CN114215501A - Control method for stable platform in rotary steering system - Google Patents
Control method for stable platform in rotary steering system Download PDFInfo
- Publication number
- CN114215501A CN114215501A CN202210165513.2A CN202210165513A CN114215501A CN 114215501 A CN114215501 A CN 114215501A CN 202210165513 A CN202210165513 A CN 202210165513A CN 114215501 A CN114215501 A CN 114215501A
- Authority
- CN
- China
- Prior art keywords
- control
- state
- neural network
- formula
- deviation
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000013528 artificial neural network Methods 0.000 claims abstract description 35
- 238000001914 filtration Methods 0.000 claims abstract description 20
- 230000009471 action Effects 0.000 claims abstract description 5
- 150000001875 compounds Chemical class 0.000 claims description 17
- 230000003044 adaptive effect Effects 0.000 claims description 8
- 239000011159 matrix material Substances 0.000 claims description 7
- 230000000087 stabilizing effect Effects 0.000 claims description 5
- 238000005553 drilling Methods 0.000 description 15
- 238000005516 engineering process Methods 0.000 description 9
- 230000000694 effects Effects 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 239000012530 fluid Substances 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000013459 approach Methods 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000004880 explosion Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000006641 stabilisation Effects 0.000 description 1
- 238000011105 stabilization Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000026683 transduction Effects 0.000 description 1
- 238000010361 transduction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
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
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)
Abstract
The invention discloses a control method of a stable platform in a rotary steering system, which comprises the steps of establishing a state equation and a neural network state observer; approximating the friction torque and the disturbance torque in the neural network state observer based on a Gaussian basis function; calculating the deviation between the expected value and the measured value; setting a virtual control law based on the deviation; performing first-order filtering on the virtual control law to obtain the deviation of a filtering result and an observation result of a state observer; finally, the self-adaptive rate and the control input voltage are obtained, and the input voltage is controlled to control the action of the torque generator in a control system of the stable platform. The invention is used for solving the problem that the control of the stable platform of the rotary guiding system is easily interfered by internal nonlinearity in the prior art, realizing the stable and rapid tracking of the command sent from the ground and leading the stable platform to provide a more stable and reliable deflecting direction and guiding force for the system.
Description
Technical Field
The invention relates to the field of drilling engineering, in particular to a control method for a stable platform in a rotary steering system.
Background
The rotary steering drilling technology is the most efficient drilling technology means for controlling the well track of the directional well in the drilling engineering; compared with the traditional directional mode of the screw drill with the bend angle, the directional drilling tool has the characteristics of small friction and torsional resistance, high drilling speed, low cost, short well construction period, smooth well track, easiness in regulation and control, capability of prolonging the length of a horizontal section and the like, and is considered as the development direction of the modern directional drilling technology. In the prior art, the rotary guiding technology in China is still in a continuous development stage, and mature rotary guiding technologies are mastered by foreign oil field service providers. In directional drilling, the curvature of a borehole track is determined by the build-up rate of a drilling tool, and the orientation and deflection performance of the drilling tool are represented by the accumulation rate and the toolface angle of the drilling tool, so that the toolface angle is an important parameter in borehole track control, and the accurate control of the toolface angle is one of main reasons for restricting the development and progress of rotary steering technology in China.
The rotary steerable tool is used as an orientation control component in a Bottom Hole Assembly (BHA), wherein the core is a stable platform, and the function of the stable platform is to adjust the tool face angle so that the tool face angle can track a given tool face angle instruction quickly and smoothly and keep stable under the condition of external disturbance as much as possible. In the application of the rotary steering drilling technology in China at present, a control method for a stable platform of a rotary steering tool is mainly open-loop control or closed-loop control combined with a classical PID algorithm; however, due to the fact that the operating environment conditions of the rotary steering tool are severe and the interference factors are numerous, uncertainty of parameters related to friction and interference torque inside the stable platform exists all the time, and many states are unknown or cannot be measured, the control effect on the stable platform is always unsatisfactory under the complex scene including strong nonlinear interference.
Disclosure of Invention
The invention provides a control method of a stable platform in a rotary steering system, which aims to solve the problem that the control of the stable platform of the rotary steering system is easily interfered by internal nonlinearity in the prior art, realize stable and rapid tracking of an instruction sent from the ground and enable the stable platform to provide a more stable and reliable deflecting direction and guiding force for the system.
The invention is realized by the following technical scheme:
a control method for stabilizing a platform in a rotary steering system comprises the following steps:
step S1, establishing a state equation of a stable platform in the rotary guiding tool;
step S2, rewriting the state equation into a state space form, and establishing a neural network state observer;
step S3, based on Gaussian function, applies value related to friction disturbance torquef(X) Carrying out approximation to obtain an approximation result
；
Step S4, an expected value of the tool face angle is input, and the deviation between the expected value and the measured value is calculatedz 1(ii) a And based on deviationz 1Setting a virtual control law;
step S5, performing first-order filtering on the virtual control law to obtain the deviation between the filtering result and the observation result of the state observerz 2；
Step S6, combining the approximation result and the deviationz 1And deviation fromz 2Obtaining the adaptive rate and controlling the input voltageu’；
Step S7, in the control system of the stable platform, to control the input voltageu’And controlling the action of the torque generator.
Aiming at disrotatory transduction in the prior artThe invention provides a control method of a stable platform in a rotary steering system, which is easy to be interfered by nonlinearity in the stable platform, and comprises the steps of firstly establishing a state equation of the stable platform, then rewriting the state equation into a state space form, establishing a neural network state observer based on the state space form, and approximating friction torque and interference torque in the neural network state observer based on a Gaussian basis function to obtain an approximation result; the method adopts a Gaussian basis function to approach the friction torque and the interference torque which can not be effectively measured in the stable platform. The desired value of the toolface angle of the rotary steerable tool is then input, which may be obtained by receiving commands from the surface, as will be appreciated by those skilled in the art. The deviation between the expected value and the measured toolface angle is calculated and defined asz 1(ii) a Based on the deviationz 1And setting a virtual control law for effectively overcoming unknown friction and interference. Specifically, the virtual control law is subjected to first-order filtering to obtain the deviation between the filtering result and the observation result of the state observer, which is defined asz 2(ii) a Finally, combining the approximation result and deviationz 1And deviation fromz 2Obtaining the adaptive rate, controlling the input voltageu’The self-adaptive rate as the control law of the neural network can be well overcome when facing actual unknown friction and interference in the underground, can maintain good control effect under the perturbation of parameters, and has stronger self-adaptability and robustness, thereby controlling the input voltageu’The torque generator is controlled to act to stably and quickly track the command sent out from the ground, so that the stable platform provides stable and reliable deflecting azimuth and guiding force for the rotary guiding system, the construction precision of the directional well is obviously improved, and the operation environment in the open hole well with severe conditions and numerous interference factors, which is faced by the rotary guiding tool, is overcome.
Further, the state equation of the stable platform is as follows:
the state variable in the state equation is x,
；x1representing the tool face angle, x2Represents an angular velocity;
a derivative representing the toolface angle;
represents the derivative of angular velocity; y represents the output toolface angle;
in the formula:umotor voltage for a torque generator;Ris a resistance;C eis the back electromotive force coefficient;C mis the motor torque coefficient;Jis the inertia moment of the motor;T fto stabilize the frictional disturbance torque in the platform; e is a modeling error;θis the tool face angle; omega is angular velocity;K W converting factors for the gyroscope;K PWM is the proportionality coefficient of pulse width modulation;K E the conversion factor of the electromagnetic torque and the current of the motor is obtained.
Further, in step S2, the state equation is rewritten into the following state space form:
in the formula:X=[x 1,x 2]Tand T represents the transpose of the vector,
representsXA derivative with respect to time;
the value related to the frictional disturbance torquef(X) Comprises the following steps:
further, the neural network state observer established in step S2 is:
in the formula:
representing the derivative with respect to time of the tool face angle or angular velocity vector derived by the state observer,
a tool face angle or angular velocity vector representing the output of the state observer;
in order to be a state observation value,
，
is a state observation;Kin order to be an observer gain vector,
，k 1、k 2are all gain values;
is composed off(X) A state estimate of (a);
the method comprises the following steps:
parameter to be determinedAIs the following Helverz matrix:
and satisfyA T P+PA=-2Q，PAndQis any given positive definite matrix;
。
it should be noted that: when the state observation object is the tool face angle, that is, X represents the tool face angle in the above description, the state observation value in this scheme
I.e. the value of the tool face angle, at this time
Representing the derivative of the toolface angle derived by the state observer,
a tool face angle representing the output of the state observer; when the state observation object is an angular velocity vector, that is, X represents an angular velocity vector in the above, the state observation value in this scheme
I.e. the value of the angular velocity vector, at this time
Representing the derivative of the angular velocity vector derived by the state observer,
the angular velocity vector representing the output of the state observer.
After the state observer is established, undetermined parameters A, B, C in a state equation in a state space form are defined: wherein by selecting appropriateKCan ensureATo satisfyA T P+PA=-2QThe helvets matrix of. By means of the state observer, efficient observation of a stable platform control system can be achieved, follow-up adaptive control over a stable platform can be fully achieved, and therefore the problem that control effects are not ideal due to uncertain reasons of friction and interference torque is solved.
Further, the approximation result
Calculated by the following formula:
in the formula (I), the compound is shown in the specification,
in order to be the weight estimation value of the neural network,
to be used for approximationf(X) Gaussian base function of (2).
Further, the virtual control law in step S4 is:
in the formula (I), the compound is shown in the specification,α 1the tool face angle obtained for the virtual control law;
is the derivative of the toolface angle;c 1is a set normal number.
In step S5, first order filtering is performed by the following low pass filter:
(ii) a In the formula (I), the compound is shown in the specification,β 1is composed ofα 1The output of the low-pass filtering of (c),τin order to be able to filter the time constant,
outputting a derivative with respect to time for the low-pass filtering;
deviation ofz 2Calculated by the following formula:
。
the traditional inversion control algorithm is easy to have the problem of differential explosion when used in the application, and the scheme adopts a dynamic surface control technology for this purpose, the method is to approximately replace differential calculation by carrying out first-order filtering through a low-pass filter, and calculate the deviation between a filtering result and an observation result of a state observer based on the calculationz 2And a sufficient basis is provided for the subsequent calculation of the self-adaptive rate and the control input voltage, so that the control input voltage finally obtained by the method can effectively overcome the operating environment with strong nonlinear interference.
Further, the adaptive rate obtained in step S6 is
：
In the formula (I), the compound is shown in the specification,γcontrol law parameters;
the output of the hidden layer of the neural network based on the Gaussian function with respect to the angular velocity;φcontrol law parameters;Wis the weight of the neural network.
According to the scheme, the self-adaptive rate is obtained through the output of the hidden layer of the neural network based on the Gaussian function on the angular velocity, the unknown interferences can be well overcome when the actual underground unknown friction and interferences are faced, and a good control effect can be maintained under the condition of parameter perturbation.
Further, the control input voltage obtained in step S6u’Comprises the following steps:
in the formula (I), the compound is shown in the specification,c 2control law parameters;Wis the weight of the neural network;
is an observation error vector;
is the output of the hidden layer of the neural network based on the Gaussian base function with respect to the angular velocity.
Further, the control system for stabilizing the platform in step S7 is a closed-loop control system.
Compared with the prior art, the invention has the following advantages and beneficial effects:
1. the control method of the stable platform in the rotary steering system can well overcome the situation of unknown friction and interference in the underground, can maintain good control effect under the condition of parameter perturbation, has stronger self-adaptability and robustness, and can realize that the tool face angle stably and quickly tracks the instruction sent from the ground, so that the stable platform provides stable and reliable deflecting azimuth and guiding force for the rotary steering system, the construction precision of a directional well is obviously improved, and the operation environment in the open hole well with severe conditions and numerous interference factors, which is faced by a rotary steering tool, is overcome.
2. The invention relates to a control method of a stable platform in a rotary steering system, which utilizes a strategy of combining an RBF neural network and a state observer to ensure that the stable platform of the rotary steering system can keep stable under the condition of friction interference and has self-adaptability under the error caused by parameter perturbation.
3. According to the control method for the stable platform in the rotary steering system, the problem of differential explosion existing in an inversion method is avoided by adding dynamic surface control, so that good tracking performance can be obtained under multiple influences, and the control method has better performance compared with the traditional controller.
4. Compared with the traditional control mode, the control method for the stable platform in the rotary guide system has the advantages that the stable platform can track a certain fixed tool face angle under the action of the control method, a given time-varying continuous tool face angle command signal is considered, the tracking effect is achieved by designing the adaptivity of the control method, and the control method has a reference significance for actual engineering.
Drawings
The accompanying drawings, which are included to provide a further understanding of the embodiments of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the principles of the invention. In the drawings:
FIG. 1 is a schematic structural diagram of a stabilization platform in an embodiment of the present invention;
FIG. 2 is a flow chart of steps of a control method according to an embodiment of the present invention;
FIG. 3 is a control flow diagram of an embodiment of the present invention;
FIG. 4 is a schematic diagram of a closed loop control system in an embodiment of the present invention;
FIG. 5 is a tool face angle trace curve in an embodiment of the present invention;
FIG. 6 illustrates tool face angle tracking error in an embodiment of the present invention;
FIG. 7 illustrates control input signals according to an embodiment of the present invention;
FIG. 8 is a friction torque approximation curve according to an embodiment of the present invention;
FIG. 9 shows z in an embodiment of the present invention1、z2The error phase of (2);
FIG. 10 is a trace error curve after friction and parameter changes in an embodiment of the present invention;
FIG. 11 is a tool face angle trace curve after friction and parameter changes in an embodiment of the present invention;
FIG. 12 is a friction torque approximation curve after friction and parameter changes in an embodiment of the present invention.
Reference numbers and corresponding part names in the drawings:
1-mud turbine, 2-drill collar, 3-upper turbine engine, 4-electronic control cabin and 5-torque generator.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to examples and accompanying drawings, and the exemplary embodiments and descriptions thereof are only used for explaining the present invention and are not meant to limit the present invention.
Example 1:
a method for controlling a stabilized platform in a rotary steerable system, as shown in fig. 2, comprises the following steps:
step S1, establishing a state equation of the stable platform in the rotary guiding tool:
the stable platform of the rotary steering system is shown in fig. 1 and comprises an upper turbine generator, a torque generator (lower turbine torque motor), an electronic control unit, a position sensor and the like. The upper turbine motor is responsible for supplying power, and the moment generated by the upper turbine motor on the main shaft of the stable platform can be regarded as a constant value because the load resistance of the upper turbine motor is not changed. The lower turbine torque motor is used as a torque generator, the Pulse Width Modulation (PWM) technology is adopted, and the output torque is adjusted by changing the current, so that the angle and the speed of a tool face are changed. The position sensor in the stabilized platform comprises a gyroscope, a linear accelerometer, an amplifier and the like. Adding a gyroscope conversion factor
Pulse width modulation proportionality coefficient
Electromagnetic torque and current conversion factor of motor
And stabilizing the internal structure of the platform to obtain a control system, and obtaining a state equation of the stabilized platform as follows:
the state isThe state variable in the process is x,
；x1representing the tool face angle, x2Represents an angular velocity;
a derivative representing the toolface angle;
represents the derivative of angular velocity; y represents the output toolface angle;
in the formula:umotor voltage for a torque generator;Ris a resistance;C eis the back electromotive force coefficient;C mis the motor torque coefficient;Jis the inertia moment of the motor;T fto stabilize the frictional disturbance torque in the platform; e is a modeling error;θis the tool face angle; omega is angular velocity;K W converting factors for the gyroscope;K PWM is the proportionality coefficient of pulse width modulation;K E the conversion factor of the electromagnetic torque and the current of the motor is obtained.
Step S2, rewriting the state equation into a state space form:
in the formula:X=[ x 1,x 2]Tand T represents the transpose of the vector,
representsXA derivative with respect to time;
establishing a neural network state observer:
in the formula:
representing the derivative with respect to time of the tool face angle or angular velocity vector derived by the state observer,
a tool face angle or angular velocity vector representing the output of the state observer;
in order to be a state observation value,
，
is a state observation;Kin order to be an observer gain vector,
，k 1、k 2are all gain values;
is composed off(X) The state estimation value of (1), namely an approximation result after approximation is carried out through a Gaussian basis function;
the method comprises the following steps:
parameter to be determinedAIs a herviz matrix as follows,
and satisfyA T P+PA=-2Q，PAndQis any given positive definite matrix;
。
wherein
Calculated by the following formula:
in the formula (I), the compound is shown in the specification,
in order to be the weight estimation value of the neural network,
to be used for approximationf(X) Gaussian base function of (2).
S3, approximating the friction torque and the disturbance torque in the neural network state observer based on the Gaussian basis function to obtain an approximation result;
step S4, an expected value of the tool face angle is input, and the deviation between the expected value and the measured value is calculatedz 1(ii) a And based on deviationz 1Setting a virtual control law:
in the formula (I), the compound is shown in the specification,α 1the tool face angle obtained for the virtual control law;
is the derivative of the toolface angle;c 1is a set normal number.
Step S5, performing first order filtering on the virtual control law:
in the formula (I), the compound is shown in the specification,β 1is composed ofα 1The output of the low-pass filtering of (c),τin order to be able to filter the time constant,
outputting a derivative with respect to time for the low-pass filtering;
obtaining the deviation between the filtering result and the observation result of the state observerz 2：
。
Step S6, combining the approximation result and the deviationz 1And deviation fromz 2Obtaining the adaptive rate and controlling the input voltageu’：
In the formula (I), the compound is shown in the specification,γcontrol law parameters;
the output of the hidden layer of the neural network based on the Gaussian function with respect to the angular velocity;φcontrol law parameters;Wis the weight of the neural network;c 2control law parameters;
is an observation error vector.
Step S7, in the closed loop control system of the stable platform, to control the input voltageu’And controlling the action of the torque generator.
The closed loop control system adopted by the embodiment is shown in FIG. 4, in FIG. 4θ d For the desired value of the face angle of the tool,sis a complex variable after laplace transformation. The definitions of the other parameters are described in this embodiment, and are not described herein.
It should be noted that the control flow chart shown in fig. 3 is an adaptive control law of a neural network based on a state observer, which is specifically designed for the stable platform characteristics according to the present application. The method comprises the steps of utilizing approximation characteristics of an RBF neural network to approximate friction torque and interference torque which cannot be measured, utilizing a state observer to control modeling errors caused by uncertain related parameters, and utilizing an inversion control method of a first-order filter to realize control.
Example 2:
on the basis of embodiment 1, a detailed method for approximating friction torque and disturbance torque in a neural network state observer by using a gaussian basis function in this embodiment is as follows:
the gaussian base function is first defined as follows:
in the formula (I), the compound is shown in the specification,c j for the center vector of the hidden layer,b j is the width of the gaussian function and,Was a weight of the neural network, the weight of the neural network,σ j (x) In order to imply the output of the layer,mtaking a positive integer larger than 1.
The approximated object is represented as:y=W T σ j (x)。
the following guidelines are given: for a given continuous functionf(x) Defining a tight set omega eR n Existence of the optimal weightW * For any given small positive numberkMake the approximation errorεThe following formula is satisfied:
in the formula (I), the compound is shown in the specification,
；
to be at the optimum weightW *A neural network estimate;
according to the above theory, in the formulaf(x) The neural network estimate of
Then the estimation error is:
。
setting up observation error terms and definingδ=ε+dThen, the following can be obtained:
in the formula (I), the compound is shown in the specification,
is composed of
With respect to the derivative of time,
is composed ofXThe derivative with respect to time, A, B are respectively the pending parameters,
and satisfy-d|≤d m，d mIs an unknown positive number.
Example 3:
in this example, the control method described in example 1 or example 2 was verified based on the actual simulation result. Before verification, parameters of each model of the stable platform need to be set, wherein typical parameters are shown in table 2.
In Table 2GIn order to stabilize the total weight of the platform,G 2in order to increase the weight of the turbine motor,G 3in order to reduce the weight of the lower turbine torque motor,r 0andrrespectively the inner diameter of the drill collar and the outer radius of the platform,r 1the average friction radius of the bearing of the turbine motor is the average friction radius of the bearing of the upper turbine motor and the lower turbine motor because the upper turbine motor and the lower turbine motor are consistent in sizer 2= r 3，μIs the viscosity system of the drilling fluid,Lin order to be able to measure the total length of the drilling fluid,f μ1、f μ2the rolling friction coefficients of the radial bearing and the axial bearing respectively,αin order to be the angle of inclination,n 0the definitions of the other parameters for the collar rotation speed are described in the above embodiments.
For RBF neural network parameters, selecting 5 neural nodes, selecting Gaussian function center, and uniformly distributing in [ -1,1 [ -1]Within the range, i.e.c=[-1,-0.5,0,0.5,1]Width of Gaussian functionb=0.5, and let the weight value initial valueW(0)=0。
The controller parameters were designed as shown in table 1:
TABLE 1 control law parameters
| Parameter(s) | γ | τ | K | φ | c 1 | c 2 |
| Value taking | 0.1 | 0.05 | 180 | 0.5 | 3 | 2 |
Adding modeling errorse(t)=cos(2t+0.5), set the ideal toolface angle input signaly d = sint, system initial statex 0 =[-0.1,0]. Verification is carried out based on the given parameters, and the results shown in FIGS. 5-9 are obtained. As can be seen from FIG. 5, the system can stably track the input ideal toolface angle signal within 0.6 s; as can be seen from FIG. 6, the steady-state tracking error of the system is 2.5 × 10-2Within. Therefore, under the condition that a plurality of friction and modeling errors exist simultaneously, the control law provided by the invention can achieve a quick and accurate tracking effect. Fig. 7 shows the control input and fig. 8 shows the friction torque approach, which can be seen to track well. FIG. 9 shows the deviationz 1、z 2It can be known that they all eventually converge.
