CN115473464A - Magnetic suspension yaw motor control method based on neural network model predictive control - Google Patents

Magnetic suspension yaw motor control method based on neural network model predictive control Download PDF

Info

Publication number
CN115473464A
CN115473464A CN202211137280.1A CN202211137280A CN115473464A CN 115473464 A CN115473464 A CN 115473464A CN 202211137280 A CN202211137280 A CN 202211137280A CN 115473464 A CN115473464 A CN 115473464A
Authority
CN
China
Prior art keywords
suspension
yaw
rotor
neural network
network model
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
CN202211137280.1A
Other languages
Chinese (zh)
Inventor
高永新
吴子文
侯利宏
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.)
Liaoning Technical University
Original Assignee
Liaoning Technical University
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 Liaoning Technical University filed Critical Liaoning Technical University
Priority to CN202211137280.1A priority Critical patent/CN115473464A/en
Publication of CN115473464A publication Critical patent/CN115473464A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
    • H02P21/00—Arrangements or methods for the control of electric machines by vector control, e.g. by control of field orientation
    • H02P21/0003—Control strategies in general, e.g. linear type, e.g. P, PI, PID, using robust control
    • H02P21/0014—Control strategies in general, e.g. linear type, e.g. P, PI, PID, using robust control using neural networks
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
    • H02P21/00—Arrangements or methods for the control of electric machines by vector control, e.g. by control of field orientation
    • H02P21/0003—Control strategies in general, e.g. linear type, e.g. P, PI, PID, using robust control
    • H02P21/0021—Control strategies in general, e.g. linear type, e.g. P, PI, PID, using robust control using different modes of control depending on a parameter, e.g. the speed
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E10/00—Energy generation through renewable energy sources
    • Y02E10/70—Wind energy
    • Y02E10/72—Wind turbines with rotation axis in wind direction

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Control Of Electric Motors In General (AREA)

Abstract

The invention discloses a magnetic suspension yaw motor control method based on neural network model predictive control, which comprises the steps that when the wind direction changes and needs yaw, a rotor converter controls the current of a rotor by adopting a PID control algorithm, so that the rotor of a magnetic suspension yaw motor is suspended upwards and is kept at a suspension balance point to realize stable suspension; controlling the current of the rotor to ensure that the rotor of the magnetic suspension yaw motor keeps stable suspension at a suspension balance point; rotating the magnetic suspension yaw motor to a wind facing position according to the expected rotating speed value omega; and when yawing, the rotor converter adopts a fuzzy neural network model prediction control strategy to control the current of the rotor, so that the rotor of the magnetic suspension yawing motor is kept at a suspension balance point in the whole yawing process. The invention realizes the cooperative control of the stator and the rotor by adopting a neural network model predictive control strategy, and ensures the real-time optimal system performance in the whole suspension yaw rotation process of the magnetic suspension yaw motor.

Description

Magnetic suspension yaw motor control method based on neural network model predictive control
Technical Field
The invention belongs to the technical field of electrical engineering, and particularly relates to a magnetic suspension yaw motor control method based on neural network model predictive control.
Background
The yaw system is an essential important component of a horizontal axis wind turbine, the existing large and medium horizontal axis wind turbines adopt a gear-driven yaw system, and have the defects of complex structure, multi-motor drive, large occupied space, high failure rate, inconvenience in maintenance (lubrication is needed, lubricating oil and lubricating grease must be replaced regularly) and the like.
The wind power magnetic levitation yaw system adopts a magnetic levitation driving technology to replace the traditional gear driving technology, and has the advantages of high wind precision, no need of lubrication, simple structure, convenience in maintenance, short power failure time, low operation and maintenance cost and the like. Through electromagnetic suspension, on one hand, the engine room is in a suspension state, and accurate wind alignment can be realized; on the other hand, the structure of the yawing system is simplified, the maintenance is simple and convenient, and the downtime can be greatly shortened.
The most key part in the wind power magnetic suspension yaw system is a magnetic suspension yaw motor, and the working principle is as follows: when the wind direction changes, firstly, direct current is introduced into the rotor of the wind power generator to realize suspension, after the suspension reaches a suspension balance point, three-phase alternating current is introduced into the stator of the wind power generator, and the rotor starts to rotate until the wind power generator reaches a wind facing position. During the rotation, the suspension control is carried out to make the suspension control be at the balance point, and the rotating speed is controlled to realize stable rotation, so that the stator and the rotor must be cooperatively controlled.
However, the magnetic levitation technology has the characteristics of high nonlinearity, strong coupling and intrinsic instability, so that the stability control is very challenging, and the current research is mostly focused on the levitation control in the fields of magnetic levitation trains, magnetic levitation bearings, magnetic levitation planar motors and the like. The linear state feedback control adopts the most suspension control strategies, but adopts a Taylor linearization method to linearize a system model at a balance point, so that the state feedback control is completed, and the robustness to air gap change is poor; some sliding mode control methods are adopted to realize robust control of the suspension system, but the application of the sliding mode control methods is still to be perfected due to the inherent buffeting problem. Aiming at the problem that Taylor linearization neglects high-order dynamic, a self-adaptive method is adopted to realize the stable control of a suspension body; some methods adopt feedback linearization and state feedback to realize suspension control, but are influenced by system parameter perturbation. In other documents, the H-infinity control is applied to the suspension system control, which improves the robustness of the system to air gap or suspension mass change, but has the defect of higher order of the controller.
Aiming at the characteristics of high nonlinearity, strong coupling and unstable essence of the magnetic suspension technology, the scheme of introducing the fuzzy neural network control theory into the yaw control system firstly performs fuzzification processing on input information, formulates a fuzzy control rule, designs a fuzzy controller and can obtain a good control effect. However, the extraction of fuzzy rules and the optimization of membership functions in fuzzy control are difficult problems which plague the fuzzy information processing technology. Fuzzy control rules and membership functions obtained by using the experience of experts have great uncertainty, so that ideal control effect cannot be obtained when the control strategy is directly applied to control.
The problems can be effectively solved by utilizing the self-learning characteristic of the neural network to extract the control rule and optimize the membership function. The study of people on the neural network starts from the beginning of the 40 th 20 th century, and along with the continuous deepening of the study of the artificial neural network, various types of neural network models appear, and the neural networks have strong self-adaption and self-learning capabilities and also have a series of advantages of parallel computing, distributed information storage, strong fault-tolerant capability and the like. The network learns the analysis and processing of the input vectors through training, and can perform functions such as calculation, memory, recognition and the like.
The fuzzy neural network inherits the respective advantages of the neural network and the fuzzy control, has high processing speed like the neural network, has self-learning capability and fuzzy logic like the fuzzy control, and has more advantages than other control modes when facing nonlinear problems. Generally, in a fuzzy neural network, input and output nodes of the neural network not only represent input and output of a fuzzy logic part, but also can be used for representing membership functions and fuzzy rules, and the parallel processing capability of the neural network can greatly improve the reasoning capability of a fuzzy system.
Disclosure of Invention
Based on the defects of the prior art, the technical problem to be solved by the invention is to provide a magnetic levitation yaw motor control method based on neural network model prediction control, and the cooperative control of a stator and a rotor is realized by adopting a neural network model prediction control strategy, so that the real-time optimal system performance of the magnetic levitation yaw motor in the whole levitation yaw rotation process is ensured.
In order to solve the technical problems, the invention is realized by the following technical scheme: the invention provides a magnetic suspension yaw motor control method based on neural network model predictive control, which comprises the following steps:
step 1, when the wind direction changes and needs to be navigated, a rotor converter controls the current of a rotor by adopting a PID control algorithm, so that the rotor of a magnetic suspension yaw motor is suspended upwards to and kept at a suspension balance point to realize stable suspension;
step 2, after stable suspension is realized, the rotor converter uses a fuzzy neural network model prediction control strategy instead to control the rotor current, so that the rotor of the magnetic suspension yaw motor keeps stable suspension at a suspension balance point;
step 3, the stator converter adopts a fuzzy neural network model prediction control strategy to control the current of the stator, so that the magnetic suspension yaw motor rotates to a wind facing position according to a rotating speed expected value omega;
and 4, while yawing, controlling the current of the rotor by using a fuzzy neural network model prediction control strategy through a rotor converter, so that the rotor of the magnetic suspension yawing motor is kept at a suspension balance point in the whole yawing process.
Further, the specific method of step 2 is:
21 According to a suspension dynamic mathematical model of the magnetic suspension yaw motor, training a suspension fuzzy neural network model;
22 Transplanting the trained suspension fuzzy neural network model into a main control chip of the rotor converter, and establishing an actual suspension fuzzy neural network model prediction control system based on the main control chip of the rotor converter;
23 Input the response output value δ m and the suspension air gap expected value δ x of the suspension fuzzy neural network model into a suspension nonlinear optimization module which determines the response output value δ m and the suspension air gap expected value δ x by minimizing a suspension cost functionDetermining an optimal control input signal, i.e. an optimal rotor current i r_opt Will optimize the rotor current i r_opt And the suspension air gap measured value delta is used as the input of a suspension fuzzy neural network model, and the optimal rotor current i is simultaneously used r_opt With the actual rotor current i r Making difference, sending the difference to PWM module via PID controller to generate drive signal of rotor converter so as to control rotor current i r And the rotor of the magnetic suspension yaw motor keeps stable suspension at the suspension balance point.
Further, the specific method of step 3 is:
31 According to a yaw dynamic mathematical model of the magnetic suspension yaw motor, training a yaw fuzzy neural network model;
32 Transplanting the trained yaw fuzzy neural network model into a main control chip of the stator converter, and establishing an actual yaw fuzzy neural network model prediction control system based on the main control chip of the stator converter;
33 Input the response output value of the yaw fuzzy neural network model and the expected value omega of the rotating speed into a yaw nonlinear optimization module, and the yaw nonlinear optimization module determines the optimal control input signal, namely the d-axis component i of the optimal stator current by minimizing a yaw cost function sd_opt And q-axis component i sq_opt Optimizing the d-axis component i of the stator current sd_opt Q-axis component i sq_opt And the rotation speed measured value omega is used as the input of the yaw fuzzy neural network model, and the d-axis component i of the optimal stator current is simultaneously used sd_opt And q-axis component i sq_opt Respectively making difference with respective actual measured value, inputting them into PI controller with amplitude limit to obtain stator voltage control quantity u sd * And u sq * U is obtained after dq/alpha beta coordinate transformation sα * And u sβ * And the SVPWM module modulates the signal to generate a driving signal, and controls the stator converter to generate required excitation voltage and current so that the magnetic suspension yaw motor rotates to a wind facing position according to a rotating speed expected value omega.
Further, the specific method of step 4 is:
41 Training a yaw suspension fuzzy neural network model according to a yaw suspension dynamic mathematical model of a magnetic suspension yaw motor;
42 Transplanting the trained yaw suspension fuzzy neural network model into a main control chip of the rotor converter, and establishing an actual yaw suspension fuzzy neural network model prediction control system based on the main control chip of the rotor converter;
43 Input the response output value of the yaw fuzzy neural network model and the expected value delta of the levitation air gap into a levitation nonlinear optimization module, and the yaw levitation nonlinear optimization module determines the optimal control input signal, namely the optimal rotor current i by minimizing a yaw levitation cost function yr_opt Will optimize the rotor current i yr_opt And taking the suspension air gap measurement value delta as an input of a yaw suspension fuzzy neural network model, and simultaneously taking the optimal rotor current i yr_opt With the actual rotor current i r Making difference, sending the difference to PWM module via PID controller to generate drive signal of rotor converter so as to control rotor current i r And the rotor of the magnetic suspension yaw motor is kept at a suspension balance point in the whole yaw process.
Further, the suspension cost function is:
Figure BDA0003851921700000051
wherein J is a suspension cost function, N p To predict the time domain length, N u To control the time domain length, i r For the control input signal, i.e. the rotor current, δ x is the desired value of the suspended air gap, δ m is the response output value of the yaw fuzzy neural network model, ρ is the weighting coefficient, and k is the current time.
Further, the yaw cost function is:
Figure BDA0003851921700000052
in the formula, J Y As a yaw cost function, N p To predict the time domain length, N u To control the time domain length, i sd 、i sq For the control input signals, i.e. the d-axis component and the q-axis component of the stator current, respectively, ω is the desired value of the rotational speed, δ m For the response output value, p, of the yaw fuzzy neural network model 1 、ρ 2 K is the current time, which is the weighting factor.
Further, the yaw levitation cost function is:
Figure BDA0003851921700000061
in the formula, J YM As a yaw levitation cost function, N p To predict the time domain length, N u To control the time domain length, i r For the control input signal, i.e. the rotor current, δ is the desired value of the floating air gap, δ m Output value, rho, for response of yaw fuzzy neural network model 3 K is the current time, which is the weighting factor.
Therefore, the magnetic suspension yaw motor control method based on the neural network model prediction control combines the fuzzy control and the neural network, utilizes the fuzzy control, efficiently integrates the experience knowledge of experts, does not need an accurate mathematical model, controls the wind power generation yaw system, and has better dynamic performance and robustness. And (3) performing real-time stable control on the suspension and yaw processes of the magnetic suspension yaw motor by adopting a fuzzy neural network model prediction control strategy: when the wind direction changes and needs to be deflected, firstly, a rotor converter adopts a fuzzy neural network algorithm to control the current of a rotor, so that the rotor is suspended upwards to and kept at a suspension balance point; secondly, controlling the rotor current by a fuzzy neural network model predictive control strategy for a rotor current transformer, so that the rotor keeps stable suspension at a balance point; and then the stator current is controlled by the stator converter by adopting a fuzzy neural network model predictive control strategy, so that the yaw motor rotates to a wind-facing position according to a specified rotating speed, and meanwhile, the rotor current is controlled by the rotor converter by adopting the fuzzy neural network model predictive control strategy, so that the rotor is kept at a balance point in the yaw process, the optimal control is realized, and the real-time optimal performance of the system in the whole suspension yaw process is ensured.
The foregoing description is only an overview of the technical solutions of the present invention, and in order to make the technical means of the present invention more clearly understood, the present invention may be implemented in accordance with the content of the description, and in order to make the above and other objects, features, and advantages of the present invention more concise and understandable, the following detailed description is given with reference to the preferred embodiments and the accompanying drawings.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the prior art descriptions will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a schematic structural diagram of a magnetic levitation yaw motor according to the present invention;
FIG. 2 is a schematic view of the magnetic levitation yaw motor stress analysis of the present invention;
FIG. 3 is an electrical connection diagram of a magnetic levitation yaw motor of the present invention;
FIG. 4 is a schematic structural diagram of a wind power magnetic levitation yaw system according to the present invention;
FIG. 5 is a model structure of the levitation and yaw levitation fuzzy neural network of the present invention;
FIG. 6 is a diagram of a yaw neural network model predictive control process of the present invention;
FIG. 7 is a model structure of the fuzzy yawing neural network of the present invention.
Reference numbers in the figures: the method comprises the following steps of 1-stator, 2-rotor, 3-disc, 4-tower, 5-guide bearing, 6-suspension frame, 7-load platform, 8-air gap sensor, 10-stator converter, 11-stator three-phase winding, 20-rotor converter, 21-rotor direct current excitation winding and 22-rotor iron core.
Detailed Description
The following describes in detail a specific embodiment of the magnetic levitation yaw motor control method based on neural network model predictive control according to the present invention with reference to the accompanying drawings.
As shown in fig. 1-4, the invention is used for a wind power magnetic suspension yaw motor, which is a non-salient pole type disc synchronous motor and comprises a stator 1, a rotor 2, a disc 3, a tower 4, a guide bearing 5, a suspension bracket 6, a load platform 7 and an air gap sensor 8. The stator 1 and the rotor 2 are oppositely and concentrically arranged; the stator 1 is fixed with the disc 3, and the disc 3 is sleeved on the tower 4 and fixed with the tower 4; the rotor 2 is fixed with the suspension frame 6; the suspension frame 6 is also fixed with a load platform 7; the guide bearing 5 is fixed with the load platform 7, fixed at the center of the load platform 7 and fixed with the tower 4; the load platform 7 is fixed with an engine room of the wind turbine generator; the air gap sensor 8 is fixed with the rotor 2; the rotor 2, the suspension 6, the load platform 7, the air gap sensor 8 and the nacelle of the wind turbine are collectively referred to as a suspension or suspension system.
As shown in fig. 1,2 and 3, the stator 1 includes a stator three-phase winding and a disc stator core, and the rotor 2 includes a rotor direct-current excitation winding 21 and a disc rotor core 22; the stator three-phase winding 11 is connected with the stator converter 10; the rotor dc excitation winding 21 is connected to the rotor converter 20. The stator converter 10 is a three-phase alternating current inverter, the main control chip of the stator converter is a DSP, and the frequency is controllable; the rotor converter 20 is a DC/DC chopper, such as a Buck circuit or an H-bridge circuit, and its main control chip is a DSP.
As shown in fig. 4, the yaw system adopts the magnetic levitation yaw motor shown in fig. 1, the rotor 2 drives the nacelle to levitate and rotate through the levitation frame 6, and the air gap sensor 8 is used for detecting a levitation air gap δ between the stator 1 and the rotor 2.
The working principle of the wind power magnetic suspension yaw system is as follows:
as shown in fig. 2, 3 and 4, the length of the air gap between the stator 1 and the rotor 2 is δ, when the wind direction changes and needs to yaw (or untwist), the rotor converter 20 supplies a direct current to the rotor direct current excitation winding 21, a main magnetic field passing through the air gap between the rotor 2 and the stator 1 and simultaneously linking the stator 1 and the rotor 2 is generated, the stator 1 is magnetized to generate an axial magnetic attraction force for dealing with the upward movement, the rotor 2 moves upward to suspend the whole nacelle upward through the suspension frame 6, and when the air gap sensor 8 detects that the air gap δ reaches a set value δ 0 (e.g. 10 mm) for adjusting rotor DC field winding 21Exciting current to balance the attraction force and the gravity of the engine room, and suspending the engine room in the air, which is called as a suspension balance point; then, the stator converter 10 outputs three-phase alternating current to supply power to the stator three-phase winding 11, a rotating magnetic field is formed in an air gap between the stator 1 and the rotor 2, and an air gap synthetic magnetic field is formed after the rotating magnetic field is coupled with the rotor 2 magnetic field. Under the action of the magnetic moment of the synthetic magnetic field, the rotor 2 drives the engine room to rotate through the suspension frame 6, and yawing is achieved. During the yaw rotation, the guide bearing 5 can ensure that the nacelle does not deviate, so that the load can normally rotate along the track.
Therefore, the magnetic suspension yaw motor performs yaw rotation at a suspension balance point, and in the process, on one hand, the rotating speed of the magnetic suspension yaw motor is controlled to meet a motion equation, and meanwhile, the suspension air gap of the magnetic suspension yaw motor is controlled to be constant, so that the magnetic suspension yaw motor rotates at the balance point to realize yaw, and therefore, the stator and the rotor must be cooperatively controlled.
For this purpose, as shown in fig. 1 to 7, the magnetic levitation yaw motor control method based on neural network model prediction control of the present invention includes the following steps:
step 1, when the wind direction changes and needs to be navigated, a rotor converter controls the current of a rotor by adopting a PID control algorithm, so that the rotor of a magnetic suspension yaw motor is suspended upwards to and kept at a suspension balance point to realize stable suspension;
step 2, after stable suspension is realized, the rotor converter is changed to a fuzzy neural network model prediction control strategy to control the rotor current, so that the rotor of the magnetic suspension yaw motor keeps stable suspension at a suspension balance point, and the specific method comprises the following steps:
21 Training a suspension fuzzy neural network model according to a suspension dynamic mathematical model of the magnetic suspension yaw motor;
first, a suspended neural network model is constructed, which predicts future output values of the neural network using current inputs and current outputs.
(1) According to the knowledge and experience of known wind driven generator yaw control, a fuzzy control system in the traditional sense is preliminarily determined;
(2) Determining a connection mode and a connection weight of a neural network according to a membership function and a fuzzy control rule of the fuzzy control system in the traditional sense;
(3) Summarizing and inducing actual experience of an operator to obtain a yaw fuzzy control rule table of the wind driven generator, converting the table into a fuzzy query table which is suitable for the corresponding relation between input quantity and output quantity of a controlled object to obtain a learning sample of fuzzy control FNN based on a neural network, performing offline training on the fuzzy control FNN based on the neural network by using an error back propagation BP algorithm, and determining the following parameters according to a learning result:
center values of normally distributed membership functions are used: a is a ij ；
Width value of membership function using normal distribution: b ij ；
Corresponding to a corresponding control rule in a fuzzy lookup table in fuzzy control, and the maximum membership degree of a fuzzy word set: w is a pq Adjusting membership function of each linguistic value and memorizing fuzzy control rule; the fuzzy neural network magnetic suspension yaw control calculation steps are as follows:
inputting a training period sample, setting an initial error, and fuzzifying an input quantity, namely:
Figure BDA0003851921700000101
Figure BDA0003851921700000102
wherein, A ij (x i ) When the input is x i The corresponding degree of membership, a ij ，b ij Expressing the center value and the width of the membership function in a normal distribution, adjusting the center value a ij The distribution of the membership function on the domain can be changed, and the width b can be adjusted ij The shape can be changed, i =1,2, for the input variable index, j =1,2.
And (3) carrying out fuzzy reasoning by adopting a composite reasoning method, wherein:
Figure BDA0003851921700000103
Figure BDA0003851921700000104
(3) Fuzzification is carried out by using a gravity center method, wherein:
Figure BDA0003851921700000105
Figure BDA0003851921700000106
where N is the number of linguistic values contained in each input linguistic variable, w pq Is the conclusion language value w pq The maximum degree of membership.
(4) Selecting an objective function
Figure BDA0003851921700000107
Calculating a total average error, when the total average error is larger than a required value, if the total average error is increased, reducing the learning rate, if the total average error is reduced, increasing the learning rate, performing offline training by using a directional multilayer back propagation BP algorithm to obtain a new membership function parameter, and then performing the above steps on the basis of the new membership function on the training period sample until the total average error is smaller than or equal to the required value;
when the total average error is less than or equal to the required value, whether a new training period sample exists or not is judged, and if not, the process is ended; if yes, inputting a next training period sample, and repeating the steps until the total average error of all training period samples is less than or equal to the required value.
The suspension dynamic mathematical model of the magnetic suspension yaw motor is obtained through the following processes:
as shown in FIG. 2, when the DC excitation winding 21 of the rotor of the magnetic levitation yaw motor is electrified, an upward axial levitation suction force F (i) is generated r δ) is:
Figure BDA0003851921700000111
in the formula i r Delta is the length of the air gap between the rotor 2 and the stator 1, k, for the rotor current 1 ＝μ 0 N 2 S/4, wherein 0 The magnetic field is vacuum magnetic conductivity, N is the number of turns of the rotor direct current excitation winding 21, and S is the effective area of the magnetic pole surface of the rotor core 22;
the suspension is subjected to an upward suspension suction force F (i) in the axial direction r Delta), downward suspension weight mg and external disturbance force f d (t), from which the mechanical equation in the vertical direction can be derived as:
Figure BDA0003851921700000112
in the formula, m is the mass of suspended matters, and g is the gravity acceleration;
the voltage equation of the rotor dc excitation winding 21 is:
Figure BDA0003851921700000113
in the formula u r Is the input voltage, R, of the rotor DC field winding 21 r Is the resistance, psi, of the rotor DC field winding 21 r For rotor flux linkage, L r Is the inductance of the rotor DC excitation winding 21 and has L r ＝2k 1 A/δ; d δ/dt is the first derivative of the suspension air gap δ with respect to time t, i.e. the axial velocity of the suspension.
In conclusion, a suspension dynamic mathematical model of the magnetic suspension yaw motor can be obtained:
Figure BDA0003851921700000114
22 Transplanting the trained suspension fuzzy neural network model into a main control chip of the rotor converter, and establishing an actual suspension fuzzy neural network model prediction control system based on the main control chip of the rotor converter;
23 Input the response output value delta m and the suspension air gap expected value delta of the suspension fuzzy neural network model into a suspension nonlinear optimization module, and the suspension nonlinear optimization module determines the optimal control input signal, namely the optimal rotor current i by minimizing a suspension cost function r_opt Will optimize the rotor current i r_opt And the suspension air gap measured value delta is used as the input of a suspension fuzzy neural network model, and the optimal rotor current i is simultaneously used r_opt With the actual rotor current i r Making difference, sending the difference to PWM module via PID controller to generate drive signal of rotor converter, and controlling rotor current i r Keeping the rotor of the magnetic suspension yaw motor in stable suspension at the suspension balance point;
the suspension cost function is as follows:
Figure BDA0003851921700000121
wherein J is a suspension cost function, N p To predict the time domain length, N u To control the time domain length, i r For control of the input signal, i.e. rotor current, δ is the desired value of the floating air gap, δ m And p is a weighting coefficient, and k is the current moment.
And 3, controlling the current of the stator by adopting a fuzzy neural network model prediction control strategy through the stator converter, and enabling the magnetic suspension yaw motor to rotate to a wind facing position according to a rotating speed expected value omega, wherein the specific method comprises the following steps:
31 According to a yaw dynamic mathematical model of the magnetic suspension yaw motor, training a yaw fuzzy neural network model;
first, a yaw neural network model is constructed, which predicts future output values of the neural network using current inputs and current outputs.
The first layer of neurons is an input layer, and the input is the deviation and the deviation change rate of the theoretical rotating angle and the actual rotating angle of the wind power generation yaw system. This layer only passes the input value to the next layer, which has a connection weight of 1.
Namely:
Figure BDA0003851921700000131
Figure BDA0003851921700000132
the second layer fuzzifies the input quantity. Each neuron represents a linguistic value expressed as a membership function. The theoretical rotation angle deviation from the actual rotation angle and the rate of change of the deviation were defined as 7 linguistic values, so the second layer had 14 total neurons. The output of each neuron corresponds to a respective membership function.
Figure BDA0003851921700000133
Figure BDA0003851921700000134
Wherein A is ij (x i ) When the input is x i The corresponding degree of membership, a ij ，b ij Expressing the center value and the width of the membership function in a normal distribution, adjusting the center value a ij The distribution of the membership function on the domain can be changed, and the width b can be adjusted ij Its shape can be changed, i =1,2, for the input variable index, j =1,2.
Performing fuzzy inference on the third layer, wherein the connection weight between neurons of the second and third layers is 1, and the total number of the neurons is 7 × 7=49, and the method adopts a product composite inference rule and comprises the following steps:
Figure BDA0003851921700000135
Figure BDA0003851921700000136
the fourth layer is a deblurring layer, and the gravity center method comprises the following steps:
Figure BDA0003851921700000137
Figure BDA0003851921700000138
where N =7 is the number of linguistic values contained in each input linguistic variable, w pq Is the conclusion language value w pq Maximum degree of membership.
Selecting an objective function
Figure BDA0003851921700000139
Wherein
Figure BDA00038519217000001310
Desired output, Δ u, for fuzzy control FNN based on neural networks mn Is the actual output of the FNN. We adopt a forward type multilayer neural network and use a back propagation BP algorithm to adjust a ij ，b ij And W ij And (4) evaluating until the objective function is less than or equal to the set value, thereby realizing the off-line memory and optimization of the membership function and the fuzzy control rule.
The yawing dynamic mathematical model of the magnetic levitation yawing motor is obtained by the following processes:
according to fig. 1, the magnetic levitation yaw motor of the present invention is a non-salient pole type synchronous disc motor, without damping windings, ignoring magnetic circuit saturation and leakage inductance of each winding, and according to the coordinate transformation principle, the dynamic voltage equation of the magnetic levitation yaw motor under dq synchronous rotation coordinate system can be obtained as follows:
Figure BDA0003851921700000141
in the formula u sd 、u sq 、u r D-axis and q-axis components of the stator voltage and the rotor voltage, i sd 、i sq 、i r D-axis and q-axis components of the stator current and the rotor current, psi sd 、ψ sq 、ψ r D-axis and q-axis components of the stator flux linkage and the rotor flux linkage, R, respectively s 、R r The resistance of the stator three-phase winding 11 and the resistance, omega, of the rotor dc field winding 21, respectively 1 The stator rotating field angular velocity.
The flux linkage equation is:
Figure BDA0003851921700000142
in the formula, L sd 、L sq Self-inductance of d-axis and q-axis of stator winding, respectively, and L for non-salient pole motor sd ＝L sq ；L m Is the mutual inductance between stator and rotor windings, L r Is the inductance of the rotor dc field winding 21.
Torque and equation of motion:
Figure BDA0003851921700000143
Figure BDA0003851921700000144
where ω is the rotational angular velocity of the rotor, n p Is the pole pair number of a magnetic suspension yaw motor, J is the total rotational inertia, T L Is the load torque.
Will be provided with
Figure BDA0003851921700000151
Substitution into
Figure BDA0003851921700000152
And notice L sd ＝L sq The following can be obtained:
Figure BDA0003851921700000153
the above formula constitutes a yaw dynamic mathematical model of the magnetic suspension yaw motor.
32 Transplanting the trained yaw fuzzy neural network model into a main control chip of the stator converter, and establishing an actual yaw fuzzy neural network model prediction control system based on the main control chip of the stator converter;
33 Input the response output value and the expected speed value omega of the yaw fuzzy neural network model into a yaw nonlinear optimization module, and the yaw nonlinear optimization module determines the optimal control input signal, namely the d-axis component i of the optimal stator current by minimizing a yaw cost function sd_opt And q-axis component i sq_opt D-axis component i of the optimum stator current sd_opt Q-axis component i sq_opt And the rotation speed measured value omega is used as the input of the yaw fuzzy neural network model, and the d-axis component i of the optimal stator current is simultaneously used sd_opt And q-axis component i sq_opt Respectively making difference with respective actual measured value, inputting them into PI controller with amplitude limit to obtain stator voltage control quantity u sd * And u sq * U is obtained after dq/alpha beta coordinate transformation sα * And u sβ * The SVPWM module modulates the magnetic levitation yaw motor to generate a driving signal, and controls the stator converter to generate required excitation voltage and current so that the magnetic levitation yaw motor rotates to a wind facing position according to a rotating speed expected value omega;
the yaw cost function is as follows:
Figure BDA0003851921700000154
in the formula, J Y As a yaw cost function, N p To predict the time domain length, N u To control the time domain length, i sd 、i sq For the control input signals, i.e. the d-axis component and the q-axis component of the stator current, respectively, ω is the desired value of the rotational speed, δ m For the response output value, p, of the yaw fuzzy neural network model 1 、ρ 2 K is the current time, which is the weighting coefficient.
And 4, while yawing, controlling the current of the rotor by using a fuzzy neural network model prediction control strategy through a rotor converter, so that the rotor of the magnetic suspension yawing motor is kept at a suspension balance point in the whole yawing process, wherein the specific method comprises the following steps of:
41 According to a yaw suspension dynamic mathematical model of the magnetic suspension yaw motor, training a yaw suspension fuzzy neural network model;
the yawing and suspending dynamic mathematical model of the magnetic levitation yawing motor is obtained by the following processes:
in the yaw process, because stator current exists, the voltage equation of the rotor direct-current excitation winding 21 is as follows:
Figure BDA0003851921700000161
the flux linkage equation is:
ψ r ＝L m i sd +L r i r
in the formula u r Is the input voltage, R, of the rotor DC field winding 21 r Is the resistance, i, of the rotor DC field winding 21 r Is the current of the rotor DC field winding 21 (i.e. the rotor current), i sd D-axis component of stator current, L m Is the mutual inductance between stator and rotor windings, L r Is the inductance of the rotor DC excitation winding 21 and has L r ＝2k 1 /δ。
By the above two formulas, it can be obtained:
Figure BDA0003851921700000162
the yawing dynamic mathematical model of the magnetic levitation yawing motor in the yawing process can be obtained by combining the following formula:
Figure BDA0003851921700000163
42 Transplanting the trained yaw suspension fuzzy neural network model into a main control chip of the rotor converter, and establishing an actual yaw suspension fuzzy neural network model prediction control system based on the main control chip of the rotor converter;
43 Input the response output value of the yaw fuzzy neural network model and the expected value delta of the levitation air gap into a levitation nonlinear optimization module, and the yaw levitation nonlinear optimization module determines the optimal control input signal, namely the optimal rotor current i by minimizing a yaw levitation cost function yr_opt Will optimize the rotor current i yr_opt And taking the suspension air gap measurement value delta as the input of a yaw suspension fuzzy neural network model, and simultaneously, taking the optimal rotor current i yr_opt With the actual rotor current i r Making difference, sending the difference to PWM module via PID controller to generate drive signal of rotor converter so as to control rotor current i r And the rotor of the magnetic suspension yaw motor is kept at a suspension balance point in the whole yaw process.
The yaw suspension cost function is as follows:
Figure BDA0003851921700000171
in the formula, J YM As a yaw levitation cost function, N p To predict the time domain length, N u To control the time domain length, i r For the control input signal, i.e. the rotor current, δ is the desired value of the floating air gap, δ m For the response output value, p, of the yaw fuzzy neural network model 3 K is the current time, which is the weighting coefficient.
Finally, it should be noted that: while the foregoing is directed to the preferred embodiment of the present invention, other and further embodiments of the invention may be devised without departing from the basic scope thereof, and the scope thereof is determined by the claims that follow.

Claims (7)

1. A magnetic suspension yaw motor control method based on neural network model predictive control is characterized by comprising the following steps:
step 1, when the wind direction changes and needs to be navigated, a rotor converter controls the current of a rotor by adopting a PID control algorithm, so that the rotor of a magnetic suspension yaw motor is suspended upwards to and kept at a suspension balance point to realize stable suspension;
step 2, after stable suspension is realized, the rotor converter uses a fuzzy neural network model prediction control strategy instead to control the rotor current, so that the rotor of the magnetic suspension yaw motor keeps stable suspension at a suspension balance point;
step 3, the stator converter adopts a fuzzy neural network model prediction control strategy to control the current of the stator, so that the magnetic suspension yaw motor rotates to a wind facing position according to a rotating speed expected value omega;
and 4, while yawing, controlling the current of the rotor by using a fuzzy neural network model prediction control strategy through a rotor converter, so that the rotor of the magnetic suspension yawing motor is kept at a suspension balance point in the whole yawing process.
2. The magnetic levitation yaw motor control method based on neural network model predictive control as claimed in claim 1, wherein the specific method of step 2 is:
21 According to a suspension dynamic mathematical model of the magnetic suspension yaw motor, training a suspension fuzzy neural network model;
22 Transplanting the trained suspension fuzzy neural network model into a main control chip of the rotor converter, and establishing an actual suspension fuzzy neural network model prediction control system based on the main control chip of the rotor converter;
23 Output value of response of suspended fuzzy neural network modelδ m and a suspension air gap desired value δ input to a suspension nonlinear optimization module which determines an optimal control input signal, i.e. an optimal rotor current i, by minimizing a suspension cost function r_opt Will optimize the rotor current i r_opt And the suspension air gap measured value delta is used as the input of a suspension fuzzy neural network model, and the optimal rotor current i is simultaneously used r_opt With the actual rotor current i r Making difference, sending the difference to PWM module via PID controller to generate drive signal of rotor converter, and controlling rotor current i r And the rotor of the magnetic suspension yaw motor keeps stable suspension at the suspension balance point.
3. The magnetic levitation yaw motor control method based on neural network model predictive control as claimed in claim 1, wherein the specific method of step 3 is:
31 According to a yaw dynamic mathematical model of the magnetic suspension yaw motor, training a yaw fuzzy neural network model;
32 Transplanting the trained yaw fuzzy neural network model into a main control chip of the stator converter, and establishing an actual yaw fuzzy neural network model prediction control system based on the main control chip of the stator converter;
33 Input the response output value and the expected speed value omega of the yaw fuzzy neural network model into a yaw nonlinear optimization module, and the yaw nonlinear optimization module determines the optimal control input signal, namely the d-axis component i of the optimal stator current by minimizing a yaw cost function sd_opt And q-axis component i sq_opt Optimizing the d-axis component i of the stator current sd_opt Q-axis component i sq_opt And taking the rotation speed measured value omega as the input of the yaw fuzzy neural network model, and simultaneously taking the d-axis component i of the optimal stator current sd_opt And q-axis component i sq_opt Respectively making difference with respective actual measured value, inputting them into PI controller with amplitude limiting to obtain stator voltage control quantity u sd * And u sq * U is obtained after dq/alpha beta coordinate transformation sα * And u sβ * After being modulated by the SVPWM module, the driving signal is generated to control the stator converter to generate the required voltageThe magnetic suspension yaw motor rotates to a wind facing position according to the expected rotating speed value omega by the exciting voltage and the current.
4. The magnetic levitation yaw motor control method based on neural network model prediction control as claimed in claim 1, wherein the specific method of step 4 is:
41 Training a yaw suspension fuzzy neural network model according to a yaw suspension dynamic mathematical model of a magnetic suspension yaw motor;
42 Transplanting the trained yaw suspension fuzzy neural network model into a main control chip of the rotor converter, and establishing an actual yaw suspension fuzzy neural network model prediction control system based on the main control chip of the rotor converter;
43 Input the response output value of the yaw fuzzy neural network model and the expected value delta of the levitation air gap into a levitation nonlinear optimization module, and the yaw levitation nonlinear optimization module determines the optimal control input signal, namely the optimal rotor current i by minimizing a yaw levitation cost function yr_opt Will optimize the rotor current i yr_opt And taking the suspension air gap measurement value delta as the input of a yaw suspension fuzzy neural network model, and simultaneously, taking the optimal rotor current i yr_opt With the actual rotor current i r Making difference, sending the difference to PWM module via PID controller to generate drive signal of rotor converter so as to control rotor current i r And the rotor of the magnetic suspension yaw motor is kept at a suspension balance point in the whole yaw process.
5. The magnetic levitation yaw motor control method based on neural network model predictive control as claimed in claim 2, wherein the levitation cost function is:
Figure FDA0003851921690000031
wherein J is a suspension cost function, N p To predict the time domain length, N u To control the time domain length, i r To controlThe input signal, namely the rotor current, delta is a suspension air gap expected value, delta m is a response output value of the yaw fuzzy neural network model, rho is a weighting coefficient, and k is the current moment.
6. The magnetic levitation yaw motor control method based on neural network model predictive control as claimed in claim 3, wherein the yaw cost function is:
Figure FDA0003851921690000032
in the formula, J Y As a yaw cost function, N p To predict the time domain length, N u To control the time domain length, i sd 、i sq For the control input signals, i.e. the d-axis component and the q-axis component of the stator current, respectively, ω is the desired value of the rotational speed, δ m For the response output value, p, of the yaw fuzzy neural network model 1 、ρ 2 K is the current time, which is the weighting factor.
7. The magnetic levitation yaw motor control method based on neural network model predictive control as claimed in claim 4, wherein the yaw levitation cost function is:
Figure FDA0003851921690000041
in the formula, J YM As a yaw levitation cost function, N p To predict the time domain length, N u To control the time domain length, i r For controlling the input signal, i.e. the rotor current, δ is the desired value of the floating air gap, δ m For the response output value, p, of the yaw fuzzy neural network model 3 K is the current time, which is the weighting factor.
CN202211137280.1A 2022-09-19 2022-09-19 Magnetic suspension yaw motor control method based on neural network model predictive control Pending CN115473464A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211137280.1A CN115473464A (en) 2022-09-19 2022-09-19 Magnetic suspension yaw motor control method based on neural network model predictive control

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211137280.1A CN115473464A (en) 2022-09-19 2022-09-19 Magnetic suspension yaw motor control method based on neural network model predictive control

Publications (1)

Publication Number Publication Date
CN115473464A true CN115473464A (en) 2022-12-13

Family

ID=84332807

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211137280.1A Pending CN115473464A (en) 2022-09-19 2022-09-19 Magnetic suspension yaw motor control method based on neural network model predictive control

Country Status (1)

Country Link
CN (1) CN115473464A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117454394A (en) * 2023-10-10 2024-01-26 盐城工学院 Image encryption method and system based on fuzzy neural network fixed time synchronization

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117454394A (en) * 2023-10-10 2024-01-26 盐城工学院 Image encryption method and system based on fuzzy neural network fixed time synchronization
CN117454394B (en) * 2023-10-10 2024-05-21 盐城工学院 Image encryption method and system based on fuzzy neural network fixed time synchronization

Similar Documents

Publication Publication Date Title
CN110401378B (en) Magnetic suspension yaw motor control method based on neural network-model predictive control
CN108183650B (en) Wind power magnetic suspension yaw motor control method based on model prediction control
Li et al. On the rejection of internal and external disturbances in a wind energy conversion system with direct-driven PMSG
CN107070342A (en) A kind of control system for permanent-magnet synchronous motor of bringing onto load state observer
CN111835261B (en) Magnetic suspension vertical axis wind turbine generator suspension control method based on adaptive neural network
CN105179164A (en) Wind energy converting system sliding mode control method and device based on T-S fuzzy model
CN110345013B (en) Magnetic suspension vertical axis wind turbine generator control method based on neural network model predictive control
CN110805523B (en) Suspension control method of wind power magnetic suspension yaw system based on neural network
CN115473464A (en) Magnetic suspension yaw motor control method based on neural network model predictive control
CN113572402A (en) Composite sliding mode speed control method and system for cylindrical permanent magnet linear synchronous motor
CN113090461B (en) Low wind speed vertical axis wind turbine suspension control method based on sliding mode neural network model prediction
CN113691176B (en) Permanent magnet direct-drive wind turbine generator control method based on neural network direct torque control
Chatri et al. Improved high-order integral fast terminal sliding mode-based disturbance-observer for the tracking problem of PMSG in WECS
Wang et al. Research on suspension control strategy based on finite control set model predictive control with state feedback control‐PID for maglev yaw system of wind turbine
CN103645637A (en) Single degree of freedom active magnetic bearing support vector machine self-adaption inverse controller construction method
CN116638544A (en) Joint module cooperative control method based on super local model
CN113343592B (en) Intelligent control method for DQN (direct current) of permanent magnet synchronous motor of new energy aircraft
Wang et al. Research on MPC-based suspension control of wind maglev yaw system
Ayub et al. Nonlinear super-twisting based speed control of PMSG-ECS using higher order sliding mode control
Shi et al. Model predictive control of direct-driven surface-mounted permanent magnet synchronous generator based on active disturbance rejection controller
CN113775474A (en) Suspension control method of vertical axis wind turbine generator based on self-adaptive neural network finite time control
Chaoui et al. Nonlinear power control of doubly fed induction generator wind turbines using neural networks
Zhi-nong et al. The intelligent control of DFIG-based wind generation
Nath et al. Speed control of neural network based energy efficient DC drive
CN114977922A (en) PMLSM position sensorless control method

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