CN115833673B - Current injection-based off-line self-learning method for flux linkage of synchronous reluctance motor - Google Patents
Current injection-based off-line self-learning method for flux linkage of synchronous reluctance motor Download PDFInfo
- Publication number
- CN115833673B CN115833673B CN202211635610.XA CN202211635610A CN115833673B CN 115833673 B CN115833673 B CN 115833673B CN 202211635610 A CN202211635610 A CN 202211635610A CN 115833673 B CN115833673 B CN 115833673B
- Authority
- CN
- China
- Prior art keywords
- axis
- current
- reluctance motor
- synchronous reluctance
- flux linkage
- 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
Images
Landscapes
- Control Of Ac Motors In General (AREA)
Abstract
A current injection-based synchronous reluctance motor flux linkage offline self-learning method relates to the technical field of synchronous reluctance motor offline parameter identification. The invention aims to solve the problems that the existing flux linkage offline self-learning method is poor in universality and high in accuracy due to the influence of data point distribution and output post-processing. The invention adopts the injection method of high-frequency zero-mean current, and in the current injection process, the motor is in a quasi-static state, the motor rotating shaft is in a disconnected load state or in a load clamping state, and the load universality is strong. And the current of the motor is controlled by adopting an active disturbance rejection control strategy, and the implementation of a current loop can be designed under the condition that the inductance parameter is unknown. And the flux linkage-current plane mutual saturation data points are obtained by the way of synchronously injecting dq axis current, so that the flux linkage self-learning data point distribution is controllable.
Description
Technical Field
The invention belongs to the technical field of off-line parameter identification of synchronous reluctance motors.
Background
The synchronous reluctance motor (Synchronous Reluctance Motor, synRM) rotor has no permanent magnet and exciting winding, and the motor has the characteristics of simple structure, low cost and good flux weakening and speed expansion, and has been widely paid attention to industrial servo application occasions in recent years. Because the motor operates by utilizing reluctance torque and works in a magnetic saturation state, inductance parameters of the motor are obviously changed along with current working points, so that the motor inductance parameters obtained only through a motor nameplate are difficult to realize high-efficiency utilization of the motor, and accurate motor flux parameters are required to be obtained through a flux linkage offline self-learning strategy before the motor is started.
In a two-phase rotating dq coordinate system, the synchronous reluctance motor generates high-frequency zero-mean torque pulsation when high-frequency zero-mean signals are injected. Because of the filtering characteristic of mechanical inertia on high-frequency torque, the motor rotor is in a static or quasi-static state when the high-frequency zero-mean signal is injected, and self-learning errors caused by the fact that the position angle of the rotor is greatly changed are avoided, so that the self-learning method based on the high-frequency zero-mean signal injection is most commonly used in the field of flux linkage offline parameter identification.
The current high-frequency zero-mean signal injection self-learning method is mainly realized by a bipolar square wave voltage injection mode. The realization thinking is that the polarity of the injection voltage is controlled so as to control the current of the motor to change in the whole current test plane, the voltage and the current are detected in real time in the voltage injection process, and the flux linkage result is obtained by calculation through a voltage equation. However, the purpose of flux linkage self-learning is to obtain the characteristic relation between flux linkage and current, and the voltage injection method can only indirectly control the current, so that the problem of sparse or even missing partial region data exists on the self-learning result, namely a flux linkage-current test plane. The flux linkage results which are not covered in the self-learning process are often estimated through data processing means such as fitting or interpolation, so that errors of the self-learning results are increased. Meanwhile, because the design inductance of the motor and the difference of the running voltage in different application occasions are very large, the amplitude of the injection voltage is also different under the self-learning strategy, and the accurate selection of the injection voltage amplitude can not be directly given only through nameplate parameters, so that the universality of the self-learning strategy is limited.
In summary, the existing flux linkage offline self-learning method has the defects of poor generality and high accuracy influenced by data point distribution and output post-processing, and restricts the application of the flux linkage offline self-learning method in the field of industrial servo.
Disclosure of Invention
The invention aims to solve the problems that the existing flux linkage offline self-learning method is poor in universality and high in accuracy and is influenced by data point distribution and output post-processing, and provides a current injection-based synchronous reluctance motor flux linkage offline self-learning method.
An off-line self-learning method for flux linkage of synchronous reluctance motor based on current injection comprises a self-saturation part and a mutual saturation part,
the self-saturation part is as follows:
injecting given current into d and q axes of synchronous reluctance motor separately in sine signal form to obtain actual d and q axes current and d and q axes flux linkage of each period of synchronous reluctance motor,
at DeltaI 1 For a section length, the d-axis and q-axis currents are segmented within a given current range, and the delta I is calculated by 1 ∈[5％I N ，15％I N ]，I N For rated current marked on the nameplate of the synchronous reluctance motor,
partitioning d-axis flux linkage and q-axis flux linkage according to the interval where the d-axis and q-axis actual current of each period is located, calculating the average value of the d-axis flux linkage and the q-axis flux linkage in each interval, taking the average value as the interval flux linkage where the d-axis flux linkage and the q-axis flux linkage is located, taking the interval intermediate current as the interval current,
constructing a flux linkage self-saturation reference curve by utilizing all interval flux linkages and corresponding interval currents, and completing flux linkage self-saturation offline self-learning;
the mutual saturation part is as follows:
injecting given currents into d axis and q axis of synchronous reluctance motor in sine signal form to obtain d and q axis flux linkage and d and q axis actual currents of synchronous reluctance motor in each period,
the d and q axis flux linkages of each period are marked in a d and q axis three-dimensional coordinate system respectively, the three axes of the d axis three-dimensional coordinate system are d and q axis actual currents and d axis flux linkages respectively, the three axes of the q axis three-dimensional coordinate system are d and q axis actual currents and q axis flux linkages respectively,
the d and q axes are respectively taken as two axes of a plane coordinate system, and a current data plane is established, and delta I is used 2 For a section length, respectively segmenting d-axis and q-axis given currents in a given current range to obtain a current data grid in a current data plane, wherein the current data grid comprises a current data gridNodes in the grid are intersection points of end points of each interval of d and q axes, and delta I is as follows 2 ∈(0，5％I N ]，I N For rated current marked on the nameplate of the synchronous reluctance motor,
and (3) corresponding nodes in the current data grid to a d-axis three-dimensional coordinate system and a q-axis three-dimensional coordinate system, calculating flux linkage corresponding to each node in the current data grid in a two-dimensional interpolation mode, and constructing a flux linkage mutual saturation reference surface by using flux linkages corresponding to all nodes to finish flux linkage mutual saturation offline self-learning.
Further, the injection frequencies of the d-axis and q-axis given currents are f/N respectively c And f, the amplitude of the given current is in the range of 110% I max ～120％I max ，
Wherein I is max For a given current peak value and
f is the rated frequency marked on the nameplate of the synchronous reluctance motor, N c Is a division factor and is a positive odd number greater than 1.
Further, a current controller is designed to enable the actual current of the synchronous reluctance motor to follow a given current under the condition of inaccurate inductance parameters, the current controller comprises a proportional controller and an extended state observer, and the specific process for controlling the synchronous reluctance motor by using the current controller is as follows:
the current i is given to the current moment by using the proportional controller * And observing current
Proportional amplification is carried out on the difference value of the two to obtain a linear state error feedback result u 0 ，
Line state error feedback result u using the observation of the extended state observer and the known term of the synchronous reluctance motor voltage equation 0 Compensating, and multiplying the compensation result by d-axis inductance L d The control quantity u is obtained and the control quantity u,
the synchronous reluctance motor is controlled by the control quantity u such that the actual current of the synchronous reluctance motor can follow the given current.
Further, the above-mentioned extended state observer model is as follows:
wherein ε r For r-axis observed current to actual current error, r-axis represents d-axis or q-axis,
observing the current for the r axis, i r For r-axis actual current, +.>
For the observation value of the expansion state of the r axis, p is a differential operator, u r For r-axis voltage, f r Known term for the r-axis of the voltage equation, beta 1 And beta 2 The first and second gain coefficients of the extended state observer, respectively.
Further, the specific method for obtaining the d and q axis actual currents of the kth period is as follows:
collecting abc three-phase current i of kth period of synchronous reluctance motor a (k)、i b (k)、i c (k)，
Obtaining d-axis and q-axis actual current i according to the following d (k) And i q (k)：
Wherein θ e (k) Rotor position angle for the kth cycle of the synchronous reluctance motor.
Further, in the above-mentioned self-saturation portion,
in the kth period, the d axis of the synchronous reluctance motor is orientedA given current injected separately
The expression of (2) is:
a given current individually injected to the q-axis of the synchronous reluctance motor in the kth period
The expression of (2) is:
wherein I is amp For a given current amplitude, N d And N q The injection period of the current is respectively given to the d axis and the q axis, f d And f q The injection frequency of the current is respectively given to the d axis and the q axis, f c Is the control frequency of the synchronous reluctance motor.
Further, in the self-saturation portion, when a given current is individually injected to the d and q axes of the synchronous reluctance motor,
the d-axis flux linkage ψ of the kth period obtained d_single (k) The method comprises the following steps:
ψ d_single (k)＝ψ d_single (k-1)+[u d,d (k-1)-R s i d_single (k)]/f c ，
the q-axis flux linkage ψ of the kth period obtained q_single (k) The method comprises the following steps:
ψ q_single (k)＝ψ q_single (k-1)+[u q,q (k-1)-R s i q_single (k)]/f c ，
wherein, psi is d_single (k-1) and ψ q_single (k-1) are d-axis flux linkages and q-axis flux linkages of the (k-1) th period when a given current is independently injected into two axes of the synchronous reluctance motor, u d,d (k-1) and u q,q (k-1) d-axis and q-axis of the (k-1) -th period when a given current is injected separately to both axes of the synchronous reluctance motorVoltage, i d_single (k) And i q_single (k) Respectively, d and q-axis actual currents of the kth period when the given current is independently injected into two axes of the synchronous reluctance motor, f c For controlling frequency of synchronous reluctance motor, R s Is the stator resistance of a synchronous reluctance motor.
Further, in the above-mentioned mutual saturation portion, the nth given current is injected simultaneously to the d and q axes of the synchronous reluctance motor
And->
The expression is as follows:
wherein k represents the period of current and k is less than or equal to f c /f d ，N＝1,2,...,N c ，N c Is a frequency division coefficient and is a positive odd number greater than 1, f d And f q The injection frequency of the current is respectively given to the d axis and the q axis, f c For controlling frequency of synchronous reluctance motor, I amp For a given magnitude of the current flow,
is the phase shift angle at the time of q-axis current injection, and +.>
Further, the d and q axis flux linkage ψ of the kth period when the current is simultaneously injected to the two axes of the synchronous reluctance motor d_double (k)、ψ q_double (k) The expression of (2) is:
ψ d_double (k)＝ψ d_double (k-1)+[u d,dq (k-1)-R s i d,dq (k)]/f c
ψ q_double (k)＝ψ q_double (k-1)+[u q,dq (k-1)-R s i q,dq (k)]/f c ，
wherein, psi is d_double (k-1) and ψ q_double (k-1) are d-axis flux linkages and q-axis flux linkages of the kth-1 th period of the synchronous reluctance motor when current is injected simultaneously, u d,dq (k-1) and u q,dq (k-1) d and q axis voltages, i, respectively, of the kth-1 th cycle of the synchronous reluctance motor when current is simultaneously injected d,dq (k) And i q,dq (k) D and q-axis actual currents of the kth period when the synchronous reluctance motor simultaneously injects current, f c For controlling frequency of synchronous reluctance motor, R s Is the stator resistance of a synchronous reluctance motor.
Further, a node Q (i d ,i q ) Corresponding d-axis flux linkage ψ d,dq (i d ,i q )：
Wherein, psi is d,dq (i d ,i q1 ) Sum phi d,dq (i d ,i q2 ) Data points Q (i) d ,i q1 ) And Q (i) d ,i q2 ) Corresponding d-axis flux linkage, Q (i) d ,i q1 ) And Q (i) d ,i q2 ) As an intermediate interpolation point of the interpolation points,
ψ d_double (i d1 ,i q1 )、ψ d_double (i d2 ,i q2 )、ψ d_double (i d2 ,i q1 ) Sum phi d_double (i d1 ,i q2 ) Data points Q (i) marked in the current data plane respectively d1 ,i q1 )、Q(i d2 ,i q2 )、Q(i d2 ,i q1 ) And Q (i) d1 ,i q2 ) Corresponding d-axis flux linkage, data point Q (i d1 ,i q1 )、Q(i d2 ,i q2 )、Q(i d2 ,i q1 ) And Q (i) d1 ,i q2 ) Respectively, four different directions from the node Q (i d ,i q ) The four nearest data points, i d And i q Respectively nodes Q (i d ,i q ) Corresponding d, q axis currents, i d1 And i q1 Respectively nodes Q (i d1 ,i q1 ) Corresponding d, q axis currents, i d2 And i q2 Respectively nodes Q (i d2 ,i q2 ) Corresponding d, q-axis currents.
According to the invention, a high-frequency zero-mean current injection method is adopted, the injection current information is obtained by the motor nameplate parameters, and in the current injection process, the motor is in a quasi-static state, so that the motor rotating shaft can be in a load disconnection state or a load clamping state, and the load universality is strong.
According to the invention, the motor current is controlled by adopting an active disturbance rejection control strategy, and the implementation of the current loop can be designed under the condition that the inductance parameter is unknown.
The method obtains flux linkage-current plane mutual saturation data points in a dq axis current simultaneous injection mode, the distribution of the data points is determined by the track planning of the injection current, the flux linkage self-learning data point distribution is controllable, the problems of irregular current data point distribution and poor uniformity in a voltage injection mode are effectively solved, and the error of a self-learning result is reduced.
Drawings
FIG. 1 is a schematic block diagram of parameter identification of a synchronous reluctance motor based on the method of the invention, wherein SynRM is the synchronous reluctance motor, a compensation function module is an inverter nonlinear compensation module, and a flux linkage calculation module is a flux linkage post-processing calculation module;
FIG. 2 is a schematic block diagram of a current loop controller designed with q-axis current as the controlled object;
FIG. 3 (a) is a graph showing the injection law of dq-axis current during mutual saturation learning;
FIG. 3 (b) is a current plan view covered by the self-learning process under the dual current injection rule shown in FIG. 3 (a);
FIG. 4 is a waveform diagram of a current injection experiment of the flux linkage self-learning method;
FIG. 5 is a schematic diagram of self-learning results of a self-saturation flux linkage obtained by a single current injection method according to the present invention, wherein a scatter diagram is a flux linkage result obtained after current injection, and a graph is a flux linkage reference value obtained by post-processing based on the injection result, (a) represents a d-axis, and (b) represents a q-axis;
fig. 6 is a schematic diagram of a cross-saturation flux linkage self-learning result obtained by the dual current injection method according to the present invention, wherein a scatter diagram is a flux linkage result obtained after current injection, and a grid diagram is a flux linkage reference value obtained by post-processing based on the injection result, (a) represents a d-axis, and (b) represents a q-axis.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention. It should be noted that, without conflict, the embodiments of the present invention and features of the embodiments may be combined with each other.
