CN111810112B - Vertical drilling deviation rectifying control method based on particle filtering and model prediction control - Google Patents
Vertical drilling deviation rectifying control method based on particle filtering and model prediction control Download PDFInfo
- Publication number
- CN111810112B CN111810112B CN202010558687.6A CN202010558687A CN111810112B CN 111810112 B CN111810112 B CN 111810112B CN 202010558687 A CN202010558687 A CN 202010558687A CN 111810112 B CN111810112 B CN 111810112B
- Authority
- CN
- China
- Prior art keywords
- drilling
- track
- actual
- alpha
- control
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH OR ROCK DRILLING; MINING
- E21B—EARTH OR ROCK DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
- E21B44/00—Automatic control systems specially adapted for drilling operations, i.e. self-operating systems which function to carry out or modify a drilling operation without intervention of a human operator, e.g. computer-controlled drilling systems; Systems specially adapted for monitoring a plurality of drilling variables or conditions
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH OR ROCK DRILLING; MINING
- E21B—EARTH OR ROCK DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
- E21B7/00—Special methods or apparatus for drilling
- E21B7/04—Directional drilling
- E21B7/10—Correction of deflected boreholes
Landscapes
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Geology (AREA)
- Mining & Mineral Resources (AREA)
- Physics & Mathematics (AREA)
- Environmental & Geological Engineering (AREA)
- Fluid Mechanics (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geochemistry & Mineralogy (AREA)
- Feedback Control In General (AREA)
- Excavating Of Shafts Or Tunnels (AREA)
Abstract
本发明提供一种基于粒子滤波和模型预测控制的垂直钻进纠偏控制方法，包括以下：建立垂直钻进过程中的钻进轨迹延伸模型；构建改进型模型预测控制器；引入粒子滤波器对实际钻进轨迹测量参数进行滤波，得到降噪后的实钻轨迹参数；将降噪后的实钻轨迹参数输入至改进型模型预测控制器，形成垂钻闭环控制。本发明提供的有益效果是：在纠偏控制中建立了一个粒子滤波器，能够有效降低测量噪声对垂直钻进纠偏控制带来的影响，提高控制精度；在模型预测控制器中引入软约束和可变优化权重，增加模型预测控制器的环境适应性。
The present invention provides a vertical drilling rectification control method based on particle filtering and model prediction control, comprising the following steps: establishing a drilling trajectory extension model during vertical drilling; building an improved model prediction controller; The drilling trajectory measurement parameters are filtered to obtain the actual drilling trajectory parameters after noise reduction; the actual drilling trajectory parameters after noise reduction are input to the improved model prediction controller to form the vertical drilling closed-loop control. The beneficial effects provided by the invention are as follows: a particle filter is established in the deviation correction control, which can effectively reduce the influence of measurement noise on the deviation correction control of vertical drilling, and improve the control accuracy; Change the optimization weight to increase the environmental adaptability of the model predictive controller.
Description
Technical Field
The invention relates to the field of drilling control, in particular to a vertical drilling deviation rectifying control method based on particle filtering and model predictive control.
Background
Vertical drilling techniques are of great importance for deep geological drilling, where the goal of vertical drilling is to drill from the wellhead down to the target formation along the plumb line. However, due to factors such as formation dip angle, anisotropy, hard-soft lithology alternation, stress and bending deformation of a lower drilling tool and the like, a drilling track is easy to deflect in engineering. When the well deviation is too great, the site construction and the original design scheme are greatly deviated, thereby reducing the resource recovery rate. Meanwhile, too large well inclination angle and azimuth angle also easily cause the complex problems of difficult tripping, deteriorated working condition of the drill rod, stuck drill sticking, key slot drill sticking and the like, and the safety of the drilling process is seriously influenced. Therefore, the deviation rectification control of the drilling track is very important.
The important means of deviation correction control in the vertical drilling process is a drilling track control method, namely the drilling track is kept to advance along a wellhead plumb line through a reasonable control means. Although the deviation rectifying control method for vertical drilling based on the traditional process and the research in the existing literature can solve the problem of trajectory deviation to a certain extent, few research results discuss the problem that the deviation rectifying control precision is affected by measurement noise in the vertical drilling process. The measurement noise may lead to a reduction in control accuracy on the one hand and increase the difficulty of the controller in handling state constraints on the other hand.
The filtering is a main method for solving the measurement noise, wherein the extended Kalman filter is one of the most commonly used filters, and is widely applied to various industrial fields. But for the noise filtering problem of the non-Gaussian nonlinear system, the extended Kalman filter has limited precision. The particle filtering theory is a filtering theory under a Bayesian framework, and can well complete filtering tasks under various environments based on Monte Carlo sampling theorem. The mechanism of the vertical drilling process is complex, and the noise of the process has non-Gaussian characteristic, so that the design of the particle filter suitable for the vertical drilling deviation rectifying process is an effective means for processing the noise of the vertical drilling measurement.
Aiming at the constraint problem, the model prediction controller can be competent for various actual industrial control tasks because of the capability of displaying and processing the constraint, and has good application effect on the vertical drilling deviation rectifying process. However, due to the influence of measurement noise, the inclination angle is easily out of limit in the deviation correction process, and when the inclination angle exceeds a certain range, the rolling optimization problem in the model prediction controller is lack of a feasible solution, so that the calculation of the controller is wrong. Further research is still needed to address this problem.
