CN112152522B - Multi-model control method for interaction of non-position sensors in full-speed range of motor - Google Patents
Multi-model control method for interaction of non-position sensors in full-speed range of motor Download PDFInfo
- Publication number
- CN112152522B CN112152522B CN202011046502.XA CN202011046502A CN112152522B CN 112152522 B CN112152522 B CN 112152522B CN 202011046502 A CN202011046502 A CN 202011046502A CN 112152522 B CN112152522 B CN 112152522B
- Authority
- CN
- China
- Prior art keywords
- motor
- speed
- model
- filter
- low
- 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
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
- H02P6/00—Arrangements for controlling synchronous motors or other dynamo-electric motors using electronic commutation dependent on the rotor position; Electronic commutators therefor
- H02P6/08—Arrangements for controlling the speed or torque of a single motor
-
- 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
- H02P6/00—Arrangements for controlling synchronous motors or other dynamo-electric motors using electronic commutation dependent on the rotor position; Electronic commutators therefor
- H02P6/34—Modelling or simulation for control purposes
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Control Of Electric Motors In General (AREA)
Abstract
The invention relates to a motor full-speed range no-position sensor interaction multi-model control method, and belongs to the field of motor control. The method approximates to the actual model of the motor by establishing a low-speed model and a medium-high-speed model of the motor; then, the state quantity of the motor is estimated by constructing model filters of low-speed and medium-high-speed systems of the motor; and then, an interactive multi-model control method is constructed for the discretized model, and a parallel structure of a plurality of sub-filters is adopted to realize the control of the motor. The method well solves the problem of inaccurate modeling or time variation of the motor, thereby remarkably improving the reliability and anti-interference performance of the motor model.
Description
Technical Field
The invention belongs to the field of motor control.
Background
Permanent magnet synchronous motors are widely used because of their high power density, high efficiency, easy control, etc. Vector control techniques for permanent magnet motors are now well developed. The vector control of the conventional high-performance permanent magnet synchronous motor generally requires a position sensor to detect a position signal of a rotor, wherein the position sensor mainly comprises a hall sensor, a photoelectric encoder, a rotary transformer and the like. Although these sensors can provide position signals for the motor system, the position sensor can increase the cost of the system, and the installation of the position sensor can affect the accuracy of the motor control system, and the position sensor belongs to precise electronic components, so that the temperature, humidity and vibration of the environment where the position sensor is located can greatly affect the position sensor, and the reliability of the motor system is reduced. Therefore, in some low cost and space demanding applications, a sensorless control approach may be used.
The permanent magnet synchronous motor position-free control method generally utilizes direct calculation or observer of motor motion relationship, electromagnetic relationship and the like to obtain the rotor position of the motor, so that the motor needs to be modeled. When the motor runs at a high speed, the friction has a small influence on the rotating speed of the motor and can be ignored. When the motor runs at a low speed, the nonlinear relation between friction and the motor rotation speed is strong, the influence on the dynamic and static performances of the system is great, and in order to overcome the adverse effect caused by friction, a low-speed model of the system needs to be accurately and rapidly built. The combination of the low-speed model control method and the medium-high speed model control method is a common method for controlling the motor in the full speed range at present, but the method is easy to oscillate when switching between different algorithms. The full-speed domain low-speed and medium-high-speed switching algorithm mainly has hysteresis switching. The principle of the hysteresis algorithm is to determine the upper limit and the lower limit of a switching interval, when the rotating speed is lower than the lower limit, a low-speed control method is adopted, when the rotating speed is higher than the upper limit, a medium-high-speed control method is adopted, and the original estimation method is kept for the range between the upper limit and the lower limit. However, the method has different errors in estimation at two rotational speed switching moments, so that larger jitter is brought to the system during switching, and smooth switching of rotational speeds cannot be realized.
The interactive multi-model algorithm is a method for approaching an actual model by establishing a plurality of models, adopts a structure with a plurality of sub-filters in parallel, can better solve the problem of inaccurate system modeling or time-varying, considers the transition between the models, and effectively avoids the jump of a filtering estimation result when the models are directly switched.
Disclosure of Invention
The full-speed domain low-speed and medium-high-speed switching algorithm generally brings larger jitter to the system, and smooth switching of the rotating speed cannot be realized. The invention provides a motor full-speed range sensorless interaction multi-model control method by combining a low-speed model and a medium-high-speed model of a motor, which can better solve the problem of inaccurate or time-varying modeling of the motor, realize smooth switching of the rotating speed and improve the reliability and anti-interference performance of the motor.
The invention relates to a motor sensorless multi-model control method, which comprises the following steps:
step one: and (3) establishing a high-speed system model in the motor, selecting motor current i and motor rotating speed omega as state quantities, taking a current measured value as a measurement value, and constructing a high-speed system model filter in the motor.
Step two: and (3) taking the nonlinear relation between friction and motor rotation speed into consideration, establishing a motor low-speed system model, selecting motor current i and motor rotation speed omega as state quantities, taking a current measured value as quantity measurement, and constructing a motor low-speed system model filter.
Step three: and constructing an interactive multi-model control method for the discretized model to obtain a state estimation value, so as to realize the control of the motor.
Wherein the third step is preferably realized by the following steps:
1. input interaction: input interaction is estimated by state at time k-1And probability of mutual transition between modelsA common decision for reinitializing the filter input;
2. according to the characteristics of each system model, selecting a proper filter for filtering to obtain a state estimation value of the modelAnd in each parallel sub-filter, the filtered residual is calculated separately>Variance->And then a corresponding likelihood function is calculated.
3. According to the Bayesian hypothesis testing principle, updating the model probability;
4. and carrying out probability weighted fusion on the results of the high-speed system model filter and the low-speed system model filter in the motor to obtain the overall state estimation value.
The invention approximates to the actual model of the motor by establishing the low-speed model and the medium-high-speed model of the motor, adopts a parallel structure of a plurality of sub-filters, realizes the control of the motor, and better solves the problem of inaccurate system modeling or time-varying, thereby obviously improving the reliability and the anti-interference performance of the system.
Drawings
FIG. 1 is a schematic flow chart of an embodiment of the present invention.
Fig. 2 is a functional block diagram of an interactive multi-model algorithm.
FIG. 3 is a schematic block diagram of a motor full speed range sensorless interactive multi-model control method designed by the invention.
Detailed Description
The present invention will be further described with reference to the accompanying drawings.
The invention provides a motor full-speed range no-position sensor interaction multi-model control method, which comprises the following steps:
step one: and (3) establishing a high-speed system model in the motor, selecting motor current i and motor rotating speed omega as state quantities, taking a current measured value as a measurement value, constructing a high-speed system model filter in the motor, and estimating the state quantity of the motor.
The high-speed mathematical model in the motor is as follows:
wherein K is T Is a moment coefficient, T L Is the load torque, J is the moment of inertia, R S And ω is the motor speed, and B is the viscous coefficient of friction.
And constructing a high-speed model filter in the motor to realize the estimation of the current and the rotating speed of the motor.
The state quantity is as follows:
X＝[iω] T (2)
each state quantity satisfies:
the state equation is recorded as follows:
wherein W is system noise, and is Gaussian white noise with the mean value of 0, and the variance matrix is marked as Q.
Selecting the current measured by the current sensor as a measurement value:
Z＝[i] T (5)
the measurement equation is recorded as:
Z＝h(X,V) (6)
wherein V is measurement noise, gaussian white noise with zero vector as mean value, and variance matrix is marked as R.
And obtaining a state estimation value according to the established state equation and the measurement equation, and realizing the control of the motor.
Step two: and (3) taking the nonlinear relation between friction and motor rotation speed into consideration, establishing a motor low-speed system model, selecting motor current i and motor rotation speed omega as state quantities, taking a current measured value as quantity measurement, constructing a motor low-speed system model filter, and estimating the motor state quantity.
The motor low-speed mathematical model is as follows:
wherein F is C For Coulomb friction coefficient, F S Is the static friction coefficient, V S Is a characteristic velocity coefficient.
And constructing a motor low-speed model filter to realize the estimation of motor current and rotation speed.
The state quantity is as follows:
X＝[iω] T (8)
each state quantity satisfies:
the state equation is recorded as follows:
wherein W is system noise, and is Gaussian white noise with the mean value of 0, and the variance matrix is marked as Q.
Selecting the current measured by the current sensor as a measurement value:
Z＝[i] T (11)
the measurement equation is recorded as:
Z＝h(X,V) (12)
wherein V is measurement noise, gaussian white noise with zero vector as mean value, and variance matrix is marked as R.
And obtaining a state estimation value according to the established state equation and the measurement equation, and realizing the control of the motor.
Step three: and constructing an interactive multi-model control method for the discretized model to obtain a state estimation value, so as to realize the control of the motor. The principle block diagram of the motor sensorless interaction multi-model control method is shown in figure 3, and mainly comprises the following four steps:
(1) Input interaction
Input interaction is estimated by state at time k-1Probability of mutual transition between models->A common decision for reinitializing the filter input.
Assuming that the transition between models is markov chain compliant, thenThe expression of (2) is:
wherein, the liquid crystal display device comprises a liquid crystal display device,representing the probability of transition of the i model to the j model at time k-1; pi ij Representing a priori Markov model transition probability, satisfying +.> The probability of the i model at time k-1. The state and covariance matrix of model j is reinitialized as shown in equation (14):
(2) Model filtering
According to the characteristics of each system model, selecting a proper filter for filtering to obtain a state estimation value of the modelAnd in each parallel sub-filter, the filtered residual is calculated separately>Variance->And then calculates the corresponding likelihood function as shown in equations (15) and (16).
(3) Model probability update
According to the Bayesian hypothesis testing principle, the model update probability is expressed as:
from equations (15) (16) (17), the model probability at time k-1, the model transition probability matrix, and the model likelihood function at time k collectively determine the model probability of the interactive multi-model algorithm.
(4) Output interactions
And carrying out probability weighted fusion on the filter results to obtain an overall state estimation value, wherein the overall state estimation value is shown in a formula (18).
The interactive multi-model estimation method is realized by iterative calculation of the four steps.
Claims (3)
1. The motor full-speed range no-position sensor interaction multi-model control method is characterized by comprising the following steps of:
step one: establishing a high-speed system model in the motor:
in the method, in the process of the invention,K T is a moment coefficient, T L Is the load torque, J is the moment of inertia, R S The motor is a stator winding, omega is the motor rotating speed, and B is the viscous friction coefficient;
selecting a motor current i and a motor rotating speed omega as state quantities, taking a current measured value as a measurement value, constructing a high-speed system model filter in the motor, and estimating the motor rotating speed and the current under the condition of high-speed running in the motor;
step two: according to the nonlinear relation between friction and motor rotation speed, a motor low-speed system model is established:
wherein F is C For Coulomb friction coefficient, F S Is the static friction coefficient, V S Is a characteristic velocity coefficient;
selecting a motor current i and a motor rotating speed omega as state quantities, taking a current measured value as quantity measurement, constructing a motor low-speed system model filter, and estimating the motor rotating speed and the current under the condition of low-speed operation of the motor;
step three: and constructing an interactive multi-model control method for the discretized model to obtain a state estimation value, so as to realize the control of the motor.
2. The motor full speed range sensorless interactive multi-model control method of claim 1, wherein: the third step is as follows:
step 3-1: input interaction is estimated by state at time k-1And probability of mutual transition between models mu ij k-1 A common decision for reinitializing the filter input;
step 3-2: according to the characteristics of each system model, selecting a filter for filtering to obtain a state estimation value of the modelAnd in each parallel sub-filter, the filtered residual is calculated separately>Variance->And then a corresponding likelihood function is calculated;
step 3-3: according to the Bayesian hypothesis testing principle, updating the model probability;
step 3-4: and carrying out probability weighted fusion on the results of the high-speed system model filter and the low-speed system model filter in the motor to obtain an overall state estimation value.
3. The motor full speed range sensorless interactive multi-model control method of claim 1, wherein: based on the built motor low-speed model and the built medium-high speed model, a structure that a motor medium-high speed system model filter and a motor low-speed system model filter are parallel is adopted, so that control of the motor is realized.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011046502.XA CN112152522B (en) | 2020-09-29 | 2020-09-29 | Multi-model control method for interaction of non-position sensors in full-speed range of motor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011046502.XA CN112152522B (en) | 2020-09-29 | 2020-09-29 | Multi-model control method for interaction of non-position sensors in full-speed range of motor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112152522A CN112152522A (en) | 2020-12-29 |
| CN112152522B true CN112152522B (en) | 2023-09-26 |
Family
ID=73895901
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011046502.XA Active CN112152522B (en) | 2020-09-29 | 2020-09-29 | Multi-model control method for interaction of non-position sensors in full-speed range of motor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112152522B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103926875A (en) * | 2014-04-18 | 2014-07-16 | 东南大学 | Method for friction compensation of ball screw feeding system |
| CN104009696A (en) * | 2014-05-08 | 2014-08-27 | 昆明理工大学 | Interactive model reference adaptive speed and stator resistance identification method based on sliding-mode control |
| JP2014220938A (en) * | 2013-05-09 | 2014-11-20 | 三菱電機株式会社 | Motor control device |
| CN110649850A (en) * | 2018-06-27 | 2020-01-03 | 中车株洲电力机车研究所有限公司 | Method for determining stator flux linkage of dual-mode voltage model |
-
2020
- 2020-09-29 CN CN202011046502.XA patent/CN112152522B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2014220938A (en) * | 2013-05-09 | 2014-11-20 | 三菱電機株式会社 | Motor control device |
| CN103926875A (en) * | 2014-04-18 | 2014-07-16 | 东南大学 | Method for friction compensation of ball screw feeding system |
| CN104009696A (en) * | 2014-05-08 | 2014-08-27 | 昆明理工大学 | Interactive model reference adaptive speed and stator resistance identification method based on sliding-mode control |
| CN110649850A (en) * | 2018-06-27 | 2020-01-03 | 中车株洲电力机车研究所有限公司 | Method for determining stator flux linkage of dual-mode voltage model |
Non-Patent Citations (5)
| Title |
|---|
| Jie Li等.Full Speed Range Hybrid Sensorless Control of Permanent Magnet Synchronous Motor Based on Weighting Dual EKF and Signal Injection.2019 22nd International Conference on Electrical Machines and Systems (ICEMS).2019,第1-6页. * |
| M. Tahan、T. Hu.Speed-Sensorless Vector Control of Surface- Mounted PMS Motor Based On Modified Interacting Multiple-Model EKF.2015 IEEE International Electric Machines & Drives Conference (IEMDC.2016,第510-515页. * |
| 张钧天等.基于IMM的永磁电机无速度传感器技术.计算技术与自动化.2016,第35卷(第4期),第39-44页. * |
| 李哲峰等.电压电流混合模型的感应电机无速度传感器矢量控制.北京交通大学学报.2007,第31卷(第2期),第99-102页. * |
| 罗慧等.PMSM自适应观测器算法低速稳定性研究.电气传动.2013,第43卷(第8期),第3-7页. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112152522A (en) | 2020-12-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109873586B (en) | Motor mechanical parameter identification method and system based on high-order sliding-mode observer | |
| CN103532464A (en) | Sensorless vector control system and method for permanent magnet synchronous motor | |
| CN108945373A (en) | Impeller driven system and control method under integrated water | |
| CN112511059B (en) | High-precision position estimation method for permanent magnet synchronous motor | |
| CN115173774B (en) | Permanent magnet synchronous motor sensorless control method and system | |
| CN105429545A (en) | Soft measurement-based control method for sensorless permanent magnet synchronous motor | |
| CN104009696B (en) | Interactive model reference adaptive speed and stator resistance identification method based on sliding-mode control | |
| CN104201967A (en) | Networking time-delay compensation and control method using active-disturbance-rejection control technology for permanent magnet synchronous motor | |
| CN107395083A (en) | PMLSM servo-control system Position And Velocity methods of estimation and device | |
| CN103956953A (en) | Sliding-mode observer based brushless direct-current motor state estimation method | |
| Wang et al. | Rotor position estimation method for permanent magnet synchronous motor based on super-twisting sliding mode observer | |
| CN113630055A (en) | Position sensor-free rocker arm servo control method based on Kalman data fusion | |
| CN106712618A (en) | Method and system for identifying rotational inertia of permanent magnet synchronous motor | |
| CN102510253B (en) | Sensorless controller of bearingless synchronous reluctance motor and method for controlling sensorless controller | |
| CN108599661A (en) | A kind of permanent magnet synchronous motor is without sensor composite control method | |
| CN109194224B (en) | Permanent magnet synchronous motor sensorless control method based on extended state observer | |
| CN112149274A (en) | Online modeling method for multi-axis engraving machine system with dead zone input nonlinearity | |
| CN110768600A (en) | PMSM speed sensorless rotor detection method | |
| CN114944801A (en) | PMSM (permanent magnet synchronous motor) position sensorless control method based on innovation self-adaptive extended Kalman | |
| CN112152522B (en) | Multi-model control method for interaction of non-position sensors in full-speed range of motor | |
| CN110661466B (en) | Quasi-proportional resonance adaptive observer and permanent magnet synchronous motor position estimation method | |
| CN107395080A (en) | Speedless sensor moment controlling system and method based on cascade non-singular terminal sliding mode observer | |
| Liu et al. | Application of a new type of super twist sliding mode control in flywheel energy storage | |
| CN109100532B (en) | Filtering speed measurement method of speed measurement generator based on adaptive interaction dual-mode algorithm | |
| Abd Samat et al. | Implementation of Sugeno FIS in model reference adaptive system adaptation scheme for speed sensorless control of PMSM |
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 | ||
| CB02 | Change of applicant information |
Address after: 210003 No. 346, Zhongshan North Road, Jiangsu, Nanjing Applicant after: 724 Research Institute of China Shipbuilding Corp. Address before: 210003 No. 346, Zhongshan North Road, Jiangsu, Nanjing Applicant before: 724TH RESEARCH INSTITUTE OF CHINA SHIPBUILDING INDUSTRY Corp. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |