CN111997583A - Rotating guide motor speed regulation method based on fuzzy neural network - Google Patents

Rotating guide motor speed regulation method based on fuzzy neural network Download PDF

Info

Publication number
CN111997583A
CN111997583A CN202010903284.0A CN202010903284A CN111997583A CN 111997583 A CN111997583 A CN 111997583A CN 202010903284 A CN202010903284 A CN 202010903284A CN 111997583 A CN111997583 A CN 111997583A
Authority
CN
China
Prior art keywords
neural network
fuzzy
layer
fuzzy neural
rotary steering
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
Application number
CN202010903284.0A
Other languages
Chinese (zh)
Inventor
宋晓健
魏春明
董晨曦
郑邦贤
穴强
李宝鹏
梁昶婧
王海
毕雨萌
李新杰
张悦
刘旭
王涛
李峰
鲍伟伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China National Petroleum Corp
CNPC Bohai Drilling Engineering Co Ltd
Original Assignee
China National Petroleum Corp
CNPC Bohai Drilling Engineering Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by China National Petroleum Corp, CNPC Bohai Drilling Engineering Co Ltd filed Critical China National Petroleum Corp
Priority to CN202010903284.0A priority Critical patent/CN111997583A/en
Publication of CN111997583A publication Critical patent/CN111997583A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • E—FIXED CONSTRUCTIONS
    • E21—EARTH DRILLING; MINING
    • E21B—EARTH DRILLING, e.g. DEEP 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 DRILLING; MINING
    • E21B—EARTH DRILLING, e.g. DEEP 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/004—Artificial life, i.e. computing arrangements simulating life
    • G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/043—Architecture, e.g. interconnection topology based on fuzzy logic, fuzzy membership or fuzzy inference, e.g. adaptive neuro-fuzzy inference systems [ANFIS]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods

Abstract

The invention discloses a speed regulation method of a rotary steering motor based on a fuzzy neural network, which is based on fuzzy neural network control and is widely applied to the field of speed regulation of nonlinear motors. The method can accurately control the rotating speed of the eccentric shaft motor of the directional rotary guide system and has better robustness and faster response speed.

Description

Rotating guide motor speed regulation method based on fuzzy neural network
Technical Field
The invention relates to the technical field of petroleum and natural gas drilling equipment, in particular to a fuzzy neural network-based speed regulation method for a rotary steering motor, which is applied to the field of rotary steering and is based on a fuzzy neural network PID controller.
Background
The rotary steering drilling technology is an important mark for advancing to automatic and intelligent drilling, is a three-dimensional borehole trajectory control drilling technology taking an underground closed-loop control system as a core, and represents the leading level of the current oil and gas drilling engineering. In recent years, drilling is carried out in horizontal wells, extended reach wells and three-dimensional multi-target wells abroad, particularly for shale gas wells in North America, the rotary steering drilling technology is widely popularized and applied, the drilling speed is increased, accidents are reduced, the drilling cost is also reduced, and the purposes of cost reduction and efficiency improvement are achieved.
The rotary guide system can be divided into a push type and a directional type according to the guide mode. The push-type rotary steering system mainly utilizes the guide block to push against the well wall, so as to generate lateral force on the drill bit and push the drill bit away from the direction, thereby achieving the purpose of changing well deviation and direction. The directional rotary steering system biases the main shaft through a biasing mechanism between the outer sleeve and the rotary main shaft, thereby providing a dip angle which is inconsistent with the axis of the well hole for the drill bit to generate a steering effect. Because the directional rotary steering drilling system is more smooth in borehole track and better in borehole quality, the directional rotary steering drilling system does not push against the borehole wall in the working process than the push-against type drilling system, and becomes the development direction of the current rotary steering system technology.
The eccentric shaft motor is one of the key guiding components of the directional rotary guiding system. The eccentric shaft motor outputs torque, the eccentric shaft is driven to rotate through the gear reduction box, and the eccentric shaft is connected with the drill bit driving shaft through the eccentrically-mounted bearing to drive the drill bit to rotate. The rotation direction of the eccentric shaft is opposite to that of the drill collar, when the rotation speeds of the eccentric shaft and the drill collar are consistent, the eccentric shaft is a non-rotating body relative to the ground, and the tool face angle is unchanged. Therefore, in order to change the tool surface, the rotation speed of the eccentric shaft motor needs to be increased or decreased, a certain deviation is generated between the rotation speed of the eccentric shaft motor and the rotation speed of the drill collar, and after the expected tool surface is reached, the rotation speed of the eccentric shaft motor is maintained to be the same as that of the drill collar again, so that the new tool surface is stable. Different from the conventional motor rotating speed single-speed closed-loop control, the directional rotary steering system requires that under the condition of only controlling the rotating speed of the eccentric shaft, the rotating speed of the eccentric shaft is controlled to reach the rotating speed of the drill collar, and simultaneously, the tool face can also reach the position expected to be controlled, namely, the double closed-loop control of a rotating speed ring and a tool face angle position ring is realized simultaneously.
The prior PID system for online speed regulation of the rotating speed of the eccentric shaft motor of the rotary steering system has unstable working state and weak anti-interference capability. Therefore, a speed regulation method of the rotary steering motor based on the fuzzy neural network is developed to overcome the defects.
Disclosure of Invention
The invention aims to make up the defects of the prior art, provides a speed regulation method of a rotary steering motor based on a fuzzy neural network, which has the advantages of high calculation precision, strong anti-interference capability, automatic updating and optimization of a model structure and good popularization performance, integrates the advantages of fuzzy control and the neural network, can carry out fuzzy logic reasoning by using limited fuzzy rule information, has good approximation capability on a nonlinear system, and has good effect on solving the problems of nonlinearity and large time lag.
In order to solve the technical problems, the invention adopts the following technical scheme:
a speed regulation method of a rotary steering motor based on a fuzzy neural network comprises the following steps:
step S1, calculating a geometric offset vector between the drilled well track and the designed well track of the directional rotary steering system;
and step S2, carrying out PID control adjustment on the angular speed of the eccentric shaft motor of the directional rotary guide system according to the calculated geometric deviation vector.
Further, the specific process of step S1 is as follows:
step S11, calculating the curvature k of the drilled borehole of the directional rotary guiding system;
step S12, calculating the device angle TFA of the drilled borehole trajectory of the directional rotary steerable systempath；
Step S13, calculating the rotary steerable system position offset vector AC' and the tool surface TFA of the rotary steerable system toolRSS。
Further, the specific process of step S2 is as follows:
s21, performing closed-loop control on a control object by adopting a conventional PID controller, and performing online adjustment on 3 parameters of the PID controller by using a fuzzy neural network so as to realize the self-adaptive control of the control parameters of the PID controller;
step S22, establishing a fuzzy neural network; the fuzzy neural network comprises an input layer, a membership function layer, a fuzzy inference layer, a normalization layer and an output layer; wherein the content of the first and second substances,
the input layer is used for transmitting the rotating speed error e and the rotating speed error change rate delta e of the rotating speed of the eccentric shaft motor to the membership function layer;
the membership function layer is used for fuzzifying the input variable and calculating membership functions of all components of the input layer relative to all language variable value fuzzy sets;
each node in the fuzzy inference layer represents a fuzzy rule in a known fuzzy rule base, and the layer is used for determining the matching condition of the fuzzy rules and calculating the fitness of each fuzzy rule;
the number of nodes of the normalization layer is the same as that of the nodes of the fuzzy inference layer, and the normalization layer is used for realizing normalization operation;
the output layer is the proportional coefficient K of three indexes of the PID controllerPIntegral coefficient KiDifferential gain coefficient KdPerforming a sharpening operation as an output;
s23, optimizing the structural parameters of the fuzzy neural network; the mean value, the standard deviation and the network connection weight of the parameter membership Gaussian function of the fuzzy neural network are continuously adjusted by a random gradient descending training method, so that the nonlinear accurate control of a real-time link and a time-delay link of a tool surface of a rotary steering system tool is realized.
Further, the specific process of step S23 is as follows:
step S231, initialization; the particle swarm initialization comprises the steps of setting particle numbers, particle initial positions, speeds, inertia factors, acceleration factors and maximum iteration times;
step S232, improving the optimal solution of the structural parameters of the fuzzy neural network by adopting a particle swarm algorithm;
step S233, increasing the cycle number by 1; if the optimal solution condition is met, obtaining an improved optimal individual;
and S234, if the evolution meets the maximum iteration times, stopping the iteration process, and obtaining the optimal solution of the structural parameters of the fuzzy neural network, otherwise, jumping to the step S232 to perform circular execution.
Compared with the prior art, the invention has the beneficial effects that:
(1) compared with the conventional control method, the real-time link rise time and the time-lag link rise time of the embodiment are greatly shortened, so that the fuzzy neural network has better robustness and tracking capability and is more suitable for on-line training and learning;
(2) after the control method is used for processing the rotating speed of the eccentric shaft motor of the pointing type rotary guide system, the real-time link overshoot can be reduced when the time link adjusting time meets the requirement.
Drawings
In order that the advantages of the invention will be readily understood, a more particular description of the invention briefly described above will be rendered by reference to specific embodiments that are illustrated in the appended drawings. Understanding that these drawings depict only typical embodiments of the invention and are not therefore to be considered to be limiting of its scope, the invention will be described and explained with additional specificity and detail through the use of the accompanying drawings.
FIG. 1 is a flow chart of a speed regulation method for a rotary steering motor based on a fuzzy neural network provided by the present invention;
FIG. 2 is a fuzzy neural network structure related to a speed regulation method of a rotary steering motor based on a fuzzy neural network provided by the present invention;
FIG. 3 is a schematic tool calculation diagram of a rotary steering system for a speed regulation method of a rotary steering motor based on a fuzzy neural network according to the present invention;
FIG. 4 is a schematic diagram of tool offset vector calculation of a rotary steering system according to a speed regulation method of a rotary steering motor based on a fuzzy neural network provided in the present invention;
FIG. 5 is a flow chart of a particle swarm algorithm for a rotating guide motor speed regulation method based on a fuzzy neural network, provided by the invention;
fig. 6 is an effect diagram of a speed regulation method for a rotary steering motor based on a fuzzy neural network provided by the invention.
Detailed Description
In the following description, numerous specific details are set forth in order to provide a more thorough understanding of the present invention. It will be apparent, however, to one skilled in the art, that embodiments of the invention may be practiced without one or more of these specific details. In other instances, well-known features have not been described in detail so as not to obscure the embodiments of the invention.
In the following description, a detailed structure will be presented for a thorough understanding of embodiments of the invention. It is apparent that the implementation of the embodiments of the present invention is not limited to the specific details familiar to those skilled in the art. The following detailed description of preferred embodiments of the invention, however, the invention is capable of other embodiments in addition to those detailed.
Embodiments of the invention are described in further detail below with reference to the accompanying drawings:
the invention provides a speed regulation method of a rotary steering motor based on a fuzzy neural network, which is based on a fuzzy neural network PID controller, and a schematic diagram of the speed regulation method is shown in figure 1.
Step S1: an offset vector between the drilled wellbore trajectory and the planned wellbore trajectory of the directional rotary steerable system is calculated.
Further, step S11: calculating the curvature k of the drilled hole of the directional rotary guiding system; the rotary steering biasing mechanism is matched with an MWD tool for use to form a rotary steering system, the device angle can be measured in real time, and the tool device angle can be controlled to be unchanged, so that the rotary steering system adopts a cylindrical spiral mode, a drilled well track belongs to a constant device angle curve, and the curvature of any point on the curve is unchanged and is equal to the average curvature of the curve;
the calculation formula of the cylindrical spiral mode borehole curvature is as follows:
Figure BDA0002660498720000031
where k is the borehole curvature, kincIs the rate of change of well deviation, kaziIs the rate of change of the well deviation orientation.
Further, step S12: calculating the device angle of the drilled borehole track of the directional rotary steering system;
Figure BDA0002660498720000032
wherein inc1、inc2Respectively is a starting point well inclination angle and a terminal point well inclination angle on the curve; azi1、azi2Respectively, the azimuth of the initial point and the azimuth of the final point on the curve.
Further, step S13: calculating a position offset vector of the rotary guide system; the position offset vector represents a geometric offset vector of the actual wellbore trajectory from the planned wellbore trajectory at a spatial location.
Establishing a wellhead coordinate system, O-XYZ (O-NEH), wherein ON is in the true north direction, OE is in the true east direction, and OH is in the gravity direction, the position and the direction of any point ON the borehole curve can be uniquely determined and represented by the position parameters (X, Y, Z), the borehole depth L, the borehole inclination angle INC and the borehole inclination azimuth AZI:
x ═ X (l), Y ═ Y (l), Z ═ Z (l), INC ═ INC (l), and az ═ aze (l).
In the actual borehole trajectory control process, an offset vector AB appears, the device angle of a tool of the rotary steering system needs to be changed, the tool is drilled along the direction of the AB, but the well depth is increased along with the increase of the tool in the drilling process, so that the actual approaching point is not the point B, but a point C deeper than the point B, and the well depth L of the point B is the point CBAnd well depth L at point CCThe spacing is not too large, as the selection is made from the design wellbore trajectory on a case-by-case basis. The position and orientation parameters of the point C are determined, the point C is projected on an offset plane to obtain C ', AC' is an actual offset vector, tAThe tangential direction of point a, the unit vector can be expressed as:
tA＝sinINCA cosAZIAi+sinINCA sinAZIAj+cosAZIAk (3)
wherein i, j, k represent unit direction vectors of X, Y, Z, respectively;
assuming that the coordinates of any point on the offset plane are (X, Y, Z), the offset plane can be expressed as:
sinINCAcosAZIA(X-XA)+sinINCAsinAZIA(Y-YA)+cosAZIA(Z-ZA)＝0 (4)
the offset vector between points a and C' is:
＝(XC'-XA)i+(YC'-YA)j+(ZC'-ZA)k (5)
the modulus of the offset vector is:
Figure BDA0002660498720000041
the setting angle of the rotary steerable system tool is offset vector AC' and wellbore height η at point AACan be expressed as:
Figure BDA0002660498720000042
AC' and ηAThe calculation principle of (2) is shown in fig. 3 and 4:
projection C' of point C on offset plane (X)C’,YC’,ZC’) The following equation is satisfied:
sinINCAcosAZIA(XC'-XA)+sinINCAsinAZIA(YC'-YA)+cosAZIA(ZC'-ZA)＝0 (8)
Figure BDA0002660498720000043
the offset vector AC' is:
AC'＝(XC'-XA)i+(YC'-YA)j+(ZC'-ZA)k (10)
the height of the well bore is defined as the height of the well boreAIs parallel to the Z axis and perpendicular to the offset plane, i.e. satisfies:
ηA＝tA×r (11)
where r is the normal plane of plane I, tAA normal plane being an offset plane;
Figure 1
wherein a is sinINCA cosAZIA，b＝sinINCA sinAZIA，c＝cosAZIA；
The combined type (7), (11) and (12):
Figure BDA0002660498720000045
wherein the position parameter and the direction parameter of the point A are known, and the position parameter of the point C can be obtained from (8) and (9).
Step S2: and carrying out PID control adjustment on the angular speed of the eccentric shaft motor of the directional rotary guiding system according to the calculated geometric deviation vector, wherein the rotating speed of the eccentric shaft motor is consistent with that of the drill collar after the tool face angle of the tool of the rotary guiding system reaches a target tool face angle.
Further, step S21: a conventional PID controller is adopted to carry out closed-loop control on a control object, and a fuzzy neural network carries out online adjustment on 3 parameters of the PID controller, thereby realizing the self-adaptive control of the control parameters of the PID controller;
the output signal is:
Figure BDA0002660498720000051
wherein Kp,Ki,KdE are respectively PID controller controlProportional coefficient, integral coefficient, derivative gain coefficient and error.
Further, step S22: establishing a fuzzy neural network, wherein the structure diagram of the fuzzy neural network is shown in FIG. 2; the fuzzy neural network consists of five layers of neural networks, namely an input layer, a membership function layer, a fuzzy inference layer, a normalization layer and an output layer;
the purpose of the input layer is to transfer the rotation speed error e and the rotation speed error change rate Δ e to the membership function layer, and the rotation speed error e and the rotation speed error change rate Δ e are transferred from the input layer to the second layer, and can be represented as follows:
yi (1)＝xi (1) i＝1,2；x1＝e；x2＝Δe (15)
the membership function layer (fuzzy layer) is mainly responsible for fuzzifying input variables, each input variable is respectively represented as a fuzzy subset { NB, NM, NS, ZO, PS, PM, PB }, wherein NB and PB respectively represent negative large and positive large, NM and PM respectively represent negative medium and middle, NS and PS respectively represent negative small and positive small, and ZO is zero; the membership function layer is used for calculating membership function of each component of the input layer relative to each fuzzy set of linguistic variable values, wherein the membership function is expressed in a Gaussian function form and is divided into 7 linguistic variables by the known values e and delta e, and the output and the input of the jth node of the layer can be expressed as follows:
Figure BDA0002660498720000052
wherein
Figure BDA0002660498720000053
σijMean and standard deviation of membership functions, y, of the jth fuzzy set of i input variables of the second layer, respectivelyij (2)The jth neuron output of the layer for the ith input variable, T1 (j)Representing membership functions, T, under e input variables2 (j)Representing a membership function under a delta e input variable;
each node in the fuzzy inference layer represents one fuzzy rule in a known fuzzy rule base, the fuzzy inference layer is mainly used for determining the matching condition of the fuzzy rules, the number of the nodes in the fuzzy inference layer is 49 according to the number of the fuzzy rules, and the fitness of each fuzzy rule is calculated;
the fitness of the fuzzy logic operation rule corresponding to each node is as follows:
ykj1j2 (3)＝T1 (j1)·T2 (j2) j1，j2＝1,2,3...7；k＝12.....49 (17)
the number of nodes of the normalization layer is the same as that of nodes of the fuzzy inference layer, and the normalization operation is realized by the normalization layer; calculating the applicability of each fuzzy control rule:
Figure BDA0002660498720000054
Figure BDA0002660498720000055
the output layer is the proportional coefficient K of three indexes of the PID controllerPIntegral coefficient KiDifferential gain coefficient KdPerforming a sharpening operation as an output, wherein the output quantity is the sum of the processed normalization layer output signals;
Figure BDA0002660498720000056
kp＝y1 (5)，ki＝y2 (5)，kd＝y3 (5) (21)
wherein, wijIs the connection weight between the normalization layer and the output layer.
Further, step S23: optimizing structural parameters of the fuzzy neural network controller; parameters of fuzzy neural network controller
Figure BDA0002660498720000061
wij，σijContinuously adjusting parameters by a random gradient descent training method to realize multivariable nonlinear accurate control on a real-time link and a time-lag link of a rotary steering tool surface;
the Particle Swarm Optimization (PSO) algorithm is a bionic-based cluster optimization algorithm, the PSO algorithm is derived from the simulation and research of predation behaviors of bird swarms, in the PSO algorithm, each optimization problem solution is a bird in a search space and is called as a 'particle', namely the position of each particle is a potential solution; iterating the standard PSO algorithm according to the formulas (22) and (23) until the optimal position meeting the minimum error requirement in the particle swarm is searched;
Vid＝ω·Vid+c1·r1(pid-xid)+c2·r2(pgd-xid) (22)
xid +＝Vid+xid (23)
wherein, omega is an inertia weight factor (the value is not negative), and the classical value is omega0＝0.9,ωend0.4; c1 and c2 are called learning constants (the values of which are also non-negative); r1 and r2 are random numbers with a value range of [0, l]；pidDimension d of the individual extreme value of the ith variable; p is a radical ofgdIs the d-dimension, x, of the global extremum of the i-th variableid +Is the particle position, x, at the next momentidThe particle position at the current moment.
Further, as shown in fig. 5, the specific process of step S23 is as follows:
step S231, initialization; the particle swarm initialization comprises the steps of setting the particle number h, the initial position of the particle, the velocity and the inertia factor omegamaxAnd ωminAcceleration factors c1 and c2, maximum number of iterations MpEtc.;
step S232, adopting a particle swarm algorithm to improve the optimal solution obtained in the step S23; firstly, calculating the fitness of the particles according to an equation (18), and updating the running speed and the positions of the particles in time according to equations (20) and (21); suppose that the best position P of the currently found particle isbestCannot be better than the current position P of the particlenewThen order Pbest＝Pnew(ii) a If the current search is being conductedGlobal best position GbestCannot be better than the current position P of the particlenowThen order Gbest＝Pnow(ii) a If the maximum iteration times of the particle swarm are not reached, the current position and the optimal position P of the particle are continuously adjustedbestAnd GbestPerforming comparison execution;
step S233: increasing the cycle number by 1, and obtaining the most individual with improved current optimal solution;
step S234: and if the evolution meets the maximum iteration times, stopping the iteration process, and obtaining the optimal solution of the structural parameters of the fuzzy neural network, otherwise, jumping to the step S232 to perform circular execution.
As shown in FIG. 6, compared with the optimization of the traditional fuzzy control PID algorithm, the dynamic response and the static characteristic of the control system of the fuzzy neural network controller are obviously superior to those of the traditional fuzzy neural network. The system has the advantages of quick response of the rotating speed, small fluctuation of the rotating speed, quick and automatic adjustment of disturbance energy of an external load, strong anti-interference capability and higher control precision of the system after steady state in network control. Therefore, the fuzzy neural network controller optimized by the ant colony algorithm has better dynamic characteristics and robustness in terms of the starting, speed tracking and load disturbance conditions of the ultrasonic motor.
In summary, the present invention is not limited to the above-mentioned embodiments, and those skilled in the art can propose other embodiments within the technical teaching of the present invention, but these embodiments are included in the scope of the present invention.

Claims (4)

1. A speed regulation method of a rotary steering motor based on a fuzzy neural network is characterized by comprising the following steps:
step S1, calculating a geometric offset vector between the drilled well track and the designed well track of the directional rotary steering system;
and step S2, carrying out PID control adjustment on the angular speed of the eccentric shaft motor of the directional rotary guide system according to the calculated geometric deviation vector.
2. The method for regulating the speed of the rotary steering motor based on the fuzzy neural network as claimed in claim 1, wherein the specific process of the step S1 is as follows:
step S11, calculating the curvature k of the drilled borehole of the directional rotary guiding system;
step S12, calculating the device angle TFA of the drilled borehole trajectory of the directional rotary steerable systempath；
Step S13, calculating the position offset vector AC of the rotary guiding system’And tool face TFA of rotary steerable system toolRSS。
3. The method for regulating the speed of the rotary steering motor based on the fuzzy neural network as claimed in claim 1, wherein the specific process of the step S2 is as follows:
s21, performing closed-loop control on a control object by adopting a conventional PID controller, and performing online adjustment on 3 parameters of the PID controller by using a fuzzy neural network so as to realize the self-adaptive control of the control parameters of the PID controller;
step S22, establishing a fuzzy neural network; the fuzzy neural network comprises an input layer, a membership function layer, a fuzzy inference layer, a normalization layer and an output layer; wherein the content of the first and second substances,
the input layer is used for transmitting the rotating speed error e and the rotating speed error change rate delta e of the rotating speed of the eccentric shaft motor to the membership function layer;
the membership function layer is used for fuzzifying the input variable and calculating membership functions of all components of the input layer relative to all language variable value fuzzy sets;
each node in the fuzzy inference layer represents a fuzzy rule in a known fuzzy rule base, and the layer is used for determining the matching condition of the fuzzy rules and calculating the fitness of each fuzzy rule;
the number of nodes of the normalization layer is the same as that of the nodes of the fuzzy inference layer, and the normalization layer is used for realizing normalization operation;
the output layer is the proportional coefficient K of three indexes of the PID controllerPIntegral coefficient KiDifferential gain coefficient KdPerforming a sharpening operation as an output;
s23, optimizing the structural parameters of the fuzzy neural network; the mean value, the standard deviation and the network connection weight of the parameter membership Gaussian function of the fuzzy neural network are continuously adjusted by a random gradient descending training method, so that the nonlinear accurate control of a real-time link and a time-delay link of a tool surface of a rotary steering system tool is realized.
4. The method for regulating the speed of the rotary steering motor based on the fuzzy neural network as claimed in claim 3, wherein the specific process of the step S23 is as follows:
step S231, initialization; the particle swarm initialization comprises the steps of setting particle numbers, particle initial positions, speeds, inertia factors, acceleration factors and maximum iteration times;
step S232, improving the optimal solution of the structural parameters of the fuzzy neural network by adopting a particle swarm algorithm;
step S233, increasing the cycle number by 1; if the optimal solution condition is met, obtaining an improved optimal individual;
and S234, if the evolution meets the maximum iteration times, stopping the iteration process, and obtaining the optimal solution of the structural parameters of the fuzzy neural network, otherwise, jumping to the step S232 to perform circular execution.
CN202010903284.0A 2020-09-01 2020-09-01 Rotating guide motor speed regulation method based on fuzzy neural network Pending CN111997583A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010903284.0A CN111997583A (en) 2020-09-01 2020-09-01 Rotating guide motor speed regulation method based on fuzzy neural network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010903284.0A CN111997583A (en) 2020-09-01 2020-09-01 Rotating guide motor speed regulation method based on fuzzy neural network

Publications (1)

Publication Number Publication Date
CN111997583A true CN111997583A (en) 2020-11-27

Family

ID=73464962

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010903284.0A Pending CN111997583A (en) 2020-09-01 2020-09-01 Rotating guide motor speed regulation method based on fuzzy neural network

Country Status (1)

Country Link
CN (1) CN111997583A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113890411A (en) * 2021-09-28 2022-01-04 哈尔滨工业大学 Multi-mode combined control method and system of standing wave linear ultrasonic motor based on Koopman state observer
CN114215501A (en) * 2022-02-23 2022-03-22 西南石油大学 Control method for stable platform in rotary steering system

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104350233A (en) * 2012-04-05 2015-02-11 普拉德研究及开发股份有限公司 Formation volumetric evaluation using normalized differential data
CN106640033A (en) * 2015-10-30 2017-05-10 中石化石油工程技术服务有限公司 State monitoring method for rotary guiding tool
WO2017106807A1 (en) * 2015-12-18 2017-06-22 Schlumberger Technology Corporation Imaging subterranean formations and features using multicoil nmr measurements
CN108474249A (en) * 2015-12-18 2018-08-31 斯伦贝谢技术有限公司 Subsurface formations and feature are imaged using azimuth phase code
CN109477376A (en) * 2016-06-03 2019-03-15 吉奥奎斯特系统公司 Detect the event in drilling well report
CN111279050A (en) * 2017-09-11 2020-06-12 吉奥奎斯特系统公司 Well planning system

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104350233A (en) * 2012-04-05 2015-02-11 普拉德研究及开发股份有限公司 Formation volumetric evaluation using normalized differential data
CN106640033A (en) * 2015-10-30 2017-05-10 中石化石油工程技术服务有限公司 State monitoring method for rotary guiding tool
WO2017106807A1 (en) * 2015-12-18 2017-06-22 Schlumberger Technology Corporation Imaging subterranean formations and features using multicoil nmr measurements
CN108474249A (en) * 2015-12-18 2018-08-31 斯伦贝谢技术有限公司 Subsurface formations and feature are imaged using azimuth phase code
CN109477376A (en) * 2016-06-03 2019-03-15 吉奥奎斯特系统公司 Detect the event in drilling well report
CN111279050A (en) * 2017-09-11 2020-06-12 吉奥奎斯特系统公司 Well planning system

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
田欢: "一种导向钻井偏心矢量智能控制方法的研究", 《中国优秀博硕士学位论文全文数据库（硕士）工程科技Ⅰ辑》 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113890411A (en) * 2021-09-28 2022-01-04 哈尔滨工业大学 Multi-mode combined control method and system of standing wave linear ultrasonic motor based on Koopman state observer
CN113890411B (en) * 2021-09-28 2022-06-14 哈尔滨工业大学 Multi-mode combined control method and system of standing wave linear ultrasonic motor based on Koopman state observer
CN114215501A (en) * 2022-02-23 2022-03-22 西南石油大学 Control method for stable platform in rotary steering system

Similar Documents

Publication Publication Date Title
CN111997583A (en) Rotating guide motor speed regulation method based on fuzzy neural network
CN109884900B (en) Design method of harvester path tracking controller based on adaptive model predictive control
CN108001447A (en) A kind of intelligent vehicle path trace front wheel angle compensating control method
CN110134006B (en) Complex borehole trajectory optimization method based on improved multi-target particle swarm optimization
CN114510063B (en) Unmanned tracked vehicle and track tracking control method and system thereof
CN108227495B (en) Self-adaptive sliding guide drilling control system and control method
CN108876065A (en) A kind of unmanned boat motion planning method based on trajectory unit
CN109683624A (en) Nonlinear robust control method for small-sized depopulated helicopter gesture stability
CN111608868A (en) Maximum power tracking adaptive robust control system and method for wind power generation system
US6101444A (en) Numerical control unit for wellbore drilling
Xia et al. Adaptive energy-efficient tracking control of a X rudder AUV with actuator dynamics and rolling restriction
CN101122777A (en) Large condenser underwater operation environment two-joint robot control method
CN110262513B (en) Design method of marine robot trajectory tracking control structure
CN115236973A (en) AGV trajectory tracking control method based on PSO Lyapunov function
CN113093526A (en) Overshoot-free PID controller parameter setting method based on reinforcement learning
CN109325288A (en) A kind of Solid Launch Vehicle population parameter based on uncertainty optimization determines method and system
CN116520883A (en) Method for planning homing track of parafoil system in complex airspace environment
CN107065537A (en) A kind of horizontal low speed that pushes away of AUV is without the motion control method in the case of output
CN112431585A (en) Automatic bit feeding method based on deep belief network
Li et al. Research on Improved Particle Swarm Optimized Fuzzy PID Control for Quadrotor UAV
CN117950398A (en) Unmanned ship planning method based on speed obstacle and fuzzy parameters
Cai et al. Trajectory Azimuth Control Based on Equivalent Input Disturbance Approach for Directional Drilling Process
Gao et al. Real Time Trajectory Parameters Measurement on Machine Learning for Intelligent Drilling
Ji et al. A Sliding Mode Control Based Stabilization Method for Directional Rotary Steering Tool-Face
CN111287659B (en) Build-up rate adjusting method based on full-rotation directional type guiding drilling tool

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: 20201127

RJ01 Rejection of invention patent application after publication