Disclosure of Invention
In view of the above, the invention provides a vertical drilling deviation correction control method based on particle filtering and model prediction control, which combines a particle filter with improved model prediction control to reduce the negative influence of measurement noise on deviation correction control. And analyzing deviation correction control requirements and process limitations in the vertical drilling process, and researching the size and distribution characteristics of measurement noise and process noise in the vertical drilling process so as to give mathematical description of the deviation correction control problem. Then, considering the problem that the deviation correction control precision is influenced by measurement noise in the vertical drilling process, a particle filter is introduced to improve the control precision. And finally, designing a model prediction controller to realize deviation correction and deviation correction of the drilling track, and improving model prediction control by introducing soft constraint and variable optimization weight so as to reduce adverse effects of measurement noise on the controller.
The invention provides a vertical drilling deviation rectifying control method based on particle filtering and model predictive control, which comprises the following steps of:
s101: establishing a three-dimensional stratum coordinate system, wherein the Z axis is in the direction of a plumb line, the X axis points to the east direction, the Y axis points to the north direction, and establishing a drilling track extension model in the vertical drilling process according to the deviation rectifying process and the noise distribution in the vertical drilling process;
s102: linearizing and discretizing the drilling track extension model, introducing a soft constraint condition and a variable optimization weight matrix based on the noise distribution characteristic in the vertical drilling process, constructing an improved model prediction controller, giving a drilling reference track, and inputting the reference track to the improved model prediction controller; starting a vertical drilling process, and measuring and obtaining actual drilling track parameters containing noise; the reference tracks are a reference well inclination angle, a reference azimuth angle and a reference horizontal displacement in the drilling process;
s103: introducing a particle filter, and filtering the actual drilling track parameter containing the noise to obtain the actual drilling track parameter with the noise reduced;
s104: and converting the actual drilling track parameters subjected to noise reduction into feedback input signals of the improved model prediction controller by combining a minimum curvature method and a drilling track extension model, and inputting the feedback input signals to the improved model prediction controller to form vertical drilling closed-loop control.
Further, in step S101, the drilling trajectory extension model is represented by equation (1):
in the formulas (1), (2) and (3), alpha is the inclination angle of the drilling track, beta is the azimuth angle of the drilling track, and alphaxIs the projection component, alpha, of the angle of inclination of the drilling path in the XOZ planeyThe projection component of the well drilling track inclination angle on the YOZ plane is shown;
the drilling speed is used;
for the horizontal X-direction component S of the drilling pathxA derivative of (a);
for horizontal Y-direction component S of drilling trackyA derivative of (a);
is alphaxA derivative of (a);
is alphayA derivative of (a); omegaSRThe rate of guidance for the drilling system;
orienting the drilling system magnetic tool at an angle; r is the build-up rate of the drilling system; mu.sxIs the component of the process noise in the X direction during the drilling process; mu.syIs the component of the process noise in the Y direction during drilling.
Further, in step S102, the drilling trajectory extension model is linearized and discretized, specifically as shown in equation (4):
equation (4) is a linear discretization state space equation of the drilling track extension model, wherein
And
filter estimates of the actual lateral displacement of the tool of the drilling system in the X-axis and the Y-axis with respect to said reference trajectory at time k,
and
filter estimates, ω, of the projections on the XOZ and YOZ planes of the skew angle of the actual drilling trajectory relative to the reference trajectory at time k, respectivelyex(k) And omegaey(k) T is the sampling period for the control increment on both planes relative to the reference steering rate.
Further, in step S102, a soft constraint condition and a variable optimization weight matrix are introduced to construct an improved model predictive controller, specifically: the prediction equation of the improved model prediction controller is as follows:
Y(k)＝Ξkx(k|k)+ΘkW(k) (5)
in the formula (5), xikAnd ΘkAre parameter matrices of a prediction equation; in formula (5), each matrix is represented by formula (6):
in the formula (6), p is a preset prediction step length, and c is a preset control step length; sex(k +1| k) and Sey(k +1| k) are actual lateral displacement deviations of a drilling tool of the drilling system relative to the reference track in the X axis and the Y axis at k +1 time predicted at k time respectively; alpha is alphaex(k +1| k) and αey(k +1| k) are respectively the deviations of the projection values of the well inclination angles of the actual drilling track at the k +1 moment predicted at the k moment relative to the reference track on the XOZ and YOZ planes;
introducing a soft constraint condition and a variable optimization weight matrix to obtain an optimization constraint condition of the improved model predictive controller, which is specifically shown as a formula (7):
in the formula (7), Q and R are weight matrixes of the state quantity Y (k) and the controlled quantity U (k) of the prediction equation respectively;
q is represented by formula (8):
wherein q issxIs a component S in the X directionxWeight of (a), qsyIs a Y-direction component SyWeight of aQ，bQ， cQAnd dQIs an angle weight factor; alpha is alpharx(m) and alphary(m) is the projection value of the well skew angle of the reference track at the moment m on the XOZ and YOZ planes;
and
the deviation of the projection value of the skew angle of the actual drilling track relative to the reference track at the moment m on the XOZ and YOZ planes is obtained; omegarx(m) and ωry(m) is a reference control quantity at the moment m; omegaex(m) and ωey(m) is the deviation of the actual controlled variable and the reference controlled variable at the moment m;
is an estimated value of the well inclination angle at the moment k; alpha is alphamaxIs a preset maximum soft constraint for the well inclination angle.
