CN106803731B - A kind of five-phase PMSM model prediction method for controlling torque - Google Patents
A kind of five-phase PMSM model prediction method for controlling torque Download PDFInfo
- Publication number
- CN106803731B CN106803731B CN201710022345.0A CN201710022345A CN106803731B CN 106803731 B CN106803731 B CN 106803731B CN 201710022345 A CN201710022345 A CN 201710022345A CN 106803731 B CN106803731 B CN 106803731B
- Authority
- CN
- China
- Prior art keywords
- voltage vector
- magnetic linkage
- torque
- vector
- indicate
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Control Of Ac Motors In General (AREA)
Abstract
The invention discloses a kind of five-phase PMSM model prediction method for controlling torque, belong to the design of polyphase machine ac control system and manufacture in power electronics and power drives field, its key step are as follows: the sector first where calculating dead beat voltage vector, input set of 4 voltage vectors (+2 big vector of 2 middle vectors) as model prediction direct torque in the sector is only chosen, accomplishes real-time update；Then prediction calculating carried out to 4 voltage vectors in input set using prediction model, combined objective function choose optimal voltage vector as system subsequent time input.The method of the present invention uses model prediction direct torque, has excellent dynamic response performance, and calculation amount is smaller, Digital Implementation is more easy, alleviates the pressure of digit chip.
Description
Technical field
It (include five phase inversions the present invention relates to polyphase machine ac control system in power electronics and power drives field
Device, five-phase PMSM) it designs and manufacture, in particular to a kind of five-phase PMSM model prediction direct torque side
Method.
Background technique
The progress of microelectric technique and power electronic technique is laid a good foundation for the development of polyphase ac speed-regulating system.It is more
The advantage of phase system essentially consists in: the vibration of motor and noise are smaller, and inverter capacity is big, good output, and reliability
By force, power density is high.
Finite aggregate model prediction direct torque (Finite-Control-Set Model Predictive Torque
Control, FCS-MPTC) it is a kind of vehicle air-conditioning algorithm.In five phase current transformers-motor driven systems, usually only select
11 vectors (big vector+zero vector, MPTC-11) or 21 vectors (big vector+middle vector+zero vector, MPTC-21) are taken to make
For input set.It, will be to the voltage vector progress time in input set in each control period in FCS-MPTC algorithm
Go through calculating.11 vectors are chosen as input set, although reducing calculation amount to a certain extent, meeting in motor stator electric current
There is biggish harmonic wave, torque pulsation is larger when causing motor stable state and control effect is bad；And 21 vectors are chosen as defeated
Although entering set enriches the quantity of domination set, and improve the control performance of system, to numerical control system bring compared with
Big computation burden can generate the other problems such as digital delay.
Summary of the invention
Technical problem to be solved by the invention is to provide a kind of five-phase PMSM model prediction direct torque sides
Method mitigates the computation burden of numerical control system, guarantees the excellent control performance of system.
In order to solve the above technical problems, the technical solution adopted by the present invention is that:
A kind of five-phase PMSM model prediction method for controlling torque, comprising the following steps:
Step 1: according to the Parker rotation transformation of extension by voltage of the five-phase PMSM under natural system of coordinates,
The physical quantitys such as electric current, magnetic linkage, torque are respectively mapped under two orthogonal coordinate systems of d1-q1 and d3-q3；
Step 2: classify to the fundamental space voltage vector of two level, five phase voltage source inventer, and calculate it is big or middle,
The amplitude of small voltage vector is respectively as follows:
In formula, UL, UM, USRespectively indicate the amplitude of large, medium and small voltage vector；UdcIndicate the bus electricity of inverter direct-flow side
Press size；
Step 3: dead beat voltage vector predicted value being gone out according to the torque of five-phase induction motor and magnetic linkage error calculation, then is extracted
The amplitude and location information of dead beat predicted voltage vector out；
Step 4: the place sector of dead beat voltage vector is determined according to the voltage vector location information being calculated；
Step 5: input set of 4 vectors as PREDICTIVE CONTROL where choosing dead beat voltage vector in sector；
Step 6: prediction and evaluation calculating being carried out to the voltage vector in domination set, the minimum value for choosing objective function is corresponding
Voltage vector is applied to inverter.
Further, the step 3 specifically:
The variable quantity △ δ of required power angle δ is calculated according to the torque error at k moment are as follows:
Te refIndicate the given value of motor speed outer ring torque；δ indicates the power angle between electric machine rotor magnetic linkage, NpTable
Show motor number of pole-pairs, LdIndicate d-axis inductance, ψfIndicate the magnetic linkage of permanent magnet, ψsIndicate stator magnetic linkage vector；
According to formulaCalculate the stator magnetic linkage error under α β coordinate system:
ψs refIndicate the given value of fundamental wave magnetic linkage；
Component U of the dead beat voltage vector predicted value under two-phase α β rest frameαAnd UβAre as follows:
Finally according to formulaAcquire the amplitude and location information of voltage vector in dead-beat control method:
Further, the objective function are as follows:
Wherein, i=m, m+1, m+2, m+3；M indicates the sector number in each sampling period where dead beat voltage vector；
ψsd1-q1 refIndicate the given value of first harmonic spatial stator magnetic linkage；Te(k+1)、ψsd1-q1(k+1) it respectively indicates the torque of k+1 moment and determines
The predicted value of sub- magnetic linkage；ψsd3-q3(k+1) the harmonic prediction value of k+1 moment stator magnetic linkage is indicated；ψsd3-q3 refIt is fixed for harmonic wave space
The given value of sub- magnetic linkage；λ1、λ2Respectively in the weight coefficient of first harmonic spatial and harmonic wave space stator magnetic linkage component.
Further, I in objective functionmaxAre as follows:
|Ilim| the current limit value for indicating setting, current amplitude after detecting the effect of a certain voltage vector is more than maximum
The voltage vector is not chosen in the limitation of value then.
Compared with prior art, the beneficial effects of the present invention are:
1) the input control collection for simplifying FCS-MPTC method, avoids a large amount of redundant computation, alleviates digital control
The computation burden of system.
2) FCS-MPTC method rapid dynamic response speed is remained, the advantages that robustness is good.
3) the present invention is based on five phase inverter motor driven systems to be analyzed, equally can be with expanded application in other multiphases electricity
Machine；Meanwhile the problems such as being related to such as power, torque pulsation, current harmonics for the occasion of other power conversions, also provides ginseng
Examine value.
Detailed description of the invention
Fig. 1 is the allomeric function block diagram of the method for the present invention.
Fig. 2 is the schematic diagram for calculating dead beat voltage vector.
Fig. 3 is the block diagram for predicting torque and magnetic linkage.
Fig. 4 is Space Rotating coordinate transform computing block diagram.
Fig. 5 is that the method for the present invention calculates duration.
Fig. 6 is the calculating duration of MPTC-21 (10 big vector+10 middle vector+zero vectors) method.
Fig. 7 is motor speed, torque and a phase current waveform (n=550r/min, T under stable situationL=8Nm).
Fig. 8 is motor speed, torque and a phase current waveform (n=550r/min, T under current intelligenceLIt is jumped by 2Nm
Fade to 8Nm).
Specific embodiment
The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.The method of the present invention significantly subtracts
Small calculation amount also realizes the PREDICTIVE CONTROL to torque, improves the dynamic response capability of system.Specifically includes the following steps:
Five-phase induction motor system includes 4 freedom degrees and zero-sequence component.Under motor normal operating condition, according to the group of extension
Symmetrical physical quantity under natural system of coordinates is respectively mapped under two orthogonal coordinate systems of d1-q1 and d3-q3 by gram rotation transformation.
Classified (table 1) to the space vector of two level, five phase voltage source inventer, and all kinds of voltage vectors are calculated according to formula (1)
Amplitude size.
The classification of 1 voltage vector group of table
In formula (1), UdcIndicate the busbar voltage size of inverter direct-flow side.Vector and d3-q3 under d1-q1 coordinate system are sat
There are corresponding relationships for vector under mark system: the large, medium and small vector under d1-q1 coordinate system can be mapped under d3-q3 coordinate system
It is small, in, big vector.
In order to reduce calculation amount, dead beat voltage vector is calculated first, where only choosing dead beat voltage vector in sector
4 vectors (2 big vector+2 middle vectors) be used as input set, avoid the redundant computation to unrelated vector.
Surface-mount type five-phase PMSM torque TeExpression formula are as follows:
In formula, δ indicates the power angle between electric machine rotor magnetic linkage, NpIndicate motor number of pole-pairs, LdIndicate d-axis inductance,
ψfIndicate the magnetic linkage of permanent magnet, ψsIndicate stator magnetic linkage.It is constant that stator magnetic linkage amplitude is generally remained in Direct Torque Control,
The torque variation of motor is adjusted by changing the size of power angle.Under normal circumstances, power angle and its variable quantity are smaller, can
Do approximate processing:
δ≈sinδ (3)
Based on this, the variable quantity △ δ of required power angle is calculated according to the torque error at k moment are as follows:
Te refIndicate the given value of motor speed outer ring torque.
The magnetic linkage error under α β coordinate system is calculated according to formula (4):
ψs refDead beat predicted voltage arrow can be obtained according to motor stator voltage equation in the given value for indicating fundamental wave magnetic linkage
The α β component size of amount:
The amplitude and location information of voltage vector in track with zero error algorithm are finally acquired according to formula (6):
Definition of Division has been carried out to resulting dead beat voltage vector in table 2.According to the voltage location angle letter being calculated
Breath is to determine the sector where it.
2 sector of table divides and domination set chooses table
According to sector number, traversal prediction is carried out to 4 voltage vectors (2 big vector+2 middle vectors) in its domination set
It calculates, chooses optimal vector further according to objective function.Objective function J are as follows:
Wherein i=m, m+1, m+2, m+3；M indicates the sector number in each sampling period where dead beat voltage vector；
ψsd1-q1 refIndicate the given value of first harmonic spatial magnetic linkage；Te(k+1)、ψsd1-q1(k+1) torque of k+1 moment and magnetic linkage are respectively indicated
Predicted value；ψsd3-q3(k+1) predicted value of k+1 moment harmonic wave magnetic linkage is indicated；λ1、λ2First harmonic spatial and harmonic wave space stator magnet respectively
Weight coefficient of the chain component in evaluation function；ψsd3-q3 refFor the given value of harmonic wave space magnetic linkage, in order to inhibit stator current
Low-order harmonic, value are disposed as 0.I in objective functionmax, current protection effect is embodied, is defined such as formula (9):
|Ilim| the current limit value for indicating setting, current amplitude after detecting the effect of a certain voltage vector is more than maximum
The voltage vector is not chosen in the limitation of value then；If this adjusting that will not influence objective function is made within the scope of given value
With.
In Fig. 1, whole system is divided into rotor flux and torque is observed, and dead beat voltage vector calculates, torque and magnetic linkage are pre-
Survey calculating, Space Rotating coordinate transform calculates four parts.The wherein particular content of major part are as follows:
1) rotor flux and torque are observed
The method of the present invention needs to go out according to magnetic linkage, torque error inverse the voltage vector for the kth moment needing to apply, therefore needs
Rotor flux is observed in real time.The given value of torque is obtained by speed error by a PI controller.
2) dead beat voltage vector calculates
Fig. 2 shows the schematic diagram calculations of dead beat voltage vector.Approximate calculation is carried out to power angle, combined torque error obtains
To the expectancy changes amount △ δ of power angle, and calculate the magnetic linkage error under α β coordinate system.According to magnetic linkage error, indifference is calculated
Beat voltage vector and its place sector, 4 voltage vectors (+2 big vector of 2 middle vectors) in sector are the side FCS-MPTC
The input set of method.
3) torque and Flux estimation calculate
In FCS-MPTC method, the precision of torque prediction calculating section directly affects the control performance of system.Obtaining current is
State variable, according to the form of standard state spatial function, state equation is written as:Discretization in order to obtain
Current of electric state equation, in sampling time TsHypothesis is made in lesser situation: thinking system input variable u in kTs~(k+
1)TsIt is constant in time；H represents the influence of back-emf, changes for electric current loop relatively slowly, it also hold that H is in kTs~(k+1)
TsIt is constant in time interval.Torque and the Flux estimation block diagram of figure are obtained, the coefficient matrix in formula is as follows:
Different voltage vectors corresponds to different torque and Flux estimation value, combined objective function, to pre- in input set
The magnetic linkage of survey is assessed with torque, choose the smallest switch state of target function value as system subsequent time input.
4) Space Rotating coordinate transform calculates
Fig. 5, which gives, transforms to extension Parker transformation matrices used in double d-q coordinate systems as five phase rest frames.It passes
What sensor detected is the phase current under usual abcde coordinate system, and prediction model is carried out under d-q rotating coordinate system,
It is therefore desirable to first will be under the current transformation to d1-q1 coordinate system and d3-q3 coordinate system of sensor detection.
Claims (3)
1. a kind of five-phase PMSM model prediction method for controlling torque, which comprises the following steps:
Step 1: voltage, electricity according to the Parker rotation transformation of extension by five-phase PMSM under natural system of coordinates
Stream, magnetic linkage, torque physical quantity are respectively mapped under two orthogonal coordinate systems of d1-q1 and d3-q3；
Step 2: classifying to the fundamental space voltage vector of two level, five phase voltage source inventer, and calculate large, medium and small electricity
The amplitude of pressure vector is respectively as follows:
In formula, UL, UM, USRespectively indicate the amplitude of large, medium and small voltage vector；UdcIndicate that the busbar voltage of inverter direct-flow side is big
It is small；
Step 3: dead beat voltage vector predicted value being gone out according to the torque of five-phase induction motor and magnetic linkage error calculation, then extracts nothing
The amplitude and location information of beat predicted voltage vector；
Step 4: the place sector of dead beat voltage vector is determined according to the voltage vector location information being calculated；
Step 5: input set of 4 vectors as PREDICTIVE CONTROL where choosing dead beat voltage vector in sector；
Step 6: prediction and evaluation calculating being carried out to the voltage vector in domination set, chooses the corresponding voltage of minimum value of objective function
Vector is applied to inverter；
The step 3 specifically:
The variable quantity △ δ of required power angle δ is calculated according to the torque error at k moment are as follows:
Te refIndicate the given value of motor speed outer ring torque；δ indicates the power angle between electric machine rotor magnetic linkage, NpIndicate electricity
Machine number of pole-pairs, LdIndicate d-axis inductance, ψfIndicate the magnetic linkage of permanent magnet, ψsIndicate stator magnetic linkage vector；
According to formulaCalculate the stator magnetic linkage error under α β coordinate system:
ψs refIndicate the given value of fundamental wave magnetic linkage；
Component U of the dead beat voltage vector predicted value under two-phase α β rest frameαAnd UβAre as follows:
Finally according to formulaAcquire the amplitude and location information of voltage vector in dead-beat control method:
2. a kind of five-phase PMSM model prediction method for controlling torque as described in claim 1, which is characterized in that
The objective function are as follows:
Wherein, i=m, m+1, m+2, m+3；M indicates the sector number in each sampling period where dead beat voltage vector；
ψsd1-q1 refIndicate the given value of first harmonic spatial stator magnetic linkage；Te(k+1)、ψsd1-q1(k+1) it respectively indicates the torque of k+1 moment and determines
The predicted value of sub- magnetic linkage；ψsd3-q3(k+1) the harmonic prediction value of k+1 moment stator magnetic linkage is indicated；ψsd3-q3 refIt is fixed for harmonic wave space
The given value of sub- magnetic linkage；λ1、λ2Respectively in the weight coefficient of first harmonic spatial and harmonic wave space stator magnetic linkage component.
3. a kind of five-phase PMSM model prediction method for controlling torque as claimed in claim 2, which is characterized in that
I in objective functionmaxAre as follows:
|Ilim| the current limit value for indicating setting, current amplitude after detecting the effect of a certain voltage vector is more than maximum value
Limitation, then do not choose the voltage vector.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710022345.0A CN106803731B (en) | 2017-01-12 | 2017-01-12 | A kind of five-phase PMSM model prediction method for controlling torque |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710022345.0A CN106803731B (en) | 2017-01-12 | 2017-01-12 | A kind of five-phase PMSM model prediction method for controlling torque |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106803731A CN106803731A (en) | 2017-06-06 |
| CN106803731B true CN106803731B (en) | 2018-12-25 |
Family
ID=58984412
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710022345.0A Active CN106803731B (en) | 2017-01-12 | 2017-01-12 | A kind of five-phase PMSM model prediction method for controlling torque |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106803731B (en) |
Families Citing this family (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107231111B (en) * | 2017-07-31 | 2020-03-10 | 福州大学 | Direct torque control method for high load capacity of five-phase permanent magnet synchronous motor |
| CN107994832A (en) * | 2017-11-24 | 2018-05-04 | 东南大学 | A kind of multichannel permanent magnet synchronous motor mixing Direct Torque Control |
| CN108667389B (en) * | 2018-04-02 | 2020-06-09 | 江苏大学 | Virtual voltage vector model-based prediction fault-tolerant control method |
| CN109495055B (en) * | 2018-09-17 | 2022-03-22 | 沈阳工业大学 | Five-phase permanent magnet synchronous motor one-phase open-circuit prediction current control method |
| CN109842338B (en) * | 2019-03-13 | 2021-07-09 | 河北工业大学 | Permanent magnet synchronous motor prediction model parameter online correction method |
| CN110112974B (en) * | 2019-05-13 | 2020-10-23 | 中国科学院深圳先进技术研究院 | Motor control method, controller, storage medium, and motor drive system |
| CN110868111B (en) * | 2019-10-22 | 2021-06-29 | 北方工业大学 | Permanent magnet synchronous motor model prediction control method based on discrete space vector modulation |
| CN110661462B (en) * | 2019-10-25 | 2021-06-08 | 中南大学 | Deadbeat torque control apparatus, method and system |
| CN110838808B (en) * | 2019-11-18 | 2021-03-16 | 东南大学 | Diagnosis-free self-repairing method for open-circuit fault of double three-phase permanent magnet synchronous motor driving system |
| CN111162714A (en) * | 2020-01-03 | 2020-05-15 | 湖南大学 | Multiphase energy storage permanent magnet motor robust prediction torque control method, system and medium considering parameter mismatch |
| CN111510040B (en) | 2020-04-10 | 2023-10-10 | 江苏大学 | Double virtual voltage vector prediction torque control method without weight coefficient for five-phase permanent magnet synchronous motor |
| CN111600524B (en) * | 2020-05-26 | 2023-09-05 | 成都运达轨道交通技术服务有限公司 | Five-phase inverter model prediction current control method based on duty ratio optimization |
| CN111800050B (en) * | 2020-06-18 | 2023-04-14 | 中国石油大学(华东) | Permanent magnet synchronous motor three-vector model prediction torque control method based on voltage vector screening and optimization |
| CN112994553A (en) * | 2021-03-15 | 2021-06-18 | 合肥恒大江海泵业股份有限公司 | Simplified model prediction voltage control method for permanent magnet motor system |
| CN113093542B (en) * | 2021-03-31 | 2022-08-12 | 吉林大学 | Motor torque optimization finite set prediction control parallel computing method |
| CN113037167B (en) * | 2021-05-20 | 2021-09-07 | 西南交通大学 | Five-phase PMSM model prediction control method for improving voltage utilization rate |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6107762A (en) * | 1996-02-06 | 2000-08-22 | S-B Power Tool Company | Speed control device for electrical motor-driven apparatus |
| CN103715973A (en) * | 2014-01-03 | 2014-04-09 | 天津大学 | Space voltage vector pulse width modulation algorithm of five-phase voltage source inverter bridge |
| CN106026833A (en) * | 2016-06-23 | 2016-10-12 | 北京理工大学 | Space vector based control method for five-phase dual-rotor permanent magnet synchronous motor |
-
2017
- 2017-01-12 CN CN201710022345.0A patent/CN106803731B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6107762A (en) * | 1996-02-06 | 2000-08-22 | S-B Power Tool Company | Speed control device for electrical motor-driven apparatus |
| CN103715973A (en) * | 2014-01-03 | 2014-04-09 | 天津大学 | Space voltage vector pulse width modulation algorithm of five-phase voltage source inverter bridge |
| CN106026833A (en) * | 2016-06-23 | 2016-10-12 | 北京理工大学 | Space vector based control method for five-phase dual-rotor permanent magnet synchronous motor |
Non-Patent Citations (2)
| Title |
|---|
| "五相永磁同步电机多目标优化直接转矩控制算法";薛诚等;《中国电机工程学报》;20160320;第36卷(第6期);第1695-1704页 * |
| "永磁同步电机模型预测直接转矩控制";牛峰等;《电机与控制学报》;20151231;第19卷(第12期);第60-67、74页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106803731A (en) | 2017-06-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106803731B (en) | A kind of five-phase PMSM model prediction method for controlling torque | |
| CN100570391C (en) | The real-time detection of permanent-magnetism synchronous motor permanent magnetic field aberration and analytical approach and device thereof | |
| CN103872951B (en) | Permanent magnet synchronous motor torque control method based on sliding mode flux linkage observer | |
| CN105897097A (en) | Current prediction control method and apparatus for permanent magnet synchronous motor (PMSM) | |
| CN107612446B (en) | Built-in permanent magnet synchronous motor model prediction torque control method | |
| CN103956957B (en) | A kind of asynchronous electromotor rotor resistance discrimination method and device | |
| CN108631672A (en) | Meter and the permanent magnet synchronous motor of optimal duty ratio modulation predict flux linkage control method | |
| CN105356810A (en) | Improved direct torque control method for model prediction of permanent magnet synchronous generator | |
| CN105790664B (en) | Permanent magnet synchronous motor model predictive control method | |
| Sain et al. | Modelling and comparative dynamic analysis due to demagnetization of a torque controlled permanent magnet synchronous motor drive for energy-efficient electric vehicle | |
| CN104052360B (en) | Controller for motor | |
| WO2014094317A1 (en) | Active front-end rectifier filter latency compensation method based on model predictive control | |
| CN109302110A (en) | A kind of asynchronous machine broad sense cascade model prediction flux linkage control method and device | |
| CN107919829B (en) | Permanent magnet synchronous motor model prediction torque control method | |
| CN112803861A (en) | Zero-vector-free algorithm for predictive control of three-vector model of permanent magnet synchronous motor | |
| Bohari et al. | Speed tracking of indirect field oriented control induction motor using neural network | |
| CN106385216A (en) | Permanent-magnet synchronous motor current predictive control steady-state error elimination method and system | |
| CN109245640A (en) | A kind of asynchronous motor forecast Control Algorithm and device | |
| Ammar et al. | Adaptive MRAC-based direct torque control with SVM for sensorless induction motor using adaptive observer | |
| CN114301336A (en) | Direct torque prediction control method for permanent magnet synchronous motor | |
| CN103986396B (en) | A kind of asynchronous machine mutual inductance parameter identification method and device | |
| Messali et al. | A resilient adaptive sliding mode observer for sensorless AC salient pole machine drives based on an improved HF injection method | |
| CN108649852B (en) | Permanent magnet synchronous motor control method for improving current loop | |
| Bilal et al. | A Comparative Study of Machine Learning Algorithms for Controlling Torque of Permanent Magnet Synchronous Motors through a Closed Loop System | |
| CN113411028B (en) | Permanent magnet synchronous motor model prediction control method and device, electronic equipment and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |