CN102629848A - Speed-sensor-less construction method for bearing-less asynchronous motor - Google Patents
Speed-sensor-less construction method for bearing-less asynchronous motor Download PDFInfo
- Publication number
- CN102629848A CN102629848A CN2012101245596A CN201210124559A CN102629848A CN 102629848 A CN102629848 A CN 102629848A CN 2012101245596 A CN2012101245596 A CN 2012101245596A CN 201210124559 A CN201210124559 A CN 201210124559A CN 102629848 A CN102629848 A CN 102629848A
- Authority
- CN
- China
- Prior art keywords
- svms
- rotating speed
- input
- asynchronous machine
- bearing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Magnetic Bearings And Hydrostatic Bearings (AREA)
Abstract
The invention discloses a speed-sensor-less construction method for a bearing-less asynchronous motor, which comprises the following steps of: firstly, constructing an embedded rotating speed subsystem for an original bearingless asynchronous motor system; then, establishing a bearingless asynchronous motor rotating speed left-inverse system of the embedded rotating speed subsystem; subsequently, adopting a static neural network and two differentiators S to form a support vector machine inverse, training the support vector machine, and adjusting and determining all weight coefficients of the support vector machine so as to realize the bearing-less asynchronous motor rotating speed left-inverse system; and finally, inversely connecting the support vector machine behind the original bearing-less asynchronous motor system in series so as to finish the speed-sensor-less construction. The speed-sensor-less control strategy of the bearing-less asynchronous motor based on support vector machine left-inverse omits a mechanical speed sensor and an interface circuit thereof of the original control system, so that the cost of the control system is lowered, and rotating speed identification can be quickly and accurately carried out in the full-speed range. The speed-sensor-less construction method for a bearing-less asynchronous motor has strong adaptability, fault tolerance and robustness.
Description
Technical field
The present invention is a kind of no bearing asynchronous machine Speedless sensor building method based on the SVMs left inverse; For the no speed operation of no bearing asynchronous machine provides a kind of new control strategy; Be applicable to the high performance control of no bearing asynchronous machine, belong to the technical field of Electric Drive control appliance.
Background technology
The high performance control of no bearing asynchronous machine be unable to do without the closed-loop control of rotating speed, thereby needs the speed of detection rotor, and the conventional motors spinner velocity adopts the mechanical type velocity transducer to detect, and has problems such as installation, connection and reliability.As far as no bearing asynchronous machine, use the mechanical type velocity transducer that bigger limitation is arranged, because transducer itself mechanically is difficult to realize motor high speed, ultrahigh speed operation, thus the performance of the good high speed performance of the serious no bearing asynchronous machine of restriction.Therefore, the Speedless sensor technology becomes the effective means that solves no this problem of bearing asynchronous machine.
Speedless sensor operation for no bearing asynchronous machine; Just at the early-stage at present; Mainly contain and adopt model reference adaptive method strategy that no bearing asynchronous machine Speedless sensor operation is studied, realize spinner velocity and stator resistance on-line identification, carried out simulation study; Adopt parameter identification method to cooperate gray system theory that rotor-position and rotating speed are carried out identification, realized the operation of permanent-magnetic electric machine with bearing Speedless sensor.Factor affecting such as rotor eccentricity, magnetic saturation, temperature rise and load variations did not really realize the accurate on-line identification of rotary speed parameter when these Speed identification methods were all considered high-speed cruising.Utilize the powerful approximation capability of static neural network based on the Speedless sensor method of neural net inverse system to nonlinear function; Break through the bottleneck of inverse system in realizing or using; But there are defectives such as local extreme points is little, mistake is learnt, operand is big in neural net self; Limit the further application of neural net inverse approach, also influenced its control performance.
Summary of the invention
The present invention seeks in order in full speed range, quick and precisely to detect the rotating speed of no bearing asynchronous machine; Realize no bearing asynchronous machine Speedless sensor operation; Improve no bearing asynchronous machine service behaviour, promote the application of no bearing asynchronous machine and a kind of no bearing asynchronous machine Speedless sensor building method is provided.
The technical scheme that the present invention adopts is: 1) no bearing asynchronous machine original system structure is included the rotating speed subsystem, the input variable of no bearing asynchronous machine original system is the stator voltage of torque winding
u s1
d ,
u s1
q And synchronous speed
ω 1, output variable is the stator current of torque winding
i s1
d With
i s1
q The input variable that includes the rotating speed subsystem is a rotating speed to be measured
ω r , output variable is the stator voltage of torque winding
u s1
d ,
u s1
q , stator current
i s1
d ,
i s1
q , degree of commentaries on classics synchronously
ω 1And the first derivative of stator current
,
2) set up the no bearing asynchronous machine rotating speed left inverse system include the rotating speed subsystem, 7 output variables that being input as of no bearing asynchronous machine rotating speed left inverse system includes the rotating speed subsystem, be output as rotating speed to be measured
ω r 3) it is contrary that the SVMs of 7 inputs of employing nodes, 1 output node adds 2 differentiator S formation SVMs, and the contrary input of SVMs is respectively stator voltage
u s1
d ,
u s1
q , stator current
i s1
d ,
i s1
q And synchronous speed
ω 1, be output as rotating speed to be measured
ω r 4) SVMs is trained, the vectorial coefficient of adjustment and definite SVMs and threshold value are with the no bearing asynchronous machine rotating speed left inverse of realization system; 5) SVMs is configured to Speedless sensor after being serially connected with no bearing asynchronous machine original system.
The invention has the beneficial effects as follows:
1. the present invention has provided the method that the left inverse system combines with SVMs, utilizes the powerful approximation capability of SVMs to nonlinear function, breaks through the bottleneck of inverse system in realizing or using.The SVMs that adopts is a strong instrument in Industrial Engineering, setting up nonlinear model; Can approach the static non linear mapping (function) of any complicacy with arbitrary accuracy; Have stronger generalization ability and adaptive ability, and can self study and self adaptation the unknown or uncertain system.
2. required input signal is the local direct measurable variable that obtains easily in the actual engineering in the Speedless sensor building method, and SVMs is contrary itself can be realized through software programming.Adopt Speedless sensor of the present invention, omitted the photoelectric encoder and the interface circuit thereof of original system, need not carry out other any change to no bearing asynchronous machine system, the realization expense is low, and is safe and reliable, is easy to Project Realization.
Description of drawings
Below in conjunction with accompanying drawing and embodiment the present invention is done further explain:
Fig. 1 is the sketch map that is included rotating speed subsystem 12 by no bearing asynchronous machine original system 11 structures;
Fig. 2 is by the schematic diagram that includes the detection rotating speed that rotating speed subsystem 12 and no bearing asynchronous machine rotating speed left inverse system 2 form;
Fig. 3 is the pie graph of SVMs contrary 32;
Fig. 4 is that SVMs contrary 32 is connected in series with no bearing asynchronous machine original system 11 and concerns sketch map.
Embodiment
Referring to Fig. 4; The present invention at first sets up the Mathematical Modeling that includes rotating speed subsystem 12 based on the Mathematical Modeling of no bearing asynchronous machine original system 11, and this includes and satisfies the determined variable bound of no bearing asynchronous machine original system 11 Mathematical Modelings between input variable and the output variable of rotating speed subsystem 12 and concern; Then set up the inversion model that includes rotating speed subsystem 12, promptly do not have bearing asynchronous machine rotating speed left inverse system 2; Adopt SVMs 31 and 2 differentiator S of 7 input nodes, 1 output node to constitute the SVMs that includes rotating speed subsystem 12 contrary 32 again with 5 input nodes, 1 output node; And, make the function of the no bearing asynchronous machine rotating speed left inverse of contrary 32 realizations of SVMs system 2 through adjusting the vectorial coefficient and the threshold value of SVMs 31; Be configured to Speedless sensor after at last SVMs contrary 32 being serially connected with no bearing asynchronous machine original system 11, can realize online in real time detection the rotating speed of no bearing asynchronous machine.Practical implementation is divided into following 5 steps successively:
1, sets up the Mathematical Modeling that includes rotating speed subsystem 12
Referring to Fig. 1,3 input variables of no bearing asynchronous machine original system 11 are to survey the stator voltage of torque winding
u s1
d ,
u s1
q And synchronous speed
ω 1, this stator voltage
u s1
d ,
u s1
q And synchronous speed
ω 1Be direct measurable variable; 2 output variables are the stator currents of the torque winding that can survey
i s1
d With
i s1
q 1 input variable that includes rotating speed subsystem 12 is a rotating speed
ω r , rotating speed
ω r It is the rotating speed to be measured of no bearing asynchronous machine; 7 output variables are to survey the stator voltage of torque winding
u s1
d ,
u s1
q , stator current
i s1
d ,
i s1
q , degree of commentaries on classics synchronously
ω 1And the first derivative of measurable variable stator current
,
No bearing asynchronous machine original system 11 structures are included the Mathematical Modeling of rotating speed subsystem 12; To adopting the no bearing asynchronous machine original system 11 of voltage control PWM inverter power supply, the Mathematical Modeling of torque winding is under the rotation d-q of torque wound rotor flux linkage orientation coordinate system:
In the formula,
L m1
,
L r1
,
L s1
,
R r1
,
R s1
,
p 1Be respectively mutual inductance, rotor and stator self-induction, rotor and stator resistance, the number of pole-pairs of no bearing asynchronous motor torque winding;
ψ r1
d ,
ψ r1
q ,
i s1
d ,
i s1
q ,
u s1
d ,
u s1
q Be respectively d axle and q axle rotor flux, d axle and q axle stator current, d axle and the q axle stator voltage of torque winding;
J,
ω 1,
ω r Be respectively moment of inertia, synchronous speed, rotating speed;
T r =
L r1
/
R r1
σ=1-
L 2 m1
/ (
L s1
L r1
).Like this, having the input that includes rotating speed subsystem 12 in the Mathematical Modeling shown in the formula (1-1) ~ (1-4) is rotating speed
ω r , output is direct measurable variable
i s1
d ,
i s1
q
Referring to Fig. 2, in order to detect rotating speed to be measured
ω r , when including rotating speed subsystem 12 reversal conditions and set up, will not have bearing asynchronous machine rotating speed left inverse system 2 and be serially connected in this and include after the rotating speed subsystem 12,7 inputs of no bearing asynchronous machine rotating speed left inverse system 2 are
u s1
d ,
u s1
q ,
i s1
d ,
,
i s1
q ,
And
ω 1, output is
ω r , can obtain identical expression formula, do not have the output of bearing asynchronous machine rotating speed left inverse system 2 like this and can duplicate the input that this includes rotating speed subsystem 12 fully.
According to the above-mentioned building method that includes the Mathematical Modeling of rotating speed subsystem 12, choose formula (1-1) and formula (1-2) as the Mathematical Modeling that includes rotating speed subsystem 12.
2, set up the inversion model that includes rotating speed subsystem 12, promptly do not have bearing asynchronous machine rotating speed left inverse system 2.
There is theorem according to inverse function; It is reversible can using conventional method of proof proof formula (1-1) and the represented rotating speed subsystem 12 that includes of formula (1-2), and the inversion model that includes the no bearing asynchronous machine rotating speed left inverse system 2 of rotating speed subsystem 12 can be represented with following nonlinear function:
?（2-1）
And confirm to include 7 stator voltages that are input as the torque winding of no bearing asynchronous machine of the inversion model of rotating speed subsystem 12
u s1
d ,
u s1
q , stator current
i s1
d ,
i s1
q And the first derivative of stator current
,
And synchronous speed
ω 1, 1 is output as rotating speed
ω r
Need to prove, the finding the solution and can skip of formula (2-1) nonlinear function, this formula is that structure and the learning training for SVMs contrary 32 provides the basis on the method, thereby can construct SVMs as shown in Figure 3 against 32.
3, adopt SVMs 31 and 2 differentiator S to construct SVMs contrary 32
Referring to Fig. 3, adopt SVMs 31 to add 2 differentiator S and constitute SVMs contrary 32 with 5 input nodes, 1 output node with 7 input nodes, 1 output node.The stator voltage of the torque winding that 5 inputs of SVMs contrary 32 are respectively no bearing asynchronous machines
u s1
d ,
u s1
q , stator current
i s1
d ,
i s1
q And synchronous speed
ω 1, 1 is output as rotating speed to be measured
ω r The input number of nodes of SVMs 31 is 7, and the output node number is 1, chooses the kernel function that gaussian kernel function is a SVMs, and the vectorial coefficient of SVMs 31 and threshold value will be confirmed in next step training.
First, second and third input of SVMs 31 is respectively first, second and third input of SVMs contrary 32; Wherein the 3rd input of SVMs 31 is output as the 4th input of SVMs 31 through a differentiator S; The 5th input of SVMs 31 is the 4th inputs of SVMs contrary 32, and it is output as the 6th input of SVMs 31 through another differentiator S; The 7th input of SVMs 31 is the 5th inputs of SVMs contrary 32.SVMs 31 is formed SVMs against 32 with two differentiator S, and the output of SVMs 31 is exactly the output of SVMs contrary 32.
4, the vectorial coefficient and the threshold value of adjustment and definite SVMs 31
Earlier with pumping signal, i.e. stator voltage
u s1
d ,
u s1
q , stator current
i s1
d ,
i s1
q And synchronous speed
ω 1Signal is added in the input of no bearing asynchronous machine original system 11, gathers the rotating speed to be measured of no bearing asynchronous machine
ω r The stator current that will not have the torque winding of bearing asynchronous machine again
i s1
d ,
i s1
q Off-line is asked its first derivative respectively, and signal is done standardization processing, the training sample set of composition SVMs
u s1
d ,
u s1
q ,
i s1
d ,
,
i s1
q ,
,
ω 1,
ω r ; At last, the regularization parameter of SVMs is set at 560, and the nuclear width setup is 1.4, thereby the vectorial coefficient and the threshold value of off-line adjustment SVMs make the function of SVMs against the no bearing asynchronous machine rotating speed left inverse of 32 realizations system 2.
5, SVMs contrary 32 is serially connected with no bearing asynchronous machine original system 11 and promptly is configured to Speedless sensor afterwards, realize online in real time detection, referring to Fig. 4 to the rotating speed of no bearing asynchronous machine.
According to the above, just can realize the present invention.
Claims (3)
1. no bearing asynchronous machine Speedless sensor building method is characterized in that adopting following steps:
1) no bearing asynchronous machine original system (11) structure is included rotating speed subsystem (12), the input variable of no bearing asynchronous machine original system (11) is the stator voltage of torque winding
u s1
d ,
u s1
q And synchronous speed
ω 1, output variable is the stator current of torque winding
i s1
d With
i s1
q The input variable that includes rotating speed subsystem (12) is a rotating speed to be measured
ω r , 7 output variables are stator voltages of torque winding
u s1
d ,
u s1
q , stator current
i s1
d ,
i s1
q , degree of commentaries on classics synchronously
ω 1And the first derivative of stator current
,
2) set up the no bearing asynchronous machine rotating speed left inverse system (2) include rotating speed subsystem (12), no bearing asynchronous machine rotating speed left inverse system (2) is input as said 7 output variables that include rotating speed subsystem (12), is output as rotating speed to be measured
ω r
3) adopt the SVMs (31) of 7 input nodes, 1 output node to add 2 differentiator S formation SVMs contrary (32), the input of SVMs contrary (32) is respectively the stator voltage of said torque winding
u s1
d ,
u s1
q , stator current
i s1
d ,
i s1
q And synchronous speed
ω 1, be output as rotating speed to be measured
ω r
4) SVMs (31) is trained, the vectorial coefficient and the threshold value of adjustment and definite SVMs (31) are to realize no bearing asynchronous machine rotating speed left inverse system (2);
5) SVMs contrary (32) is serially connected with no bearing asynchronous machine original system (11) and is configured to Speedless sensor afterwards.
2. a kind of no bearing asynchronous machine Speedless sensor building method according to claim 1 is characterized in that: first, second, third input of SVMs in the step 3) (31) is respectively first, second, third input of SVMs contrary (32); The 3rd input of SVMs (31) is output as the 4th input of SVMs (31) through a differentiator S; The 5th input of SVMs (31) is the 4th input of SVMs contrary (32); The 5th input of SVMs (31) is output as the 6th input of SVMs (31) through another differentiator S, and the 7th input of SVMs (31) is the 5th input of SVMs contrary (32); The output of SVMs (31) is the output of SVMs contrary (32).
3. a kind of no bearing asynchronous machine Speedless sensor building method according to claim 1, it is characterized in that: the vectorial coefficient and the threshold value determination method of SVMs in the step 4) (31) are: elder generation is with the stator voltage of torque winding
u s1
d ,
u s1
q , stator current
i s1
d ,
i s1
q And synchronous speed
ω 1Signal is added in the input of no bearing asynchronous machine original system (11), gathers rotating speed to be measured
ω r Again with stator current
i s1
d ,
i s1
q Off-line is asked its first derivative respectively, forms the training sample set of SVMs; And choose the kernel function of gaussian kernel function as SVMs (31), and setting regularization parameter is 800, the nuclear width is 1.62, and SVMs (31) trained, thus the vectorial coefficient and the threshold value of definite SVMs (31).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012101245596A CN102629848A (en) | 2012-04-26 | 2012-04-26 | Speed-sensor-less construction method for bearing-less asynchronous motor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012101245596A CN102629848A (en) | 2012-04-26 | 2012-04-26 | Speed-sensor-less construction method for bearing-less asynchronous motor |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102629848A true CN102629848A (en) | 2012-08-08 |
Family
ID=46588025
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2012101245596A Pending CN102629848A (en) | 2012-04-26 | 2012-04-26 | Speed-sensor-less construction method for bearing-less asynchronous motor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102629848A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103051275A (en) * | 2012-11-27 | 2013-04-17 | 江苏大学 | Construction method of flux linkage observing device of electric automobile induction motor |
| CN103414417A (en) * | 2013-07-23 | 2013-11-27 | 江苏大学 | Bearing-free asynchronous motor suspension system fault-tolerant controller and construction method thereof |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009136085A (en) * | 2007-11-30 | 2009-06-18 | Hitachi Ltd | Controller of ac motor |
| CN101640513A (en) * | 2009-08-12 | 2010-02-03 | 江苏大学 | Support vector machine (SVM) inverse controller of two-motor variable-frequency speed-regulating system and construction method thereof. |
| CN101814892A (en) * | 2010-03-04 | 2010-08-25 | 江苏大学 | Support vector machine inverse system composite controller based on bearingless synchronous reluctance motor |
| CN102130647A (en) * | 2011-01-10 | 2011-07-20 | 江苏大学 | Speed-sensor-less construction method for detecting rotating speed of bearing-less asynchronous motor |
-
2012
- 2012-04-26 CN CN2012101245596A patent/CN102629848A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009136085A (en) * | 2007-11-30 | 2009-06-18 | Hitachi Ltd | Controller of ac motor |
| CN101640513A (en) * | 2009-08-12 | 2010-02-03 | 江苏大学 | Support vector machine (SVM) inverse controller of two-motor variable-frequency speed-regulating system and construction method thereof. |
| CN101814892A (en) * | 2010-03-04 | 2010-08-25 | 江苏大学 | Support vector machine inverse system composite controller based on bearingless synchronous reluctance motor |
| CN102130647A (en) * | 2011-01-10 | 2011-07-20 | 江苏大学 | Speed-sensor-less construction method for detecting rotating speed of bearing-less asynchronous motor |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103051275A (en) * | 2012-11-27 | 2013-04-17 | 江苏大学 | Construction method of flux linkage observing device of electric automobile induction motor |
| CN103051275B (en) * | 2012-11-27 | 2015-07-08 | 江苏大学 | Construction method of flux linkage observing device of electric automobile induction motor |
| CN103414417A (en) * | 2013-07-23 | 2013-11-27 | 江苏大学 | Bearing-free asynchronous motor suspension system fault-tolerant controller and construction method thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Chang et al. | Sensorless position estimation of switched reluctance motor at startup using quadratic polynomial regression | |
| CN109412488A (en) | A kind of permanent magnet synchronous motor dynamic matrix control method | |
| CN102130647B (en) | Speed-sensor-less construction method for detecting rotating speed of bearing-less asynchronous motor | |
| CN103501148A (en) | Method for controlling operation of non-radial displacement sensor of bearingless permanent magnetic synchronous motor | |
| CN103259479A (en) | Method for observing left inverse state of neural network of permanent magnet synchronous motor | |
| Hachicha et al. | Finite element method for induction machine parameters identification | |
| CN109194208A (en) | Speed Sensorless Control Method | |
| CN102769424A (en) | Support vector machine inverse-based bearingless asynchronous motor control system | |
| CN102510253B (en) | Sensorless controller of bearingless synchronous reluctance motor and method for controlling sensorless controller | |
| Amrane et al. | Unknown input observer design for fault sensor estimation applied to induction machine | |
| Jannati et al. | Speed sensorless fault-tolerant drive system of 3-phase induction motor using switching extended kalman filter | |
| JP6850458B1 (en) | AI-equipped motor state estimation system and machine learning method for motor models | |
| Song et al. | Identification of PMSM based on EKF and elman neural network | |
| Comanescu et al. | Reduced order observers for rotor position estimation of nonsalient PMSM | |
| CN102629848A (en) | Speed-sensor-less construction method for bearing-less asynchronous motor | |
| Comanescu et al. | Full order EMF observer for PMSM—design, analysis and performance under improper speed signal | |
| Abbas Khan et al. | Improved MRAS based speed estimation for a vector controlled switched reluctance motor drive | |
| Jin-Feng et al. | Control of switched reluctance motors based on improved BP neural networks | |
| CN103051275B (en) | Construction method of flux linkage observing device of electric automobile induction motor | |
| Wang et al. | Sensorless speed control of permanent magnet synchronous motors by neural network algorithm | |
| CN113890442A (en) | Method for detecting current fault of wind driven generator rotor by particle swarm sliding-mode observer | |
| Cortajarena et al. | Influence of the rotor angle precision in control of interior permanent magnet synchronous machine drives and improvement method using sensorless estimator with Hall sensors | |
| Nguyen et al. | Efficient Sensorless Speed Estimation of Electrical Servo Drives Using a Full‐Order Nonsingular Terminal Sliding Mode Observer | |
| De La Guerra et al. | Global observability analysis of the SR Motor under sensorless operation | |
| CN106301107B (en) | A kind of pump-storage generator stationary frequency starting rotor position estimate method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20120808 |