Further, in step S103, the particle filter is specifically a basic particle filter.
Further, in step S104, the feedback input signal specifically includes: estimation of projection values of inclination angles of actual drilling paths on XOZ and YOZ planes
Estimation of the components of the horizontal displacement of the actual drilling trajectory in the X-direction and Y-direction
In step S104, after the feedback input signal is input into the improved model predictive controller, the output signal is the facing angle of the magnetic tool
And the guidance ratio omegaSRThe final actual controlled variable is specifically as shown in formula (9):
the beneficial effects provided by the invention are as follows: a particle filter is established in the deviation correction control, so that the influence of measurement noise on the deviation correction control of the vertical drilling can be effectively reduced, and the control precision is improved; soft constraints and variable optimization weights are introduced into the model predictive controller, and the environmental adaptability of the model predictive controller is improved.
Drawings
FIG. 1 is a schematic structural diagram of a vertical drilling deviation rectification control method based on particle filtering and model predictive control according to the present invention;
FIG. 2 is a block diagram of a pilot drill based vertical drilling system of the present invention
FIG. 3 is a schematic diagram of the track extension of the present invention
FIG. 4 is a schematic view of the build rate r of the present invention;
FIG. 5 is a graph showing the single-pass filtering effect of the particle filter according to the present invention;
FIG. 6 is a schematic diagram of the variable optimization weights of the present invention;
FIG. 7 is a diagram of the simulation effect of deviation rectification control according to the present invention;
FIG. 8 is a diagram showing the effect of the deviation rectifying control simulation control quantity of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be further described with reference to the accompanying drawings.
Referring to fig. 1, an embodiment of the present invention provides a vertical drilling deviation rectification control method based on particle filtering and model prediction control, including the following steps:
s101: establishing a three-dimensional stratum coordinate system, wherein the Z axis is in the direction of a plumb line, the X axis points to the east direction, the Y axis points to the north direction, and establishing a drilling track extension model in the vertical drilling process according to the deviation rectifying process and the noise distribution in the vertical drilling process;
s102: linearizing and discretizing the drilling track extension model, introducing a soft constraint condition and a variable optimization weight matrix based on the noise distribution characteristic in the vertical drilling process, constructing an improved model prediction controller, giving a drilling reference track, and inputting the reference track to the improved model prediction controller; starting a vertical drilling process, and measuring and obtaining actual drilling track parameters containing noise; the reference tracks are a reference well inclination angle, a reference azimuth angle and a reference horizontal displacement in the drilling process;
s103: introducing a particle filter, and filtering the actual drilling track parameter containing the noise to obtain the actual drilling track parameter with the noise reduced;
s104: and converting the actual drilling track parameters subjected to noise reduction into feedback input signals of the improved model prediction controller by combining a minimum curvature method and a drilling track extension model, and inputting the feedback input signals to the improved model prediction controller to form vertical drilling closed-loop control.
For convenience in explaining the symbols in the following formulas, the present invention unifies the definitions as follows:
in the variables, superscript with sharp brackets is an estimated value output by the particle filter;
the superscripts in the variables without tip brackets are all actual values;
in the variables, the variables with r subscripts are reference values of given reference tracks;
in the variables, the variables with e subscripts are all deviation values of actual values relative to reference quantities;
step S101, specifically: the vertical drilling system used in actual geological drilling is shown in fig. 2 and mainly comprises a screw-guided drilling tool, a driller room, a drill rod, a drill bit, an inclinometer and a turntable. The whole vertical drilling deviation correction control process comprises the steps of measuring track parameters by an inclinometer, calculating a next control instruction, and adjusting the rotation state of a turntable and an underground screw drill tool so as to perform directional deviation correction. It is worth noting that when the rotary table and the screw drill rotate simultaneously, the system is in a composite drilling state, and the system is not inclined; only the rotating disc is stopped rotating, the system is in a directional deflecting state, and the system provides a certain deflecting rate. By adjusting the ratio of the operating times of the two states, the system can provide different build rates.
According to the above analysis, a trajectory extension model is given as a control object model starting from the viewpoint of the drill kinematics. And establishing a three-dimensional stratum coordinate system, wherein the Z axis is in the plumb line direction, the downward direction is the positive direction, the X axis points to the east direction, and the Y axis points to the north direction. With reference to fig. 3, the trajectory extension model is shown in equations (1), (2) and (3):
the overall goal of deviation control is to simultaneously adjust the inclination angle αx,αyAnd horizontal displacement Sx,SyThe drilling trajectory is returned to the vertical line, that is, the state quantity is zero. The input of the system is a reference drilling track, the system is a plumb line in vertical drilling, and the adjustable parameter of the system is a guidance ratio omegaSRAngle with magnetic tool face
Determines the direction of drilling, omegaSRThe proportion of the system in a directional deflecting state in a control period to the drilling time is indicated.
Due to the harsh environment in the well, the measurement inevitably has some noise. The vertical drilling process needs to keep a lower well inclination angle, so the accuracy of deviation correction control is very sensitive to measurement noise. The inclinometer mainly uses an acceleration sensor and a fluxgate sensor for track measurement, and as a conventional sensor, measurement noise of the inclinometer mainly comes from electronic thermal noise, the noise distribution generally follows normal distribution, and the maximum measurement noise of a well angle can even reach 1.5 degrees along with the increase of the well depth. For process noise, as shown in FIG. 4, based on actual drilling process data, the process noise causes the system maximum build rate r to float between 1.4-9.2/30 m, which approximately follows a gamma (3,2) distribution that is scaled in scale and amplitude.
Because of geological drilling measurement limitation, fixed point measurement technology is often adopted in engineering, namely drilling is stopped after a certain distance is drilled, and track parameters are measured, wherein the drilling distance is generally the length of one drill rod. Meanwhile, in the vertical drilling process, in order to ensure the track quality, the drilling well inclination angle is required to be kept smaller than alphamax. Once a exceeds amaxThe drilling system should preferentially reduce the angle of the well in order to ensure the quality of the drilling trajectory. Furthermore, the steering tool whiplash capability r is limited.
Aiming at the problem of measurement noise of a non-Gaussian nonlinear process in a drilling process, the invention designs a particle filter for the measurement noise. The particle filter is based on the Monte Carlo sampling theorem, and can better solve the filtering problem under the non-Gaussian non-linear condition.
Step S103 specifically includes: the particle filters used more widely mainly include basic particle filters, particle filters based on optimization algorithms, particle filters combined with other filters. The basic particle filter is fast and easy to realize, and comprehensively estimates the actual state of the system based on the prior probability distribution and the measured value, and the precision of the basic particle filter depends on the prior knowledge or the measured value precision. Particle filtering based on particle optimization algorithms primarily adjusts the particle distribution in real time based on measurements, in order to expect that the particles can approach the late distribution region, with accuracy dependent on the measurement accuracy. Particle filters combined with other filters use a combination of two filtering algorithms to alter the particle distribution through the other filter in order to expect the particles to be able to approach the region of the postpropagation distribution with a precision that depends on the filter precision.
However, since the vertical drilling process has a small angle of inclination and the measurement noise may be as high as 1.5 °, manual experience is more important than the measurement value for the filtering problem of the vertical drilling process. This makes the basic particle filter based on a priori knowledge more advantageous. Therefore, the invention adopts a basic particle filter algorithm to design a particle filter in the vertical drilling process, and the pseudo code of the algorithm is shown in table 1.
TABLE 1 vertical drilling Process particle Filter
To verify the effectiveness of the particle filter, a numerical simulation is designed based on a vertical drilling trajectory extension model. Setting the measurement noise to obey a normal distribution vα,xk,υα,ykN (0,0.49), the maximum measurement noise is 1.5. The process noise follows a gamma distribution (10 x μ)x,k+6)～Γ(3,2)，(10*μy,k+6) to Γ (3,2), i.e. with a maximum error of 4 °/30 m. Comparing the Particle Filter (PF) with the Extended Kalman Filter (EKF), the particle filter based on Particle Swarm Optimization (PSOPF), and the Extended Kalman Particle Filter (EKPF), the single filtering result is shown in fig. 5, and the results of the 100 monte carlo experiments are shown in table 2:
TABLE 2100 Monte Carlo Filter results
From the single filtering result, errors of the particle filter, the particle filter based on particle swarm optimization and the extended Kalman particle filter are far smaller than measurement noise, and the extended Kalman filter diverges because process noise is gamma distribution. As can be seen from the table I, the particle filter based on particle swarm optimization and the extended Kalman particle filter lose the priori knowledge, and the measured values with larger errors are adopted to adjust the particle distribution, so that the filtering error is increased. The basic particle filtering performance is good, and the filtering precision can be further improved through other priori knowledge in drilling, such as engineering logging data, adjacent well data and the like.
To make the controller more robust and capable of displaying process control limits, the deskew control selects a model predictive controller.
Step S102 specifically includes: first, a prediction equation of a model predictive controller needs to be designed. Based on the vertical drilling track extension model established in the step 1, linearization and discretization are required to be carried out on the model so as to simplify the design difficulty of the controller. Oblique angle alpha of wellxAnd alphayDisplacement S of drilling tool in X-axis and Y-axisxAnd SxAs the state quantity, the guidance ratio ωxAnd omegayTo control the quantity (guidance ratio omega)xAnd omegayCan be determined by the guidance ratio omegaSRAngle with magnetic tool face
Obtained by calculation, as shown in (1). In order to guarantee the control precision, the invention outputs the filter
The feedback signal is taken as the feedback signal of the model prediction controller, and the objective factor of smaller well inclination angle in the vertical drilling process is considered
And is
Then a linear model of the trajectory extension of the vertical drilling process can be obtained:
for the reason that the measurement while drilling system in engineering does not dynamically measure the track parameters, but stops drilling once every certain distance, generally the length of a drill rod, the model cannot be directly used for controller design, discretizes the model, replaces the derivative with the difference quotient, and after arrangement, the linear discrete state space equation is as follows:
wherein
And
the lateral displacements of the drilling tool in the X-axis and the Y-axis respectively with respect to the reference trajectory,
and
the projections of the inclination angle of the real drilling track relative to the reference track on the XOZ and YOZ planes respectively, v is the drilling speed, omegaex(k) And omegaey(k) Is the control increment on both planes relative to the reference guidance ratio. It is worth mentioning that the reference point on the reference track is consistent with the vertical depth of the current drilling tool.
Based on the discrete state space equation, assuming p as the prediction step length and c as the control step length, the model prediction controller prediction equation can be written as:
Y(k)＝Ξkx(k|k)+ΘkW(k) (6)
wherein the meaning of each matrix is:
x(k|k)＝[Sex(k|k) aex(k|k) Sey(k|k) aey(k|k)]T
based on the constraint analysis in step 1, in the vertical drilling process, in order to ensure the drilling track quality, the inclination angle of the well is generally kept less than alpha as much as possiblemaxOnce the well deviation exceeds αmaxThe drilling system should preferentially reduce the angle of the well in order to ensure the quality of the drilling trajectory. Furthermore, the steering tool whiplash capability r is limited. Therefore, the deviation correction control system is aimed at by combining constraint conditionsWe choose the following optimization problem:
in the above equation, Q and R are weight matrices of the state and the controlled variable, respectively, a larger Q value can ensure that the tracking error is smaller, but may cause oscillation, and a larger R value can ensure that the controlled variable changes more smoothly. In the course of trajectory deviation correction, the well inclination angle will be close to alphamaxTo complete the rectification more quickly, however. The inclination angle is not free from fluctuation due to the influence of measurement noise, and the extreme condition can cause the inclination angle to greatly exceed alphamaxWhen the angle of the well
Time (omega)maxMaximum build rate that can be provided for one cycle of the actuator), the above optimization problem has no feasible solution, resulting in model predictive controller calculation errors. To solve the problem, the invention introduces a soft constraint and a variable optimization weight, wherein the soft constraint ensures that the model predictive controller always has a feasible solution, and the variable optimization weight based on the sigmoid function is combined to exceed alpha at the inclination anglemaxThe system is made to reduce well deviation preferentially to ensure drilling track quality. The variable optimization weights are given by the following equation and shown in fig. 6:
combining soft constraints and variable optimization weights, the sort optimization problem is shown as (9):
wherein Q is
Step S104 specifically includes: because the optimized control quantity output of the controller is a control increment relative to the reference control quantity, the actual control quantity needs to be increased by the reference control quantity on the basis of the control increment; meanwhile, according to the model predictive control law, the actual control increment should take the first value u (k) of the optimized and calculated U (k) sequence, and finally the actual control quantity is obtained as follows:
finally, numerical simulation is designed to verify the deviation rectification control method. According to field data of a vertical well, at 600m, the offset in the x direction is 8.82m, the offset in the y direction is 1.51m, the well deviation is 1.5 degrees, and the azimuth is 35.9 degrees. In order to observe the effectiveness of the soft constraint and the variable optimization weight in the method of the invention more clearly, the initial skew angle is properly changed to 5.83 degrees, and the azimuth angle is 56.6 degrees. According to the foregoing process analysis, it is assumed that the measurement noise follows Gaussian vkN (0,0.49), which means that the maximum measurement noise is around 1.4 °, the process noise follows a gamma distribution (12 x μ °)k+6) to Γ (3,2), which means that the maximum process noise is around 3.4 °/30 m. The model prediction parameters are shown in table 3, and the simulation results are shown in fig. 7 and 8.
TABLE 3 simulation parameters
The control method proposed by the present invention is compared with the basic model predictive control method and the model predictive control method with only particle filters, respectively. Compared with a basic model prediction control method, the basic model prediction control method improves the fluctuation trend of the track to a certain extent, but is difficult to stabilize the inclination angle under the condition of larger measurement noise, so that the final track still has larger horizontal position deviation. Compared with the model prediction control method only provided with the particle filter, the model prediction control method only provided with the particle filter has control calculation errors at 600m and 643m respectively, and the obtained guidance ratio exceeds 100%, has no practical significance in vertical drilling. Meanwhile, as can be seen from the inclination angle value between 720m and 820m, the inclination angle in the model predictive control method only with the particle filter is easier to exceed alpha due to the lack of variable optimization weightmaxTherefore, the drilling trajectory quality is lower than that of the method of the present invention.
The control method proposed by the present invention is compared with the basic model predictive control method and the model predictive control method with only particle filters, respectively. Compared with a basic model prediction control method, the basic model prediction control method improves the fluctuation trend of the track to a certain extent, but is difficult to stabilize the inclination angle under the condition of larger measurement noise, so that the final track still has larger horizontal position deviation. Compared with the model prediction control method only provided with the particle filter, the model prediction control method only provided with the particle filter has control calculation errors at 600m and 643m respectively, the obtained guiding rate exceeds 100%, and the method has no practical significance in vertical drilling. Meanwhile, as can be seen from the inclination angle value between 720m and 820m, the inclination angle in the model predictive control method only with the particle filter is easier to exceed alpha due to the lack of variable optimization weightmaxTherefore, the drilling trajectory quality is lower than that of the method of the present invention.
The beneficial effects of the implementation of the invention are as follows: a particle filter is established in the deviation correction control, so that the influence of measurement noise on the deviation correction control of the vertical drilling can be effectively reduced, and the control precision is improved; soft constraints and variable optimization weights are introduced into the model predictive controller, and the environmental adaptability of the model predictive controller is improved.
The features of the above-described embodiments and embodiments of the invention may be combined with each other without conflict.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (5)
1. A vertical drilling deviation rectifying control method based on particle filtering and model predictive control is characterized in that: the method specifically comprises the following steps:
s101: establishing a three-dimensional stratum coordinate system, wherein the Z axis is in the direction of a plumb line, the X axis points to the east direction, the Y axis points to the north direction, and establishing a drilling track extension model in the vertical drilling process according to the deviation rectifying process and the noise distribution in the vertical drilling process;
s102: linearizing and discretizing the drilling track extension model, introducing a soft constraint condition and a variable optimization weight matrix based on the noise distribution characteristic in the vertical drilling process, constructing an improved model prediction controller, giving a drilling reference track, and inputting the reference track to the improved model prediction controller; starting a vertical drilling process, and measuring and obtaining actual drilling track parameters containing noise; the reference tracks are a reference well inclination angle, a reference azimuth angle and a reference horizontal displacement in the drilling process;
s103: introducing a particle filter, and filtering the actual drilling track parameter containing the noise to obtain the actual drilling track parameter with the noise reduced;
s104: converting the actual drilling track parameters after noise reduction into feedback input signals of the improved model prediction controller by combining a minimum curvature method and a drilling track extension model, and inputting the feedback input signals to the improved model prediction controller to form vertical drilling closed-loop control;
in step S103, the particle filter is specifically a basic particle filter;
in step S101, the drilling trajectory extension model is as shown in formula (1):
in the formulas (1), (2) and (3), alpha is the inclination angle of the drilling track, beta is the azimuth angle of the drilling track, and alphaxIs the projection component, alpha, of the angle of inclination of the drilling path in the XOZ planeyThe projection component of the well drilling track inclination angle on the YOZ plane is shown;
the drilling speed is used;
for the horizontal X-direction component S of the drilling pathxA derivative of (a);
for horizontal Y-direction component S of drilling trackyA derivative of (a);
is alphaxA derivative of (a);
is alphayA derivative of (a); omegaSRThe rate of guidance for the drilling system;
orienting the drilling system magnetic tool at an angle; r is the build-up rate of the drilling system; mu.sxIs the component of the process noise in the X direction during the drilling process; mu.syIs the component of the process noise in the Y direction during drilling.
2. The vertical drilling deviation rectification control method based on particle filtering and model predictive control as claimed in claim 1, characterized in that: in step S102, the drilling trajectory extension model is linearized and discretized, specifically as shown in formula (4):
equation (4) is a linear discretization state space equation of the drilling track extension model, wherein
And
filter estimates of the actual lateral displacement of the tool of the drilling system in the X-axis and the Y-axis with respect to said reference trajectory at time k,
and
filter estimates, ω, of the projections on the XOZ and YOZ planes of the skew angle of the actual drilling trajectory relative to the reference trajectory at time k, respectivelyex(k) And omegaey(k) T is the sampling period for the control increment on both planes relative to the reference steering rate.
3. The vertical drilling deviation rectification control method based on particle filtering and model predictive control as claimed in claim 2, characterized in that: in step S102, a soft constraint condition and a variable optimization weight matrix are introduced to construct an improved model predictive controller, specifically: the prediction equation of the improved model prediction controller is as follows:
Y(k)＝Ξkx(k|k)+ΘkW(k) (5)
in the formula (5), xikAnd ΘkAre parameter matrices of a prediction equation; in formula (5), each matrix is represented by formula (6):
in the formula (6), p is a preset prediction step length, and c is a preset control step length; sex(k +1| k) and Sey(k +1| k) are actual lateral displacement deviations of a drilling tool of the drilling system relative to the reference track in the X axis and the Y axis at k +1 time predicted at k time respectively; alpha is alphaex(k +1| k) and αey(k +1| k) are respectively the deviations of the projection values of the well inclination angles of the actual drilling track at the k +1 moment predicted at the k moment relative to the reference track on the XOZ and YOZ planes;
introducing a soft constraint condition and a variable optimization weight matrix to obtain an optimization constraint condition of the improved model predictive controller, which is specifically shown as a formula (7):
in the formula (7), Q and R are weight matrixes of the state quantity Y (k) and the controlled quantity U (k) of the prediction equation respectively;
q is represented by formula (8):
wherein q issxIs a component S in the X directionxWeight of (a), qsyIs a Y-direction component SyWeight of aQ，bQ，cQAnd dQIs an angle weight factor; alpha is alpharx(m) and alphary(m) is the projection value of the well skew angle of the reference track at the moment m on the XOZ and YOZ planes;
and
the deviation of the projection value of the skew angle of the actual drilling track relative to the reference track at the moment m on the XOZ and YOZ planes is obtained; omegarx(m) and ωry(m) is a reference control quantity at the moment m; omegaex(m) and ωey(m) is the deviation of the actual controlled variable and the reference controlled variable at the moment m;
is an estimated value of the well inclination angle at the moment k; alpha is alphamaxIs a preset maximum soft constraint for the well inclination angle.
4. The vertical drilling deviation rectification control method based on particle filtering and model predictive control as claimed in claim 1, characterized in that: in step S104, the feedback input signal specifically includes: estimation of projection values of inclination angles of actual drilling paths on XOZ and YOZ planes
Estimation of the components of the horizontal displacement of the actual drilling trajectory in the X-direction and Y-direction
5. The vertical drilling deviation rectification control method based on particle filtering and model predictive control as claimed in claim 1, characterized in that: in step S104, after the feedback input signal is input into the improved model predictive controller, the output signal is the facing angle of the magnetic tool
And the guidance ratio omegaSRThe final actual controlled variable is specifically as shown in formula (9):
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010558687.6A CN111810112B (en) | 2020-06-18 | 2020-06-18 | Vertical drilling deviation rectifying control method based on particle filtering and model prediction control |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010558687.6A CN111810112B (en) | 2020-06-18 | 2020-06-18 | Vertical drilling deviation rectifying control method based on particle filtering and model prediction control |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111810112A CN111810112A (en) | 2020-10-23 |
| CN111810112B true CN111810112B (en) | 2021-12-03 |
Family
ID=72845207
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010558687.6A Active CN111810112B (en) | 2020-06-18 | 2020-06-18 | Vertical drilling deviation rectifying control method based on particle filtering and model prediction control |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111810112B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117072064A (en) * | 2023-08-28 | 2023-11-17 | 山东大学 | A jet intelligent correction control system and method for deep hole drilling rigs |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102425409A (en) * | 2011-11-10 | 2012-04-25 | 中国石油天然气集团公司 | Well deviation and tool surface measuring device and vertical drilling equipment |
| CN103266878A (en) * | 2013-05-03 | 2013-08-28 | 中国石油天然气集团公司 | Straightening control method and straightening control device for automatic vertical drilling system |
| AU2013400710A1 (en) * | 2013-09-17 | 2016-01-21 | Halliburton Energy Services, Inc. | Removal of stick-slip vibrations in a drilling assembly |
| CN108894768A (en) * | 2018-06-25 | 2018-11-27 | 中国地质大学（武汉） | A kind of drilling trace design method and system based on bat algorithm and wellbore stability |
| CN208257796U (en) * | 2018-04-28 | 2018-12-18 | 中国电子科技集团公司第二十二研究所 | Mine-used I. S signal receiving device |
| CN109477360A (en) * | 2016-06-02 | 2019-03-15 | 通用电气(Ge)贝克休斯有限责任公司 | Drilling track control based on depth |
| CN110456646A (en) * | 2019-08-23 | 2019-11-15 | 中国矿业大学 | Design method of self-adaptive sliding mode control system for drilling arm swing angle of rock bolter based on EKF |
| CN110513043A (en) * | 2019-07-15 | 2019-11-29 | 中国地质大学（武汉） | A deflection correction control method in vertical drilling process based on automatic steering drilling tool |
| CN110792430A (en) * | 2019-11-20 | 2020-02-14 | 中国地质大学（北京） | A method and device for measuring inclination while drilling based on multi-sensor data fusion |
| CN110886606A (en) * | 2019-11-20 | 2020-03-17 | 中国地质大学（北京） | A method and device for inertial inclination measurement assisted by feature quantities while drilling |
| CN111144053A (en) * | 2019-12-19 | 2020-05-12 | 中国地质大学（武汉） | Drilling process build-up rate prediction method and system based on genetic algorithm optimization |
Family Cites Families (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8783382B2 (en) * | 2009-01-15 | 2014-07-22 | Schlumberger Technology Corporation | Directional drilling control devices and methods |
| CN103883312B (en) * | 2013-07-11 | 2017-02-08 | 中国石油化工股份有限公司 | Universal method for forecasting in-target situation of guide drilling |
| CN106030031B (en) * | 2013-12-06 | 2019-11-19 | 哈里伯顿能源服务公司 | Computer-implemented method and system for controlling bottom hole assembly to follow planned wellbore path |
| CN106661938B (en) * | 2014-09-03 | 2021-05-25 | 哈里伯顿能源服务公司 | Automated Wellbore Trajectory Control |
| CN107542447A (en) * | 2016-06-23 | 2018-01-05 | 中国石油化工股份有限公司 | A kind of method for wellbore trace monitoring |
| GB2571863B (en) * | 2017-01-31 | 2021-11-24 | Halliburton Energy Services Inc | Optimal trajectory control for rotary steerable systems |
| CA3047407C (en) * | 2017-01-31 | 2021-06-22 | Halliburton Energy Services, Inc. | Sliding mode control techniques for steerable systems |
| CN106919805B (en) * | 2017-04-11 | 2019-03-26 | 中国石油大学(北京) | The real method and device for boring shale formation PDC drill bit rock drillability of prediction |
| US11360233B2 (en) * | 2017-09-12 | 2022-06-14 | Schlumberger Technology Corporation | Seismic image data interpretation system |
| US11268370B2 (en) * | 2018-03-26 | 2022-03-08 | Baker Hughes, A Ge Company, Llc | Model-based parameter estimation for directional drilling in wellbore operations |
| CN110765648A (en) * | 2019-11-08 | 2020-02-07 | 长江大学 | Visual control device and control method for petroleum drilling trajectory |
-
2020
- 2020-06-18 CN CN202010558687.6A patent/CN111810112B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102425409A (en) * | 2011-11-10 | 2012-04-25 | 中国石油天然气集团公司 | Well deviation and tool surface measuring device and vertical drilling equipment |
| CN103266878A (en) * | 2013-05-03 | 2013-08-28 | 中国石油天然气集团公司 | Straightening control method and straightening control device for automatic vertical drilling system |
| AU2013400710A1 (en) * | 2013-09-17 | 2016-01-21 | Halliburton Energy Services, Inc. | Removal of stick-slip vibrations in a drilling assembly |
| CN109477360A (en) * | 2016-06-02 | 2019-03-15 | 通用电气(Ge)贝克休斯有限责任公司 | Drilling track control based on depth |
| CN208257796U (en) * | 2018-04-28 | 2018-12-18 | 中国电子科技集团公司第二十二研究所 | Mine-used I. S signal receiving device |
| CN108894768A (en) * | 2018-06-25 | 2018-11-27 | 中国地质大学（武汉） | A kind of drilling trace design method and system based on bat algorithm and wellbore stability |
| CN110513043A (en) * | 2019-07-15 | 2019-11-29 | 中国地质大学（武汉） | A deflection correction control method in vertical drilling process based on automatic steering drilling tool |
| CN110456646A (en) * | 2019-08-23 | 2019-11-15 | 中国矿业大学 | Design method of self-adaptive sliding mode control system for drilling arm swing angle of rock bolter based on EKF |
| CN110792430A (en) * | 2019-11-20 | 2020-02-14 | 中国地质大学（北京） | A method and device for measuring inclination while drilling based on multi-sensor data fusion |
| CN110886606A (en) * | 2019-11-20 | 2020-03-17 | 中国地质大学（北京） | A method and device for inertial inclination measurement assisted by feature quantities while drilling |
| CN111144053A (en) * | 2019-12-19 | 2020-05-12 | 中国地质大学（武汉） | Drilling process build-up rate prediction method and system based on genetic algorithm optimization |
Non-Patent Citations (5)
| Title |
|---|
| "State-of-the-Art BHA Program Produces Un-Precedented Results";David C-K Chen, Min Wu;《International Petroleum Technology Conference》;20081205;全文 * |
| "垂直钻井系统纠斜机构脉宽调制控制研究";王燕，等;《石油钻探技术》;20150228;第43卷(第2期);全文 * |
| "基于虚拟仪器自动垂直钻井工具纠斜控制理论与实验研究";龚勋;《中国优秀硕士学位论文全文数据库 (工程科技Ⅰ辑)》;20100215(第2期);全文 * |
| "导向钻井工具姿态动态测量的自适应滤波方法";贺艳，等;《西安石油大学学报(自然科学版)》;20160630;第31卷(第6期);全文 * |
| "离散序列曼彻斯特解码在EM-MWD中的应用研究";陆承达，等;《科学技术与工程》;20150731;第15卷(第7期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111810112A (en) | 2020-10-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Kremers et al. | Model-based robust control of directional drilling systems | |
| EP3933166B1 (en) | Attitude measurement method | |
| CN110886606B (en) | Characteristic quantity-while-drilling assisted inertial inclinometry method and device | |
| CN110792430B (en) | A method and device for measuring inclination while drilling based on multi-sensor data fusion | |
| EP3640426A1 (en) | Attitude reference for tieback/overlap processing | |
| CA3099823A1 (en) | In-situ downhole measurement correction and control | |
| CN108571287B (en) | The hole trajectory control system of Kernel-based methods control | |
| WO2013032652A1 (en) | Minimum strain energy waypoint-following controller for directional drilling using optimized geometric hermite curves | |
| CN110513043B (en) | A control method for inclination correction in vertical drilling process based on automatic steerable drilling tools | |
| EP0793000A2 (en) | Method for correcting directional surveys | |
| CN114137836B (en) | Vertical drill deviation correction control feedback correction method and device based on Gaussian mixture model | |
| CN102954804B (en) | Sensor orthogonal calibration method for oil drilling measurement | |
| CN110440746A (en) | A kind of no-dig technique subterranean drill bit posture fusion method based on the decline of quaternary number gradient | |
| CN111810112B (en) | Vertical drilling deviation rectifying control method based on particle filtering and model prediction control | |
| Zhang et al. | A deviation correction strategy based on particle filtering and improved model predictive control for vertical drilling | |
| CN119878009A (en) | Drilling well track monitoring and controlling method based on multi-sensor fusion | |
| US12024995B2 (en) | Perturbation based well path reconstruction | |
| Zhang et al. | Model predictive control strategy based on improved trajectory extension model for deviation correction in vertical drilling process | |
| WO2023107126A1 (en) | Error-space feedback controller for drill bit steering | |
| CN109902890B (en) | Horizontal well landing target-in-target evaluation method and system | |
| CN113514025A (en) | Geomagnetic and deflection synchronous correction method and system for deep true 3D deformation monitoring | |
| CN114579934B (en) | Single-vector attitude and heading information extraction method | |
| CN119466718B (en) | Dynamic measurement method and system for attitude parameters of rotary guiding tool | |
| CN116066080B (en) | Magnetic method measurement method for radial horizontal well track | |
| Zhang et al. | Design of Deviation Correction Control System in Vertical Drilling Process |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |