary drilling speed change of the drilling tool are used for giving instructions at the ground surface, and the control, measurement and feedback through the liquid pulse are performed at the bottom of the well through the rotary guide.

Disadvantages of rotary steerable directional drilling are:

(a) High cost

At present, the comprehensive cost of using a rotary guide for one well in a certain oil field is 17 ten thousand yuan/day. The drilling cycle of the deflecting section of a well is generally one week, and a rotary guide may be used in a horizontal section, and the cost of using the rotary guide for a well is more than 100 ten thousand yuan on average.

(B) Slow drilling

Because of the limitation of the rotary guiding equipment, indexes such as pressure, drilling speed and the like applied from a wellhead are limited, so that high-speed drilling of the drill bit is affected. Such as a 215.9mm wellbore, is typically incapable of applying a weight on bit of 150kN to 200kN to the rotary steerable, but some poorly drillable formations must be effective with high weight on bit, high rotational speeds.

(C) High risk

For the well bore which can have serious underground leakage or fracture zone easy to collapse, the complex such as drill sticking and the like can occur, and even the accident of burying the rotary guiding instrument can cause serious loss.

(D) Limited conditions of use

If a high temperature environment is often present at the bottom of the well, the electronic components of the screw guide may fail and cannot be used.

Second kind: directional drilling with curved screw

The proportion of horizontal wells in oilfield development is higher and higher, and the track control is mostly carried out by adopting a curved screw rod and MWD drilling tool combination in the current horizontal well drilling process. The track control is generally carried out by adopting a static well inclination angle method, and the general flow is as follows:

(a) Static in-situ inclinometry: the static well inclination angle is well inclination angle data measured by the drilling tool in a static state. In the field top drive drilling process, each time one upright post (3 single posts, which are 28.5m in total) is drilled, the drill bit is lifted off the bottom of the well to measure the static well inclination angle and azimuth angle once. The measuring process is to stop the pump for 5 minutes, then start the pump to the circulation displacement, wait for the transmission of the well bottom measuring data to the ground, and predict about 10-15 minutes, the total time for completing one static inclinometry on site is generally more than 30 minutes.

(B) The directional engineer uses static measurement data to analyze through drilling track calculation software, predicts the change trend of the well bottom track, and selects the subsequent drilling mode (sliding directional drilling or compound drilling) and drilling parameters according to the design section.

(C) Static measurement of well inclination is affected by factors such as slurry medium, slurry pump efficiency, driller operation, etc., and sometimes the inclination measurement is unsuccessful, the measurement needs to be carried out again according to the steps.

The defects of the current bent screw directional drilling are as follows:

(1) Inaccurate well inclination angle measurement

Because the "PDC drill bit + bent screw + resistivity + MWD + … …" drill assembly is generally used, the measurement point is about 18m-22m away from the zero length of the drill bit position, resulting in that the measured data cannot immediately reflect the actual situation of the drill bit position, and the measures taken lag, which affects the accurate control of the track.

(2) Well inclination angle measurement interval is large

The measurement is generally carried out after one upright post is punched, the length of the upright post of the land and sea drilling machine is between 19 meters and 30 meters, the interval between each measurement point is large, and the measurement is delayed when a problem is found. While MWD is capable of transmitting current well inclination and azimuth in real time, such dynamic information is generally not effectively utilized.

(3) Directional sliding drilling inefficiency

For the reasons, the change trend of the well bottom track predicted by the directional engineer is sometimes inaccurate, and the deviation between the track and the designed track needs to be corrected urgently after the deviation is excessive, so that the composite drilling is abandoned and changed into the directional sliding drilling, the situation that directional inclination increase is needed for a while, directional inclination reduction is needed for a while, and the well bore is not bent and smooth is caused in construction. Frequent directional sliding drilling is due to the fact that the drilling tool does not rotate, and the speed of the well bore easy-to-hold-down machine is low; and there is a risk of sticking. In order to eliminate the pressure and prevent the drill sticking, the drilling tool needs to be frequently moved, and the drilling efficiency is low. Meanwhile, when directional sliding drilling is performed, the tool surface can swing greatly and is unstable due to fluctuation of screw torque, the tool surface is difficult to control, and the directional efficiency is low.

(4) Directional sliding drilling easy-to-damage drill bit

In directional sliding drilling, because static friction is large, a large weight on bit is sometimes applied to overcome the static friction in order to push the drill rod forward. Once the drill rod begins to slide, the dynamic friction is quickly reduced, and the drill bit can collide with the stratum severely, so that the drill bit is damaged abnormally. After the drill bit is damaged, the rate of penetration may be affected and may further result in unnecessary tripping conditions. Every trip of tripping needs to take out all drilling tools in the well, and the drilling tools are restarted after being replaced, so that the effective working time is wasted for more than 30-50 hours according to different well depths.

(5) Subjective factors affect a relatively large extent

The actual work is up to the degree in a few meters according to the requirements of the geosteering operator. The work effect is related to the orientation engineer.

Thus, there is currently a lack of a solution to reduce drilling costs while improving the efficiency of current construction methods.

Disclosure of Invention

In a first aspect, an embodiment of the present invention provides a method for dynamically controlling a petroleum drilling trajectory, capable of predicting a trajectory state parameter change rate (a well inclination angle, an azimuth angle, etc.) of a current bit position in real time without using a rotary guide, so as to accurately control a borehole trajectory, improve a drilling speed, and reduce damage to a bit, thereby reducing costs and enhancing efficiency, the method comprising:

collecting dynamic track control data at the same depth;

Determining the current drilling state according to the dynamic track control data;

calculating the track state parameter change rate of the current drilling state according to the dynamic track control data;

taking a characteristic value in the dynamic track control data as input, taking the track state parameter change rate of the current drilling state as output, and training by using a machine learning mode to obtain a prediction model corresponding to the current drilling state;

According to the dynamic track control data sequence and the track state parameter change rate sequence of the current section, predicting the track state parameter change rate of the newly drilled well section of the current drill bit by using a prediction model, and calculating the track state parameter of the position of the current drill bit according to the predicted track state parameter change rate predicted value;

Taking the track state parameter of the current drill bit as a reference, taking the target track state parameter as an objective function, taking the maximum value of the track state parameter change rate corresponding to the well section unit footage as a constraint condition, dynamically solving by using a prediction model corresponding to the current drilling state to obtain at least one group of recommended track control parameters and the track state parameter change rate corresponding to the recommended track control parameters, and calculating the drilling distance reaching the target track state parameter;

and controlling the drilling track according to the track control parameters and the drilling distance.

In a second aspect, an embodiment of the present invention further provides a dynamic control device for a petroleum drilling track, which can predict parameters such as a well inclination angle and an azimuth angle of a current bit position in real time without using a rotary guide, so as to accurately control the drilling track, improve a drilling speed, reduce damage of the bit, and thereby reduce cost and increase efficiency, and the device includes:

the data acquisition module is used for acquiring dynamic track control data at the same depth;

The drilling state determining module is used for determining the current drilling state according to the dynamic track control data;

the data preprocessing module is used for calculating the track state parameter change rate of the current drilling state according to the dynamic track control data;

The model training module is used for training by taking a characteristic value in the dynamic track control data as input and the track state parameter change rate of the current drilling state as output and using a machine learning mode to obtain a prediction model corresponding to the current drilling state;

The drill bit well angle prediction module is used for predicting the change rate of the track state parameters of the newly drilled well section of the current drill bit by using the prediction model according to the dynamic track control data sequence and the track state parameter change rate sequence of the current section, and calculating the track state parameters of the position of the current drill bit according to the predicted change rate prediction value of the track state parameters;

The track control parameter recommendation module is used for dynamically solving by using a prediction model corresponding to the current drilling state, taking track state parameters of the current drill bit as a reference, taking target track state parameters as a target function and taking the maximum value of track state parameter change rate corresponding to the unit footage of a well section as a constraint condition, obtaining at least one group of recommended track control parameters and track state parameter change rates corresponding to the at least one group of recommended track control parameters, and calculating the drilling distance reaching the target track state parameters;

and the track control module is used for controlling the drilling track according to the track control parameters and the drilling distance.

In a third aspect, an embodiment of the present invention further provides a computer device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, where the processor implements the method for dynamically controlling the petroleum drilling track when executing the computer program.

In a fourth aspect, embodiments of the present invention further provide a computer readable storage medium storing a computer program, which when executed by a processor, implements the above-described method for dynamically controlling a petroleum drilling trajectory.

In a fifth aspect, embodiments of the present invention also provide a computer program product comprising a computer program which, when executed by a processor, implements the above-described method of dynamic control of petroleum drilling trajectories.

In the embodiment of the invention, the dynamic track control data under the same depth is collected; determining the current drilling state according to the dynamic track control data; calculating the track state parameter change rate of the current drilling state according to the dynamic track control data; taking a characteristic value in the dynamic track control data as input, taking the track state parameter change rate of the current drilling state as output, and training by using a machine learning mode to obtain a prediction model corresponding to the current drilling state; according to the dynamic track control data sequence and the track state parameter change rate sequence of the current section, predicting the track state parameter change rate of the newly drilled well section of the current drill bit by using a prediction model, and calculating the track state parameter of the position of the current drill bit according to the predicted track state parameter change rate predicted value; taking the track state parameter of the current drill bit as a reference, taking the target track state parameter as an objective function, taking the maximum value of the track state parameter change rate corresponding to the well section unit footage as a constraint condition, dynamically solving by using a prediction model corresponding to the current drilling state to obtain at least one group of recommended track control parameters and the track state parameter change rate corresponding to the recommended track control parameters, and calculating the drilling distance reaching the target track state parameter; and controlling the drilling track according to the track control parameters and the drilling distance. Compared with the scheme of adopting rotary directional drilling and curved screw directional drilling in the prior art, the embodiment of the invention takes the characteristic value in the dynamic trajectory control data corresponding to the current drilling state as input, takes the change rate of the trajectory state parameter of the current drilling state as output, trains a corresponding prediction model, can be used for predicting the trajectory state parameter of the current bit position and recommending the trajectory control parameter, thereby calculating the accurate drilling distance, effectively controlling the drilling trajectory, improving the drilling speed, reducing the bit damage and reducing the cost and enhancing the efficiency.

Drawings

In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. In the drawings:

FIG. 1 is a flow chart of a method for dynamically controlling petroleum drilling trajectories in an embodiment of the present invention;

FIG. 2 is a graph showing the effect of implementing two methods of using a progressive cavity drilling tool to achieve full interval drilling in accordance with the present invention;

FIG. 3 is a drawing of a drilling pattern for a three-well horizontal leg in an embodiment of the invention;

FIG. 4 is a schematic diagram of a method according to an embodiment of the present invention applied to a well;

FIG. 5 is a chart of dynamic well inclination and weight on bit, rotational speed change for a well 4760m-4900m in accordance with an embodiment of the present invention;

FIG. 6 is a schematic diagram of a dynamic control device for petroleum drilling trajectories in an embodiment of the present invention;

Fig. 7 is a schematic diagram of a computer device according to an embodiment of the present invention.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the embodiments of the present invention will be described in further detail with reference to the accompanying drawings. The exemplary embodiments of the present invention and their descriptions herein are for the purpose of explaining the present invention, but are not to be construed as limiting the invention.

Terms related to the embodiments of the present invention will be explained first.

Wellbore trajectory: in petroleum drilling, the trajectory of the borehole of the drilling well is often adjusted artificially, commonly known as "directional", according to different needs of geology, engineering, etc. (e.g. to ensure that the wellbore progresses along the formation of the hydrocarbon reservoir in an undulating manner to achieve higher yields). During drilling, the well inclination angle and the azimuth angle of the well extension need to be adjusted, the well inclination angle change per unit drilling distance is called a well inclination angle change rate, and the azimuth angle change rate is the same. The method of adjustment is called "whipping" and includes ramping up (increasing the well angle), ramping down (decreasing the well angle), and stabilizing (maintaining the current well angle constant). In order to realize deflecting, two modes of rotary guiding equipment (rotary guiding) or a bent screw drilling tool are generally adopted.

Well section: when a horizontal well starts to drill, the general borehole track is in the vertical direction, and the section becomes a straight well section; when the well is quickly drilled to reach the target layer, the well track is gradually flattened and the target layer keeps parallel, and the well section is increased from the well section to the well inclination angle between the horizontal direction, which is called as a deflecting section; after entering the destination layer, the well track is kept in the destination layer, and the section becomes a horizontal section (the well inclination angle is about 90 °);

And (3) rotation guide: spin-steering is a valuable precision instrument, easily damaged (because of the large number of mechanical and electrical components), and equipment that is harsh in terms of use (e.g., applied pressure, bottom hole temperature, etc.). The rotary guiding tool can actively control the drilling direction of the drill bit after the rotary guiding tool goes to the bottom of the well and gives out instructions through the ground, so that the well track can be accurately drilled to the expected azimuth angle and the well bevel angle.

Screw drilling tool: the screw drilling tool is a downhole power drilling tool which uses drilling fluid circulated in a well bore as power and converts the fluid pressure energy into mechanical energy, and can provide additional rotating force for a drill bit while a drill rod rotates. The method is simple and cheap to use, stable in performance and high in efficiency, but the control of the well track is not accurate.

Curved screw drilling tool: the curved screw drilling tool front section has a certain angle of curvature (e.g., 1 °, 1.25 °, 1.5 °, 1.75 °, etc.), with which the well angle and azimuth can be adjusted by directional drilling. The bending screw drilling tool comprises two modes of compound drilling and directional sliding drilling, wherein the compound drilling refers to simultaneous rotation of a drill rod and the screw drilling tool, and the drilling speed is higher; the directional sliding drilling means that only the screw rotor rotates and the drill rod does not rotate, the well bevel angle and the azimuth angle are controlled through the bent angle of the screw drilling tool in a directional mode, and the directional sliding drilling is generally used when the well bevel angle and the azimuth angle are required to be adjusted rapidly in a large range. When directional sliding drilling is performed, the elbow of the screw rod is aligned to a certain direction (commonly called a swinging tool surface), which is called a tool surface, so that drilling in a specified direction can be performed more effectively. In the composite drilling process, the curved screw drilling tool transmits the current well inclination angle and azimuth angle to the ground through an MWD (Wireless while drilling inclinometer); tool face data is also dynamically transmitted during directional slide drilling, where the tool face may be subject to wobble.

Inclinometry: the above-described well inclination angle measured during drilling is called dynamic inclinometry. Dynamic inclinometry can have errors because the device is continuously moving during the measurement. The drilling tool stops rotating and drilling fluid circulates and the downhole drilling tool is stationary for measurements called static well angle measurements. The static inclinometry distance is different according to different conditions, when a top drive exists in the current land and offshore drilling sites, the spacing is a drill rod column, and the length is about 28.5m (the length of a drill rod single is about 9.5m in general); for a land drilling machine without a top drive, the distance is generally about 9.5m for a drill rod single part. Static well angles are more accurate, but can impact production efficiency because of intermittent downtime (typically taking more than 30 minutes in total to complete a static survey on site). The measurement of the data is typically performed by a dedicated measurement nipple-wireless measurement while drilling instrument, i.e. MWD. The MWD is behind the nonmagnetic drill pipe or collar above the screw drilling tool, away from the drill bit, typically a distance of 18-22 m.

The inventors believe that the whipstock rules may vary from well zone to well zone, from horizon lithology to horizon, from bent screw combination to bent screw combination (including position, number, diameter, type differences of centralizers), and with wear of the bit (PDC) and centralizer, under the same parameters. If the physical model is to be built with a particularly great difficulty, it involves geomechanics, rock mechanics, drilling mechanics, etc., and the calculation is very large, the calculation time of a conventional machine may be in hours.

From the basic logic, no changes in the drilling assembly occur in the same pass, and the geological conditions, drilling fluid properties, wellbore trajectory changes are generally small, which can be ignored. The actual whipstock situation has a great correlation with the current construction parameters. During compound drilling, under the conditions of different drilling weights and rotating speeds, the change rate of the well-making oblique angle and the azimuth angle also changes; in the case of sliding drilling, the well-forming oblique angle, the azimuth angle change rate and the tool face swing amplitude are different under the condition of different drilling pressures. The displacement is also a related parameter, but is weakly related, since the screw will rotate the drill bit.

The principle of the embodiment of the invention is that an artificial intelligence method is used, under the condition of using a bent screw, continuous machine learning is carried out according to drilling depth (such as 1 meter) or time interval (such as 55 seconds) based on dynamic track control data according to the combined drilling and sliding drilling working conditions, well inclination angle and azimuth angle of the current bit position are predicted in real time based on a learned AI model and current construction parameters, correction and error analysis are carried out by combining with static well inclination angle data acquired subsequently, and more accurate empirical track closed-loop control under all working conditions of the whole well is realized.

FIG. 1 is a flow chart of a method for dynamically controlling petroleum drilling trajectories in an embodiment of the invention, comprising:

Step 101, collecting dynamic track control data at the same depth;

102, determining the current drilling state according to dynamic track control data;

Step 103, calculating the track state parameter change rate of the current drilling state according to the dynamic track control data;

104, training by taking a characteristic value in the dynamic track control data as input and the track state parameter change rate of the current drilling state as output in a machine learning mode to obtain a prediction model corresponding to the current drilling state; ;

Step 105, predicting the change rate of the track state parameter of the newly drilled well section of the current drill bit by using a prediction model according to the dynamic track control data sequence and the track state parameter change rate sequence of the current section, and calculating the track state parameter of the position of the current drill bit according to the predicted change rate predicted value of the track state parameter;

Step 106, dynamically solving by using a prediction model corresponding to the current drilling state with track state parameters of the current bit position as a reference, target track state parameters as an objective function and the maximum value of track state parameter change rate corresponding to the well section unit footage as a constraint condition, obtaining at least one group of recommended track control parameters and track state parameter change rates corresponding to the at least one group of recommended track control parameters, and calculating the drilling distance reaching the target track state parameters;

and 107, performing drilling track control according to the track control parameters and the drilling distance.

Compared with the scheme of adopting rotary guide directional drilling and bent screw directional drilling in the prior art, the embodiment of the invention takes the characteristic value in the dynamic track control data corresponding to the current drilling state as input, takes the track state parameter change rate of the current drilling state as output, trains a corresponding prediction model, can be used for predicting the track state parameter of the current bit position and recommending the track control parameter, thereby calculating the accurate drilling distance, effectively controlling the drilling track, improving the drilling speed, reducing the bit damage and reducing the cost and enhancing the efficiency.

The above steps 101-107 are iterative processes, i.e. the above procedure is performed after each acquisition of the dynamic trajectory control data at the same depth, so step 107 is fed back to step 101.

Each step is described in detail below.

In step 101, dynamic trajectory control data at the same depth is acquired.

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

When the current drilling state is a composite drilling state, the drilling data comprise drilling pressure, rotating speed, well depth and displacement, the data collected by the wireless inclinometer comprise well inclination angle and azimuth angle, the track state parameter change rate comprises deflecting capability and azimuth angle change rate, the track state parameter comprises well inclination angle and azimuth angle, and the track control parameter comprises drilling pressure, rotating speed and displacement;

When the current drilling state is a directional sliding drilling state, the drilling data comprise drilling pressure, well depth and displacement, the data collected by the wireless inclinometer while drilling comprise well inclination angle, tool face and azimuth angle, the change rate of track state parameters comprise deflecting capability, azimuth angle change rate and tool face swing amplitude, the track state parameters comprise well inclination angle, azimuth angle and tool face swing amplitude, and the track control parameters comprise drilling pressure and displacement;

The method further comprises the steps of:

The drilling data collected from the wellhead and the data collected from the wireless inclinometer are aligned at the same depth as a set of data, with the same depth as the depth offset value.

Specifically, data collected from the wellhead may be collected using a drilling parameter or a comprehensive logging tool.

It is emphasized that the drilling data includes weight on bit, rotational speed, well depth and displacement, but may also include other drilling data, as may be desired. The data collected by the wireless inclinometer may include other data besides the well inclination angle and azimuth angle, and also depends on the requirements. Since the measurement point to the current MWD is a distance L from the drill bit, the data collected 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 for use as a set of data. The physical implications of this data are the wellbore interval, azimuth and toolface data measured by the current measurement point MWD, and the weight on bit, rotational speed, displacement (optional) when historically drilled to that location. This depth offset value, and thus the time offset, is taken into account in all subsequent processing logic.

The data acquisition mode can be directly acquired from an instrument or a related system, and can also be manually input.

In step 102, determining a current drilling state according to the dynamic trajectory control data;

In one embodiment, determining the current drilling status based on the dynamic trajectory control data includes:

if the drilling pressure is high and the well depth is changed, determining that the current drilling state is in;

If the rotating speed is acquired, determining the drilling state as a composite drilling state; and if the rotating speed is not acquired, determining the drilling state as a directional sliding drilling state.

And if the drilling pressure is not found and the well depth is not changed, determining that the drilling state is not currently in progress.

Specifically, when judging whether the drilling state is currently in, the sitting card and the hanging weight can be considered.

The drilling identification can be realized by adopting a state machine or other using rule judgment modes, and an artificial intelligent model can be trained to carry out intelligent identification. And judging whether the drilling state is a composite drilling state or a directional sliding drilling state, and respectively entering a corresponding composite drilling working condition processing sub-process and a corresponding directional sliding drilling working condition processing sub-process. These two sub-flows correspond.

In step 103, calculating the track state parameter change rate of the current drilling state according to the dynamic track control data;

In one embodiment, calculating the rate of change of the trajectory state parameter of the current drilling state based on the dynamic trajectory control data comprises:

judging the situation difference of the current dynamic trajectory control data and the last dynamic trajectory control data corresponding to the current drilling state;

When the situation difference exceeds the expected change, the current dynamic track control data is regarded as a first piece of next dynamic track control data corresponding to the current drilling state;

And when the situation difference does not exceed the preset range, calculating the track state parameter change rate of the current drilling state.

Specifically, the embodiment of the invention respectively introduces the data preprocessing corresponding to the composite drilling working condition processing sub-flow and the directional sliding drilling working condition processing sub-flow.

In an embodiment, determining a difference between the current dynamic trajectory control data and a last dynamic trajectory control data corresponding to the current drilling state includes:

if the difference between the depth of the current dynamic track control data and the depth of the last dynamic track control data corresponding to the current drilling state is larger than a first preset value D _max, or the difference between the acquisition time of the current dynamic track control data and the acquisition time of the last dynamic track control data corresponding to the current drilling state is larger than a second preset value T _max, or data in different drilling states exist between the current dynamic track control data and the last dynamic track control data corresponding to the current drilling state, it is determined that the difference between the situations of the current dynamic track control data and the last dynamic track control data corresponding to the current drilling state exceeds the expected change.

(1) Composite drilling condition treatment sub-process:

If the difference between the depth of the current dynamic track control data and the depth of the last piece of composite drilling data corresponding to the current drilling state is larger than a first preset value D _max, or the difference between the acquisition time of the current dynamic track control data and the acquisition time of the last piece of composite drilling data corresponding to the current drilling state is larger than a second preset value T _max, or data in different drilling states exist between the current dynamic track control data and the last piece of composite drilling data corresponding to the current drilling state, determining that the situation difference between the current dynamic track control data and the last piece of composite drilling data corresponding to the current drilling state exceeds the expected change.

Under the composite drilling working condition, the dynamic track control data is composite drilling data, and determining that the situation difference between the current dynamic track control data and the last dynamic track control data corresponding to the current drilling state exceeds the expected change means that the hidden characteristic is changed greatly, and taking the current point as the first piece of data of the next section of sample data, and waiting for the next piece of data at the moment.

(2) The directional sliding drilling working condition treatment sub-flow:

If the difference between the depth of the current dynamic track control data and the depth of the last piece of directional sliding drilling data corresponding to the current drilling state is larger than a first preset value D _max, or the difference between the acquisition time of the current dynamic track control data and the acquisition time of the last piece of directional sliding drilling data corresponding to the current drilling state is larger than a second preset value T _max, or data in different drilling states exist between the current dynamic track control data and the last piece of directional sliding drilling data corresponding to the current drilling state, it is determined that the difference between the situations of the current dynamic track control data and the last piece of directional sliding drilling data corresponding to the current drilling state exceeds the expected change.

The sub-process is consistent with the overall steps of the composite drilling condition processing sub-process. The main difference is that the wellhead rotation speed is not acquired, because the wellhead rotation speed does not exist at the moment, but the tool face and azimuth acquisition are increased; 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.

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

the whipstock capacity was calculated using the following formula:

ΔS_i＝(S_i-S_i-1)/(D_i-D_i-1)

Wherein Δs _i is the deflecting capability of the 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, and D _i-1 is the well depth in the previous dynamic trajectory control data;

The azimuth angle change rate is calculated by adopting the following formula:

ΔA_i＝(A_i-A_i-1)/(D_i-D_i-1)

Δa _i is the azimuth angle change rate of the current dynamic track control data, a _i is the azimuth angle in the current dynamic track control data, and a _i-1 is the azimuth angle in the last piece of dynamic track control data;

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

ΔW _i is the tool face swing amplitude of the current dynamic track control data, W _i is the tool face in the current dynamic track control data, and W _i-1 is the tool face in the last dynamic track control data.

In the embodiment of the invention, the characteristic value in the dynamic trajectory control data corresponding to the current drilling state is required to be used as input, the trajectory state parameter change rate of the current drilling state is required to be used as output, and the prediction models corresponding to different drilling states are trained.

In an embodiment, taking a characteristic value in dynamic trajectory control data as input, taking a trajectory state parameter change rate of a current drilling state as output, training by using a machine learning mode, and obtaining a prediction model corresponding to the current drilling state, wherein the method comprises the following steps:

After obtaining a new set of data each time, or according to a preset interval time, training a prediction model according to the following steps:

When the current drilling state is a composite drilling state, starting from the second dynamic track control data in the current section dynamic track control data, taking the average value of the weight on bit, the average value of the rotating speed and the average value of the displacement as characteristic values according to the starting and stopping time of the last dynamic track control data and the corresponding drilling of the current dynamic track control data; training by using the deflecting capability and the azimuth angle change rate as track state parameter change rates in a machine learning mode 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 piece of dynamic track control data in the current section of dynamic track control data, taking an average value of the drilling weight, an average value of the displacement and an average value of the tool surface as characteristic values according to the starting and stopping time of the last piece of dynamic track control data and the corresponding drilling time of the current dynamic track control data; training by using the deflecting capability, the azimuth angle change rate and the tool face swing amplitude as the track state parameter change rate in a machine learning mode to obtain a prediction model corresponding to the directional sliding drilling state.

The molecular flow is described in detail below.

(1) Composite drilling condition treatment sub-process:

Starting from the second piece of dynamic track control data in the current section of dynamic track control data, namely for each piece of adjacent dynamic track control data (excluding the first piece) in each section, taking an average value WOB _i of the weight on bit WOB, an average value RPM _i of the rotating speed RPM and an average value MFI _i of the displacement MFI as characteristic values according to the start-stop time when the last piece of dynamic track control data and the current dynamic track control data are correspondingly drilled.

A predictive model M _m corresponding to the multiple-input multiple-output composite drilling state is trained based on the above sequence using machine learning.

The input can be adjusted according to actual needs, such as increasing the inclination depth of the measurement point by the input. The basic input combinations are as follows:

{WOB_i,RPM_i,MFI_i}＝"{ΔS_i,ΔA_i}

(2) The directional sliding drilling working condition treatment sub-flow:

Starting from the second piece of dynamic track control data in the current section of dynamic track control data, namely for each piece of adjacent dynamic track control data (not including the first piece) in each section, taking an average value WOB _i of the weight on bit WOB, an average value MFI _i (optional) of the displacement MFI and an average value W _i of the tool face as characteristic values according to the starting and ending time of corresponding drilling of the last piece of dynamic track control data and the current dynamic track control data.

A prediction model M _m corresponding to the multi-input multi-output directional sliding drilling state is trained based on the above sequence by using a machine learning mode. The input can be adjusted according to actual needs, such as increasing the inclination depth of the measurement point by the input. The basic input combinations are as follows:

{WOB_i,MFI_i,W_i}＝"{ΔS_i,ΔA_i,ΔW_i}

In the above embodiments, the machine learning algorithm need not be limited to a specific algorithm, and may be replaced according to the development of IT technology. Considering the time series characteristics of the declination angle data and the fact that the adjacent data is more similar to the characteristics of the current data compared with the earlier data, a data mining or artificial intelligent model based on a Long Short-Term Memory network (LSTM), a normal differential equation (ODE, ordinary Differential Equations), a transducer and the like, or a machine learning algorithm combined with the model, such as a convolutional neural network (CNN-BiLSTM-attribute) integrating an Attention mechanism, can be adopted, so that the closer data occupy more weight.

After each acquisition of a new set of data, the model needs to be retrained. The interval training mode can also be adopted in consideration of the performance of the on-site computer.

In an embodiment, the method further comprises:

Training a pre-training model corresponding to different drilling states according to construction data of the historical well;

and performing incremental training on the prediction models corresponding to the different drilling states according to the pre-training models corresponding to the different drilling states.

The embodiment of the invention needs to predict the change rate of the track state parameters and predict the well inclination angle and the azimuth angle, and the embodiment of the invention needs to respectively predict the two sub-processes, and for each sub-process, the two states are respectively adopted. In the busy embodiment, three modes of prediction exist, namely, predicting and obtaining the state of the current position of the drill bit; secondly, different track control parameters are listed, the track state parameter change rate of the current bit position is predicted, and a reference is provided for the next step of adjusting construction parameters; thirdly, given the deflecting requirement of the section to be drilled, such as the expected change value of the well deflection angle in how much distance and the constraint value of the change rate of the track state parameters, the change rate of the track state parameters at different depths in the next step is automatically recommended.

Each state is divided into a simple working condition and a complex working condition.

A) Simple working condition

(1) Composite drilling condition treatment sub-process:

When the current drilling state is a composite drilling state, if the variation amplitude of the drilling pressure, the rotation speed and the displacement of the shaft drilling between the depth of the current measuring point and the depth of the position of the drill bit is smaller than the preset variation amplitude delta WOB _max,ΔRPM_max,ΔMFI_max, taking an average value of the drilling pressure, the rotation speed and the displacement; inputting an average WOB _n of the bit pressure, an average RPM _n of the rotating speed and an average MFI _n of the displacement into a prediction model corresponding to the composite drilling state, and predicting the deflecting capacity and the azimuth angle change rate of the newly drilled well section of the current drill bit to obtain a deflecting capacity predicted value and an azimuth angle change rate predicted value;

(2) The directional sliding drilling working condition treatment sub-flow:

When the current drilling state is a directional sliding drilling state, if the change amplitude of the drilling weight, the displacement and the tool face of the shaft drilling between the depth of the current measuring point and the depth of the position of the drill bit is smaller than the preset change amplitude delta WOB _max,ΔMFI_max,ΔW_max, taking average values of the drilling weight, the displacement and the tool face; and (3) inputting the average WOB _n of the bit pressure, the average MFI _n of the displacement and the average W _n of the tool face into a prediction model corresponding to the directional sliding drilling state, and predicting the deflecting capacity, the azimuth change rate and the tool face swing amplitude of the newly drilled well section of the current bit to obtain a deflecting capacity predicted value, an azimuth change rate predicted value and a tool face swing amplitude predicted value.

In one embodiment, calculating the track state parameter of the current bit position according to the predicted value of the change rate of the track state parameter, includes:

the well inclination angle of the current position of the drill bit is calculated by adopting the following formula:

S＝S_i+(L×ΔS_n)

S is the well inclination angle of the current drill bit, S _i is the well inclination angle of the current measuring point, deltaS _n is the predicted value of the deflecting ability of the newly drilled well section, and L is the distance between the current drill bit and the current measuring point;

the azimuth angle of the current position of the drill bit is calculated by adopting the following formula:

A＝A_i+(L×ΔA_n)

Wherein A is the azimuth angle of the current bit position, A _i is the azimuth angle of the current measuring point, L is the distance between the bit and the measuring point, and DeltaA _n is the predicted value of the azimuth angle change rate of the new drilling section;

The tool face swing amplitude of the current position of the drill bit is calculated by adopting the following formula:

ΔW＝ΔW_n

Wherein DeltaW is the azimuth angle of the current position of the drill bit, deltaW _n is the predicted value of the swing amplitude of the tool face.

B) Complex working conditions

(1) Composite drilling condition treatment sub-process:

When the current drilling state is a composite drilling state, if the variation amplitude of any one of the drilling weight, the rotating speed and the displacement of the shaft drilling between the depth of the current measuring point and the depth of the position of the drill bit is not smaller than the preset variation amplitude delta WOB _max,ΔRPM_max,ΔMFI_max, and the segmentation length of the dynamic trajectory control data is larger than the minimum segmentation SEG _min, segmenting the dynamic trajectory control data to form a plurality of data segments [ SEG ₁,SEG₂,…,SEG_n ], and taking an average value for each data segment; inputting the average value of the bit pressure, the average value of the rotating speed and the average value of the displacement of each data segment into a corresponding prediction model, and predicting the deflecting capability and the azimuth angle change rate of each data segment to obtain a deflecting capability predicted value and An azimuth angle change rate predicted value [ delta S ₁,ΔA₁],[ΔS₂,ΔA₂ ], … [ delta Sn, delta An ];

(2) The directional sliding drilling working condition treatment sub-flow:

When the current drilling state is a directional sliding drilling state, if the change amplitude of any one of drilling weight, displacement and tool face of the shaft drilling between the depth of the current measuring point and the depth of the position of the drill bit is not smaller than a preset change amplitude delta WOB _max,ΔMFI_max,ΔW_max, and the segment length of the dynamic track control data is larger than the minimum segment SEG _min, segmenting the dynamic track control data to form a plurality of data segments [ SEG ₁,SEG₂,…,SEG_n ], and averaging each data segment; inputting the average value of the bit pressure, the average value of the displacement and the average value of the tool face of each data segment into a corresponding prediction model, predicting the deflecting capability, the azimuth angle change rate and the tool face swing amplitude of each data segment, and obtaining the deflecting capability predicted value, the azimuth angle change rate predicted value and the tool face swing amplitude predicted value of each data segment [ΔS₁,ΔA₁,ΔW₁],[ΔS₂,ΔA₂,ΔW₂],…[ΔS_n,ΔA_n,ΔW_n].

S is the well inclination angle of the current bit, S _k is the well inclination angle corresponding to the kth data segment of the current measuring point, L _k is the segment length corresponding to the kth data segment, delta S _k is the new drilling-in well section deflecting ability predicted value corresponding to the kth data segment, and n is the number of data segments;

Wherein A is the azimuth of the current bit position, A _k is the azimuth corresponding to the kth data segment of the current measurement point, and DeltaA _k is the new drilling-in azimuth change rate predicted value corresponding to the kth data segment;

ΔW＝ΔW_n

Wherein DeltaW is the azimuth angle of the current bit position, deltaW _n is the tool face swing amplitude predicted value corresponding to the last data segment.

In step 106, dynamically solving by using a prediction model corresponding to the current drilling state with track state parameters of the current bit position as a reference, target track state parameters as an objective function and the maximum value of track state parameter change rate corresponding to the well section unit footage as a constraint condition, obtaining at least one group of recommended track control parameters and track state parameter change rates corresponding to the recommended track control parameters, and calculating the drilling distance reaching the target track state parameters;

(1) Composite drilling condition treatment sub-process:

According to the well inclination angle S and the azimuth angle A of the current bit position, taking a target well inclination angle and the azimuth angle as an objective function, taking the maximum well inclination angle change rate and the azimuth angle change rate of a unit footage well as constraint conditions, using a trained prediction model to carry out dynamic solution, obtaining 1 to a plurality of recommended weight on bit, rotating speed and displacement combinations, and corresponding deflecting capacity delta S and azimuth angle change rate delta A, and calculating the drilling distance reaching the target well inclination angle and the target azimuth angle.

(2) The directional sliding drilling working condition treatment sub-flow:

According to the well inclination angle S and the azimuth angle A of the current bit position, taking a target well inclination angle and an azimuth angle as an objective function, taking the maximum well inclination angle change rate and the azimuth angle change rate of a unit footage well and the maximum swing amplitude of a tool face as constraint conditions, carrying out dynamic solution by using a trained prediction model to obtain 1 to a plurality of recommended bit pressure and displacement combinations, and the corresponding deflecting capacity delta S, azimuth angle change rate delta A and tool face swing amplitude W, and calculating the drilling distance reaching the target well inclination angle and azimuth angle.

The solution process in the above process may use conventional parameter iteration, particle swarm iteration (ITERATIVE PARTICLE SWARM algorithm), and other methods to traverse the potential parameter space.

The recommended weight on bit and displacement parameters can be used as real-time reminding for operators, and further can be accessed into a drilling machine control system to realize real-time automatic closed-loop control.

In addition to the above prediction scheme, other schemes are also provided in the embodiments of the present invention.

(1) Prediction using classical formula

Modeling the drilling tool mechanics to obtain a drilling tool mechanics model;

And calculating at least one group of recommended track control parameters and corresponding track state parameter change rates according to the drilling tool mechanical model.

The modeling of drilling tool mechanics is solved, and the modeling comprises calculation and analysis of bending screw BHA guiding force and guiding capability, research of influence rules of key parameters on the BHA drill bit guiding force, inversion calculation of balanced lateral force of stratum, design of BHA structural parameters when different configurations of sliding guiding and composite drilling capability and the like.

The method has strong interpretability, but needs to collect a large amount of drilling tools and geological information, has large calculation amount, is not tightly combined with measured data, and can have larger in-out of calculation results and actual conditions.

The main difference is that this approach does not use artificial intelligence.

(2) Grouping calculation of measured data

Grouping the dynamic track control data according to equidistant depth intervals;

Analyzing the track control rules of different groups;

and predicting at least one group of recommended track control parameters and the corresponding track state parameter change rate according to the analyzed track control law.

The method is characterized in that the acquired data is kept at a fixed interval, and the data are grouped fixedly according to the value range, so that the method is a relative quantification method. In the case of irregular or discontinuous data spacing, additional special processing is required. In addition, the grouping value range of the rotating speed and the bit pressure is preset by a person, is not necessarily reasonable, and potential parameter relations can be coarsened, so that a continuous space is changed into a discrete space.

The main difference is that the patent does not need to group parameters and does not require equal spacing, and a computer is intelligent to find rules.

(3) Iterative solution mode based on artificial intelligent model

After the depth deviation is carried out on the collected dynamic track control data under the same depth, respectively training artificial intelligent models corresponding to different drilling states;

based on the measuring points and all previous dynamic track control data as a sample sequence, using an artificial intelligent model, and adopting the following steps to carry out iterative solution until the track state parameters of the current position of the drill bit are solved:

taking the average value of the measured dynamic track control data of the next predicted depth as input, and predicting the track state parameter of the next depth;

and (3) taking the predicted value of the track state parameter and the actually measured track state parameter as a set of new data, and putting the new data into the tail end of the sample sequence.

The above prediction process can be expressed as:

based on the real measurement points and all previous data as sequences, taking the measured weight on bit and the rotating speed (averaged and the processing logic is the same as above) of the next predicted depth as input, and predicting the well inclination angle, azimuth angle and tool face of the next depth;

then, the predicted well inclination angle, azimuth angle, tool face and actually measured bit pressure and rotating speed are taken as a group of new data, and are put into the tail end of a sample sequence;

and then taking the actually measured drilling pressure and rotating speed of the next predicted depth as input to predict the next group of well inclination angle, azimuth angle and tool face. And (3) iterating in a circulating way until the well inclination angle, the azimuth angle and the tool face of the current position of the drill bit are solved.

In general, there is a difference between the dynamic measurement of the well inclination angle and the static measurement (one column 28.5 m) of plus or minus 0.1-0.2 degrees. After each static measurement point, an analysis of the error interval and the reverse distribution error is required.

(1) Composite drilling condition treatment sub-process:

When the current drilling state is a composite drilling state, calculating the difference value between the dynamic well inclination angle and the static well inclination angle of the last static measurement point after each static measurement point, and taking the difference value as an error interval of the well inclination angle, and calculating the difference value between the dynamic azimuth angle and the static azimuth angle of the last static measurement point, and taking the difference value as an error interval of the azimuth angle; distributing the error interval and azimuth angle of the well inclination angle to each dynamic measurement point according to the depth of the last two static measurement points and the ratio of the depth between every two adjacent points for the covered dynamic measurement points, and correcting the value of the last dynamic measurement point to be the value of the corresponding static measurement point; retraining a corresponding prediction model after the value of the dynamic measurement point is corrected;

(2) The directional sliding drilling working condition treatment sub-flow:

When the current drilling state is a directional sliding drilling state, after each static measuring point, calculating the difference value between the dynamic well inclination angle and the static well inclination angle of the last static measuring point, and taking the difference value as an error interval of the well inclination angle, calculating the difference value between the dynamic azimuth angle and the static azimuth angle of the last static measuring point, and taking the difference value as an error interval of the azimuth angle, and calculating the difference value between the dynamic tool face and the static tool face of the last static measuring point, and taking the difference value as a tool face interval of the azimuth angle; distributing the error interval of the well oblique angle, the error interval of the azimuth angle and the error interval of the tool face to each dynamic measurement point according to the depth of the last two static measurement points and the ratio of the depth between every two adjacent points for the covered dynamic measurement points, and correcting the value of the last dynamic measurement point to be the value of the corresponding static measurement point; and after the value of the dynamic measurement point is corrected, retraining the corresponding prediction model.

A specific embodiment is given below to illustrate a specific application of the method and apparatus according to the embodiments of the present invention.

Fig. 2 is a diagram showing two effects of implementing full-well section drilling by using a screw drilling tool, and the parameters such as a well inclination angle, an azimuth angle and the like of the current bit position can be predicted in real time under the condition that the screw drilling tool is not used, so that the accurate control of the well track is realized, the drilling speed is improved, the damage of the bit is reduced, and the cost and the efficiency are improved.

Fig. 3 is a drawing of a three-open horizontal section drilling plate of a well in an embodiment of the invention, fig. 4 is a schematic diagram of the application of the method in the embodiment of the invention in the well, fig. 3 and fig. 4 are a case of realizing continuous track control of composite drilling by using a combination of a PDC bit, a double-screw drill and an MWD drill by utilizing dynamic track control data per meter, performing parameter grouping test and control prediction on a section to be drilled, creating a three-open horizontal section well depth 3962 m-52991 m of the well (combination of a GT55DRs PDC bit of DBS, a bent screw (phi 200 holds) +phi 205 holds+resistivity+nonmagnetic suspension and the like), a single drilling rule 1329m, and an average mechanical drilling rate of 12.7m/h, wherein the drilling rule is continuously 4m from the well depth 4017m-5291m, and the maximum length record of the bent screw drilling of the horizontal section in a block water-based slurry is created.

The green curve in fig. 4 represents the top drive rotational speed as a function of well depth, the blue dot represents the weight on bit applied as a function of well depth, the red curve represents the drilling time as a function of well depth, and the upper two dotted line combination curves in fig. 4 represent the well inclination angle and azimuth angle, respectively.

As can be seen from fig. 4, by continuously recording the data of the dynamic well inclination angle, the corresponding weight on bit, the rotating speed and the like measured along with the well depth in the composite drilling process, the data of grouping (the weight on bit is 50-100kN, the rotating speed is 40-80 rpm) are initially tested, and the control rule of the drilling tool combination under different data is mastered;

And carrying out error analysis on the recorded dynamic well inclination angle through the static measurement point well inclination angle data, wherein the error of the underground combined dynamic well inclination angle is +/-0.15.

Analyzing, studying and judging according to the dynamic well inclination angle change rules of the different dynamic track control data, and controlling and predicting the track of the section to be drilled; the subsequent well section compares and analyzes the predicted track state parameters and the actually measured track state parameters, and further continuously optimizes and corrects the drilling track control parameters (drilling pressure and rotating speed), so that the control and prediction accuracy is further improved;

and the method is circularly reciprocated, and the dynamic track control data is used for continuously analyzing, adjusting and continuously predicting, so that the good effect of 1274m continuous composite drilling of the well is realized.

Fig. 5 is a chart of dynamic well inclination angle, drilling pressure and rotating speed change of a certain well 4760m-4900m in an embodiment of the invention, and fig. 5 is a chart of control data of dynamic track per meter, PDC bit, screw double-support and MWD drilling tool combination, control prediction of a drilling section to be drilled is realized by using drilling pressure and rotating speed, and composite drilling continuous track control (local well section of the certain well 4760m-4900 m) is realized, wherein a light green area is marked as an inclination increasing well section, and a light orange area is marked as a inclination decreasing well section.

The green curve in fig. 5 represents the top drive rotational speed as a function of well depth, the blue dot represents the applied weight on bit as a function of well depth, the red curve represents the drilling time as a function of well depth, and the upper dotted line combination curve in fig. 5 represents the well inclination angle.

As can be seen from the graph of the change in the inclination angle of each meter of well:

6m well sections (middle and later periods of use of the drill bit and the centralizer) with the well depths of 4764m to 4770m, the bit weight of 120kN, the rotating speed of 60rpm, changing the well inclination angle from a declination trend to an inclination increasing trend and realizing inclination increase at an inclination increasing angle of 0.6 degrees/10 m;

a20 m well section with the well depth of 4812m to 4812m (abrasion of the drill bit and the centralizer occurs at the later use period), the weight on bit of 60kN-75kN and the rotating speed of 80rpm, the well inclination angle is changed from an inclination increasing trend to an inclination decreasing trend and is inclined at an inclination decreasing angle of 0.6 degrees/10 m;

A 14m well section with the well depth of 4886m to 4900m (more serious abrasion occurs at the later use period of the drill bit and the centralizer), the drilling pressure of 110-120kN and the rotating speed of 30rpm, the well inclination angle is changed from a declination trend to an inclination increasing trend, and the inclination increasing is still realized at an inclination increasing angle of 0.6 degrees/10 m;

The continuous analysis of data such as the dynamic well inclination angle, the drilling pressure and the rotating speed is integrated gradually, the drilling pressure and the rotating speed are regulated timely, and the continuous track prediction is carried out, so that the continuous control of the horizontal section track in a designed target box body (target oil layer) is realized.

The embodiment of the invention also provides a dynamic control device for the petroleum drilling track, the principle of which is similar to that of the dynamic control method for the petroleum drilling track, and the description is omitted here.

FIG. 6 is a schematic diagram of a dynamic control device for petroleum drilling trajectories according to an embodiment of the present invention, including:

The data acquisition module 601 is configured to acquire dynamic track control data at the same depth;

The drilling state determining module 602 is configured to determine a current drilling state according to the dynamic trajectory control data;

a data preprocessing module 603, configured to calculate a track state parameter change rate of the current drilling state according to the dynamic track control data;

The model training module 604 is configured to train by using a machine learning manner with a feature value in the dynamic trajectory control data as an input and a trajectory state parameter change rate of a current drilling state as an output, so as to obtain a prediction model corresponding to the current drilling state;

The drill bit well inclination angle prediction module 605 is configured to predict a change rate of a track state parameter of a new drilling section of the current drill bit by using a prediction model according to a dynamic track control data sequence and a track state parameter change rate sequence of the current section, and calculate a track state parameter of a position where the current drill bit is located according to a predicted change rate prediction value of the track state parameter obtained by prediction;

the track control parameter recommendation module 606 is configured to dynamically solve, using a prediction model corresponding to a current drilling state, with a track state parameter of a current bit position as a reference, a target track state parameter as an objective function, and a maximum value of a track state parameter change rate corresponding to a well section unit footage as a constraint condition, to obtain at least one set of recommended track control parameters and a track state parameter change rate corresponding to the at least one set of recommended track control parameters, and calculate a drilling distance reaching the target track state parameter;

The track control module 607 is used for controlling the drilling track according to the track control parameter and the drilling distance.

In one embodiment the dynamic trajectory control data includes drilling data acquired from a wellhead and data acquired from a wireless inclinometer;

the data acquisition module is used for:

In one embodiment, the drilling status determination module is to:

if the well depth changes, determining that the current drilling state exists;

And if the well depth is unchanged, determining that the drilling state is not currently in progress.

In one embodiment, the data preprocessing module is configured to:

If the difference between the depth of the current dynamic track control data and the depth of the last dynamic track control data corresponding to the current drilling state is larger than a first preset value, or the difference between the acquisition time of the current dynamic track control data and the acquisition time of the last dynamic track control data corresponding to the current drilling state is larger than a second preset value, or data in different drilling states exist between the current dynamic track control data and the last dynamic track control data corresponding to the current drilling state, it is determined that the situation difference between the current dynamic track control data and the last dynamic track control data corresponding to the current drilling state exceeds the expected change.

In one embodiment, the data preprocessing module is configured to:

the whipstock capacity was calculated using the following formula:

ΔS_i＝(S_i-S_i-1)/(D_i-D_i-1)

The azimuth angle change rate is calculated by adopting the following formula:

ΔA_i＝(A_i-A_i-1)/(D_i-D_i-1)

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

In one embodiment, the model training module is to:

In one embodiment, the drill bit well inclination prediction model is specifically used for:

When the current drilling state is a composite drilling state, if the variation amplitude of the drilling pressure, the rotation speed and the displacement of the shaft drilling between the depth of the current measuring point and the depth of the position of the drill bit is smaller than the preset variation amplitude, averaging the drilling pressure, the rotation speed and the displacement; inputting the average value of the bit pressure, the average value of the rotating speed and the average value of the displacement into a prediction model corresponding to the composite drilling state, and predicting the deflecting capability and the azimuth angle change rate of the newly drilled well section of the current drill bit to obtain a deflecting capability predicted value and an azimuth angle change rate predicted value;

When the current drilling state is a directional sliding drilling state, if the variation amplitude of the drilling weight, the displacement and the tool face of the shaft drilling between the depth of the current measuring point and the depth of the position of the drill bit is smaller than the preset variation amplitude, averaging the drilling weight, the displacement and the tool face; and inputting the average value of the bit pressure, the average value of the displacement and the average value of the tool face into a prediction model corresponding to the directional sliding drilling state, and predicting the deflecting capability, the azimuth angle change rate and the tool face swing amplitude of the newly drilled well section of the current drill bit to obtain a deflecting capability predicted value, an azimuth angle change rate predicted value and a tool face swing amplitude predicted value.

S＝S_i+(L×ΔS_n)

A＝A_i+(L×ΔA_n)

ΔW＝ΔW_n

According to the dynamic track control data sequence and the track state parameter change rate sequence of the current section, predicting the track state parameter change rate of the newly drilled well section of the current drill bit by using a prediction model, wherein the method comprises the following steps:

when the current drilling state is a composite drilling state, if the variation amplitude of any one of drilling weight, rotating speed and displacement of the shaft drilling between the depth of the current measuring point and the depth of the position of the drill bit is not smaller than the preset variation amplitude, and the segmentation length of the dynamic track control data is larger than the minimum segmentation, segmenting the dynamic track control data to form a plurality of data segments, and averaging each data segment; inputting the average value of the bit pressure, the average value of the rotating speed and the average value of the displacement of each data segment into a corresponding prediction model, and predicting the deflecting capability and the azimuth angle change rate of each data segment to obtain a deflecting capability predicted value and an azimuth angle change rate predicted value of each data segment;

When the current drilling state is a directional sliding drilling state, if the change amplitude of any one of drilling weight, displacement and tool face of the shaft drilling between the depth of the current measuring point and the depth of the position of the drill bit is not smaller than the preset change amplitude, and the segmentation length of the dynamic track control data is larger than the minimum segmentation, segmenting the dynamic track control data to form a plurality of data segments, and averaging each data segment; and inputting the average value of the bit pressure, the average value of the displacement and the average value of the tool face of each data segment into a corresponding prediction model, and predicting the deflecting capability, the azimuth angle change rate and the tool face swing amplitude of each data segment to obtain a deflecting capability predicted value, an azimuth angle change rate predicted value and a tool face swing amplitude predicted value of each data segment.

ΔW＝ΔW_n

In an embodiment, the apparatus further comprises a formula prediction module for:

modeling the drilling tool mechanics to obtain a drilling tool mechanics model;

In an embodiment, the apparatus further comprises a packet calculation module for:

Analyzing the track control rules of different groups;

In an embodiment, the apparatus further comprises an artificial intelligence model solving module for:

In an embodiment, the apparatus further comprises a correction module for:

In summary, the method and the device provided by the embodiment of the invention have the following beneficial effects:

(1) The horizontal well section or the highly-inclined stable well section reduces the use of the rotary guide or completely eliminates the rotary guide, thereby saving the cost

Under the condition of not using a rotary guide, not only is the equipment cost saved, but also higher weight on bit and rotating speed parameters can be used, and the drilling speed, especially the stratum with poor drillability, is greatly improved. If the drilling tool is used in a horizontal section, the drilling period of the horizontal well can be further shortened, and the cost is greatly saved.

(2) Without additional data acquisition means

The bent screw is used as a relatively simple and low-cost device, and the existing acquired data is utilized without additional data acquisition. No special grouping test is needed, the rule is automatically acquired, and no interference is caused to construction. And the continuous measurement and transmission data of the measurement while drilling instrument are utilized, no extra special drilling stopping measurement is needed, and the production time efficiency is high. By adopting the method, the number of times and time of static well inclination angle measurement can be reduced, and the drilling time efficiency can be improved.

(3) The composite time rate is improved, and the sliding drilling duty ratio is reduced

Advanced research and judgment are carried out, a basis is provided for advanced determination of follow-up measures, drilling parameters are adjusted more timely, well inclination angle, azimuth angle and tool face are controlled, and track control is finer. The directional sliding drilling process is avoided or greatly reduced, the single-pass drilling length and speed are improved, and the difficulty that the curved screw drilling tool assembly is difficult to track control in a long horizontal section (especially water-based mud) is basically solved. When in sliding drilling, the stability of the tool surface is maintained as much as possible, and the given sliding drilling parameters are more accurate. The drill-in length and the rate of penetration under the same conditions for a typical block are expected to be improved by about 10%.

(4) Reducing downhole complications and bit damage and corresponding losses

Through reducing directional sliding drilling, the occurrence of underground complex conditions such as sticking and sticking is effectively avoided, the damage to a drill bit is reduced, the service life of the drilling bit is prolonged, the drilling ruler is improved, and the use cost of the drill bit is saved. Further, additional tripping operations caused by damage to the drill bit and advanced aging are reduced, and the construction time is further shortened.

(5) Keeping the track of the well smooth

The well formed by the composite drilling is smoother, and the well completion electrical measurement and casing running operation are smoother. The fine and efficient track control is beneficial to prolonging the service life of the drill bit, improving the machine speed and the drilling rule and reducing the drilling cost.

(6) Reducing human error

Full-automatic learning, prediction and control, without human interaction intervention, can reduce the requirements for directional engineers and reduce errors due to human negligence.

An embodiment of the present invention further provides a computer device, and fig. 7 is a schematic diagram of the computer device in the embodiment of the present invention, where the computer device 700 includes a memory 710, a processor 720, and a computer program 730 stored in the memory 710 and capable of running on the processor 720, and the processor 720 implements the above-mentioned method for dynamically controlling the petroleum drilling track when executing the computer program 730.

The embodiment of the invention also provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program realizes the dynamic control method of the petroleum drilling track when being executed by a processor.

The embodiment of the invention also provides a computer program product, which comprises a computer program, wherein the computer program realizes the dynamic control method of the petroleum drilling track when being executed by a processor.

It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

The foregoing description of the embodiments has been provided for the purpose of illustrating the general principles of the invention, and is not meant to limit the scope of the invention, but to limit the invention to the particular embodiments, and any modifications, equivalents, improvements, etc. that fall within the spirit and principles of the invention are intended to be included within the scope of the invention.

Claims

1. A method for dynamically controlling petroleum drilling trajectories, comprising:

collecting dynamic track control data at the same depth;

2. The method of claim 1, wherein the dynamic trajectory control data comprises drilling data acquired from a wellhead and data acquired from a wireless inclinometer;

The method further comprises the steps of:

3. The method of claim 2, wherein determining the current drilling status based on the dynamic trajectory control data comprises:

if the rotating speed is acquired, determining the drilling state as a composite drilling state; if the rotating speed is not acquired, determining that the drilling state is a directional sliding drilling state;

4. The method of claim 2, wherein calculating a rate of change of a trajectory state parameter of a current drilling state based on the dynamic trajectory control data comprises:

5. The method of claim 2, wherein determining a difference in conditions of the current dynamic trajectory control data and the last dynamic trajectory control data corresponding to the current drilling state comprises:

6. The method of claim 2, wherein calculating a rate of change of a trajectory state parameter of a current drilling state based on the dynamic trajectory control data comprises:

the whipstock capacity was calculated using the following formula:

ΔS_i＝(S_i-S_i-1)/(D_i-D_i-1)

The azimuth angle change rate is calculated by adopting the following formula:

ΔA_i＝(A_i-A_i-1)/(D_i-D_i-1)

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

7. The method of claim 2, wherein training using machine learning with the characteristic value in the dynamic trajectory control data as input and the rate of change of the trajectory state parameter of the current drilling state as output to obtain the predictive model corresponding to the current drilling state comprises:

8. The method as recited in claim 7, further comprising:

9. The method of claim 2, wherein predicting the rate of change of the trajectory state parameter of the newly drilled section of the current drill bit using the predictive model based on the sequence of dynamic trajectory control data and the sequence of rate of change of the trajectory state parameter of the current section comprises:

When the current drilling state is a directional sliding drilling state, if the change amplitude of the drilling weight, the displacement and the tool face of the shaft drilling between the depth of the current measuring point and the depth of the position of the drill bit is smaller than the preset change amplitude, averaging the drilling weight, the displacement and the tool face; and inputting the average value of the bit pressure, the average value of the displacement and the average value of the tool face into a prediction model corresponding to the directional sliding drilling state, and predicting the deflecting capability, the azimuth angle change rate and the tool face swing amplitude of the newly drilled well section of the current drill bit to obtain a deflecting capability predicted value, an azimuth angle change rate predicted value and a tool face swing amplitude predicted value.

10. The method of claim 9, wherein calculating the trajectory state parameter for the current bit location based on the predicted rate of change of the predicted trajectory state parameter comprises:

S＝S_i+(L×ΔS_n)

A＝A_i+(L×ΔA_n)

ΔW＝ΔW_n

11. The method of claim 2, wherein predicting the rate of change of the trajectory state parameter of the newly drilled section of the current drill bit using the predictive model based on the sequence of dynamic trajectory control data and the sequence of rate of change of the trajectory state parameter of the current section comprises:

12. The method of claim 11, wherein calculating the trajectory state parameter for the current bit location based on the predicted rate of change of the trajectory state parameter, comprises:

ΔW＝ΔW_n

13. The method as recited in claim 1, further comprising:

modeling the drilling tool mechanics to obtain a drilling tool mechanics model;

14. The method as recited in claim 1, further comprising:

Analyzing the track control rules of different groups;

15. The method as recited in claim 1, further comprising:

16. The method as recited in claim 1, further comprising:

17. A petroleum drilling trajectory dynamic control device, comprising:

18. The apparatus of claim 17, further comprising a formula prediction module configured to:

modeling the drilling tool mechanics to obtain a drilling tool mechanics model;

19. The apparatus of claim 17, further comprising a packet computation module to:

Analyzing the track control rules of different groups;

20. The apparatus of claim 17, further comprising an artificial intelligence model solving module for:

21. The apparatus of claim 17, further comprising a correction module to:

When the current drilling state is a composite drilling state, calculating the difference value between the dynamic well inclination angle and the static well inclination angle of the last static measuring point after each static measuring point, and taking the difference value as an error interval of the well inclination angle; distributing the error interval and azimuth angle of the well inclination angle to each dynamic measurement point according to the depth of the last two static measurement points and the ratio of the depth between every two adjacent points for the covered dynamic measurement points, and correcting the value of the last dynamic measurement point to be the value of the corresponding static measurement point; retraining a corresponding prediction model after the value of the dynamic measurement point is corrected;

22. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method of any of claims 1 to 16 when executing the computer program.

23. A computer readable storage medium, characterized in that the computer readable storage medium stores a computer program which, when executed by a processor, implements the method of any of claims 1 to 16.

24. A computer program product, characterized in that the computer program product comprises a computer program which, when executed by a processor, implements the method of any of claims 1 to 16.