In addition, parameters such as rotational inertia are caused by temperature and drilling fluid in consideration of actual downhole operationJLoad resistorRAnd actually, the friction term considered in the embodiment is not the whole friction torque, and other friction torque exists in the actual operationUnknown frictional disturbances are not taken into account. Therefore, the embodiment further doubles the friction torque, reduces the resistance and the moment of inertia by 20%, and changes the frequency of the modeling error term to makee(t)=cos(5t+0.5) to obtain the results shown in FIGS. 10 to 12. It can be seen that the tracking of the tool face angle is still good, a given input signal can be tracked within 0.7s, and the steady state error is also kept at 3 × 10-2Within.
Specifically, the following are mentioned:
in fig. 5 and 11, the abscissa is time in seconds(s); the ordinate is the tool face angle in radians (rad); in the drawingsy d For the desired angle of the tool face,yactual tool face angle;
the abscissa of fig. 6 and 10 is time in seconds(s); the ordinate is the tool face angle tracking error in radians (rad);
the abscissa in fig. 7 is time in seconds(s); the ordinate is the controller input voltage in volts (v);
the abscissa of fig. 8 and 12 is time in seconds(s); the ordinate is the frictional disturbance torque in newtons per meter (N.M); in the drawingsT f In order to be the actual moment of force,
is an approximated moment;
in fig. 9, the abscissa is the deviation z1 in the present embodiment, and the ordinate is the deviation z2 in the present embodiment.
TABLE 2 rotating guidance System stabilized platform model parameters
In summary, the present embodiment can fully prove that the control method provided by the present invention can be well overcome in the face of actual unknown downhole friction and interference, can maintain a good control effect under the perturbation of parameters, and has extremely strong adaptability and robustness.
The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are merely exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.
It should be noted that, in this document, terms such as "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Claims (10)
1. A control method for stabilizing a platform in a rotary steering system is characterized by comprising the following steps:
step S1, establishing a state equation of a stable platform in the rotary guiding tool;
step S2, rewriting the state equation into a state space form, and establishing a neural network state observer;
step S3, based on Gaussian function, applies value related to friction disturbance torquef(X) Carrying out approximation to obtain an approximation result
；
Step S4, an expected value of the tool face angle is input, and the deviation between the expected value and the measured value is calculatedz 1(ii) a And based on deviationz 1Setting a virtual control law;
step S5, performing first-order filtering on the virtual control law to obtain the deviation between the filtering result and the observation result of the state observerz 2；
Step S6, combining the approximation result and the deviationz 1And deviation fromz 2Obtaining the adaptive rate and controlling the input voltageu’；
Step S7, in the control system of the stable platform, to control the input voltageu’And controlling the action of the torque generator.
2. The method of claim 1, wherein the stable platform of the rotary steerable system has the following equation of state:
the state variable in the state equation is x,
；x 1representing the angle of the tool face,x 2Represents an angular velocity;
represents the derivative of the toolface angle with respect to time;
represents the derivative of angular velocity with respect to time; y represents the output toolface angle;
in the formula:umotor voltage for a torque generator;Ris a resistance;C eis the back electromotive force coefficient;C mis the motor torque coefficient;Jis the inertia moment of the motor;T fto stabilize the frictional disturbance torque in the platform; e is a modeling error;θis the tool face angle; omega is angular velocity;K W converting factors for the gyroscope;K PWM is the proportionality coefficient of pulse width modulation;K E the conversion factor of the electromagnetic torque and the current of the motor is obtained.
3. The method of claim 2, wherein in step S2, the state equation is rewritten into the following state space form:
in the formula:X=[ x 1,x 2]Tand T represents the transpose of the vector,
representsXA derivative with respect to time;
the value related to the frictional disturbance torquef(X) Comprises the following steps:
4. the method for controlling a stable platform in a rotary steerable system according to claim 3, wherein the neural network state observer established in step S2 is:
in the formula:
representing the derivative with respect to time of the tool face angle or angular velocity vector derived by the state observer,
a tool face angle or angular velocity vector representing the output of the state observer;
in order to be a state observation value,
，
is a state observation;Kin order to be an observer gain vector,
，k 1、k 2are all gain values;
the method comprises the following steps:
parameter to be determinedAIs the following Helverz matrix:
and satisfyA T P+PA=-2Q，PAndQis any given positive definite matrix;
。
5. the method of claim 4, wherein the approximation result is used to control a stable platform in a rotary steerable system
Calculated by the following formula:
in the formula (I), the compound is shown in the specification,
in order to be the weight estimation value of the neural network,
to be used for approximationf(X) Gaussian base function of (2).
6. The method as claimed in claim 3, wherein the virtual control law in step S4 is as follows:
in the formula (I), the compound is shown in the specification,α 1the tool face angle obtained for the virtual control law;
is the derivative of the toolface angle;c 1is a set normal number.
7. The method as claimed in claim 6, wherein the step S5 is performed by a low pass filter with a first order filtering:
(ii) a In the formula (I), the compound is shown in the specification,β 1is composed ofα 1The output of the low-pass filtering of (c),τin order to be able to filter the time constant,
outputting a derivative with respect to time for the low-pass filtering;
deviation ofz 2Calculated by the following formula:
。
8. the method as claimed in claim 7, wherein the adaptive rate obtained in step S6 is
：
In the formula (I), the compound is shown in the specification,γcontrol law parameters;
the output of the hidden layer of the neural network based on the Gaussian function with respect to the angular velocity;φcontrol law parameters;Wis the weight of the neural network.
9. The method as claimed in claim 7, wherein the control input voltage obtained in step S6 is the control input voltageu’Comprises the following steps:
in the formula (I), the compound is shown in the specification,c 2control law parameters;Wis the weight of the neural network;
is an observation error vector;
is the output of the hidden layer of the neural network based on the Gaussian base function with respect to the angular velocity.
10. The method as claimed in claim 1, wherein the control system for stabilizing the platform in step S7 is a closed loop control system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210165513.2A CN114215501A (en) | 2022-02-23 | 2022-02-23 | Control method for stable platform in rotary steering system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210165513.2A CN114215501A (en) | 2022-02-23 | 2022-02-23 | Control method for stable platform in rotary steering system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114215501A true CN114215501A (en) | 2022-03-22 |
Family
ID=80709340
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210165513.2A Pending CN114215501A (en) | 2022-02-23 | 2022-02-23 | Control method for stable platform in rotary steering system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114215501A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117627616A (en) * | 2024-01-25 | 2024-03-01 | 中国矿业大学 | Hydraulic jumbolter self-adaptive anti-disturbance global terminal sliding mode rotating speed control method |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN200947580Y (en) * | 2006-09-08 | 2007-09-12 | 西安石油大学 | Electricity-driven unit of rotary guide drilling based on power factor correction |
| US20090090555A1 (en) * | 2006-12-07 | 2009-04-09 | Nabors Global Holdings, Ltd. | Automated directional drilling apparatus and methods |
| CN105064917A (en) * | 2015-06-30 | 2015-11-18 | 中国石油天然气股份有限公司 | Rotary guide system and control method thereof |
| US20170067332A1 (en) * | 2015-09-03 | 2017-03-09 | Schlumberger Technology Corporation | Rotary steerable roll stabilized control system |
| CN106873358A (en) * | 2015-11-19 | 2017-06-20 | 欧姆龙株式会社 | Control Device And Control Method |
| CN110500034A (en) * | 2019-08-30 | 2019-11-26 | 中国石油集团川庆钻探工程有限公司 | Method for establishing neural network model, determining torsional pendulum drill string parameters and directionally drilling |
| CN111997583A (en) * | 2020-09-01 | 2020-11-27 | 中国石油集团渤海钻探工程有限公司 | Rotating guide motor speed regulation method based on fuzzy neural network |
| CN113361124A (en) * | 2021-06-22 | 2021-09-07 | 中国石油大学(华东) | Tool face angle estimation method of rotary steering drilling tool system |
| CN113482590A (en) * | 2021-08-04 | 2021-10-08 | 西南石油大学 | Bending screw rod deflecting parameter control method and system based on downhole robot |
-
2022
- 2022-02-23 CN CN202210165513.2A patent/CN114215501A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN200947580Y (en) * | 2006-09-08 | 2007-09-12 | 西安石油大学 | Electricity-driven unit of rotary guide drilling based on power factor correction |
| US20090090555A1 (en) * | 2006-12-07 | 2009-04-09 | Nabors Global Holdings, Ltd. | Automated directional drilling apparatus and methods |
| CN105064917A (en) * | 2015-06-30 | 2015-11-18 | 中国石油天然气股份有限公司 | Rotary guide system and control method thereof |
| US20170067332A1 (en) * | 2015-09-03 | 2017-03-09 | Schlumberger Technology Corporation | Rotary steerable roll stabilized control system |
| CN106873358A (en) * | 2015-11-19 | 2017-06-20 | 欧姆龙株式会社 | Control Device And Control Method |
| CN110500034A (en) * | 2019-08-30 | 2019-11-26 | 中国石油集团川庆钻探工程有限公司 | Method for establishing neural network model, determining torsional pendulum drill string parameters and directionally drilling |
| CN111997583A (en) * | 2020-09-01 | 2020-11-27 | 中国石油集团渤海钻探工程有限公司 | Rotating guide motor speed regulation method based on fuzzy neural network |
| CN113361124A (en) * | 2021-06-22 | 2021-09-07 | 中国石油大学(华东) | Tool face angle estimation method of rotary steering drilling tool system |
| CN113482590A (en) * | 2021-08-04 | 2021-10-08 | 西南石油大学 | Bending screw rod deflecting parameter control method and system based on downhole robot |
Non-Patent Citations (1)
| Title |
|---|
| WAN MIN等: "Observer-based adaptive neural network control for stabilized platform in rotary steerable system with unknown input dead-zone", 《TRANSACTIONS OF THE INSTITUTE OF MEASUREMENT AND CONTROL》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117627616A (en) * | 2024-01-25 | 2024-03-01 | 中国矿业大学 | Hydraulic jumbolter self-adaptive anti-disturbance global terminal sliding mode rotating speed control method |
| CN117627616B (en) * | 2024-01-25 | 2024-03-26 | 中国矿业大学 | Hydraulic jumbolter self-adaptive anti-disturbance global terminal sliding mode rotating speed control method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107202584B (en) | Planet accurate landing anti-interference guidance method | |
| US8115331B2 (en) | Wind turbine generator, active damping method thereof, and windmill tower | |
| US10001004B2 (en) | Closed loop model predictive control of directional drilling attitude | |
| CN102279564B (en) | Flight simulation rotating table control system and method applying intelligent PID (Proportion Integration Differentiation) controller | |
| CN111324142B (en) | Disturbance compensation control method for missile pilot | |
| CN113359479B (en) | Composite anti-interference control method for DGVSCMG frame servo system | |
| US10876389B2 (en) | Advanced toolface control system for a rotary steerable drilling tool | |
| US10883355B2 (en) | Nonlinear toolface control system for a rotary steerable drilling tool | |
| CN111025904B (en) | Magnetic levitation ball continuous sliding mode control method based on extended state observer | |
| CN114215501A (en) | Control method for stable platform in rotary steering system | |
| CN102033548A (en) | RBF neural network-based servo control system and method | |
| WO2016076828A1 (en) | Feedback based toolface control system for a rotary steerable drilling tool | |
| CN111781829A (en) | Neural network control method for backlash compensation of turntable servo system | |
| CN116009568A (en) | Aircraft robust control method based on angular acceleration feedback | |
| US10221672B2 (en) | Rotary steerable roll stabilized control system | |
| CN107795317A (en) | A kind of rotary valve method for controlling number of revolution of measurement while drilling instrument | |
| CN114625005A (en) | Backstepping anti-interference rotating speed control method for control moment gyro frame servo system | |
| CN106059413B (en) | A kind of fly wheel system method for controlling number of revolution of direct current generator driving | |
| CN113525415A (en) | Automatic-driving automobile layered anti-interference path tracking system based on rack force observation | |
| CN107959453A (en) | A kind of improved MRAS speed observation procedure | |
| CN111102887B (en) | Projectile body roll angle control method based on reaction flywheel | |
| Chen et al. | Sliding mode control for manipulator based on fuzzy switching gain adjustment | |
| CN112696981B (en) | Full closed loop interference rate compensation self-stabilization control method under geodetic coordinate system | |
| Song et al. | Tracking control of rotary steerable toolface in directional drilling | |
| Chen et al. | Adaptive Control of Stabilized Platform for Rotating Steering System |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220322 |
|
| RJ01 | Rejection of invention patent application after publication |