According to the current injection-based synchronous reluctance motor flux linkage offline self-learning method, a rated current I is applied N The injection current of the motor=8.2a is chosen to be 13A to ensure that the self-learning current can cover the current range of the motor in normal operation. For the motor of the present embodiment with a rated frequency of 100Hz, the d-axis is set to be a large inductance axis, and the q-axis is set to be a small inductance axis. In order to ensure the following property of injection current and the uniformity of distribution of a test plane, the q-axis injection frequency is selected to be 100Hz of rated frequency of a motor, the d-axis injection frequency is selected to be 20Hz, and the frequency division coefficient N c ＝5。
Referring to fig. 2, a current controller that is independent of motor inductance parameters is designed with unknown motor flux (inductance) parameters so that the actual current of the synchronous reluctance motor can follow a given current under inaccurate inductance parameter conditions. The current controller includes a proportional controller and an extended state observer. The specific design process of the extended state observer (Extended State Observer, ESO) is as follows:
dividing a voltage equation into an input term, a known term and a disturbance term, and rewriting a synchronous reluctance motor voltage equation into a current form:
wherein p is a differential operator, R s For stator resistance, ω of synchronous reluctance motor e For electric angular velocity, L d And L q Respectively d and q axis inductances, u d And u q D and q axis voltages, respectively, E is the extended back emf, expressed as: e= (L) d -L q )·(ω e ·i d -pi q )。
The perturbation term is an extended back emf related term expressed as: E/L d The extended state observer model built with this perturbation as an extended state is as follows:
wherein ε r For r-axis observed current to actual current error, r-axis represents d-axis or q-axis,
observing the current for the r axis, i r For r-axis actual current, +.>
As the observation value of the expansion state of the r axis, u r For r-axis voltage, f r Known term for the r-axis of the voltage equation, beta 1 And beta 2 The first and second gain coefficients of the extended state observer, respectively.
According to the design principle of the extended state observer, selecting the characteristic value corresponding to the error state equation as the bandwidth of ESO, and calculating the design value of the parameter in ESO:
This embodiment includes a self-saturation portion and a mutual saturation portion.
The self-saturation part is as follows:
the d and q axes of the synchronous reluctance motor are respectively and independently injected with given currents in the form of sine signals. Specifically, a given current is independently injected to the d-axis of the synchronous reluctance motor in the kth period
The expression of (2) is:
a given current individually injected to the q-axis of the synchronous reluctance motor in the kth period
The expression of (2) is:
wherein I is amp For a given current amplitude, N d And N q The injection period of the current is respectively given to the d axis and the q axis, f d And f q The injection frequency of the current is respectively given to the d axis and the q axis, f c Control for synchronous reluctance machinesAnd (5) frequency control.
When a given current is injected, a current controller is utilized to control the synchronous reluctance motor, and the specific process is as follows:
the current i is given to the current moment by using the proportional controller * And observing current
Proportional amplification is carried out on the difference value of the two to obtain a linear state error feedback result u 0 。
Line state error feedback result u using the observation of the extended state observer and the known term of the synchronous reluctance motor voltage equation 0 Compensating, and multiplying the compensation result by d-axis inductance L d The control amount u is obtained.
The synchronous reluctance motor is controlled by the control quantity u such that the actual current of the synchronous reluctance motor can follow the given current.
In the process, the abc three-phase current i of the kth period of the synchronous reluctance motor is collected a (k)、i b (k)、i c (k) The d-axis and q-axis actual currents i are obtained according to the following formula d (k) And i q (k)：
Wherein θ e (k) Rotor position angle for the kth cycle of the synchronous reluctance motor.
The d-axis flux linkage ψ of the kth period is obtained according to the following d_single (k)：
ψ d_single (k)＝ψ d_single (k-1)+[u d,d (k-1)-R s i d_single (k)]/f c 。
The q-axis flux linkage ψ of the kth period is obtained according to the following q_single (k)：
ψ q_single (k)＝ψ q_single (k-1)+[u q,q (k-1)-R s i q_single (k)]/f c 。
Wherein, psi is d_single (k-1) and ψ q_single (k-1) are d-axis flux linkages and q-axis flux linkages of the (k-1) th period when a given current is independently injected into two axes of the synchronous reluctance motor, u d,d (k-1) and u q,q (k-1) the d-axis voltage and the q-axis voltage of the kth-1 th period when the given current is independently injected into two axes of the synchronous reluctance motor, i d_single (k) And i q_single (k) The actual currents of d and q axes of the kth period when the given currents are independently injected into the two axes of the synchronous reluctance motor are respectively.
The amplitude of the given current ranges from 110% I max ～120％I max ，I max For a given current peak value and
at DeltaI 1 For a section length, the d-axis and q-axis currents are segmented within a given current range, and the delta I is calculated by 1 ∈[5％I N ，15％I N ]。
Partitioning d-axis flux linkages and q-axis flux linkages according to intervals where d-axis and q-axis actual currents of each period are located, calculating average values of the d-axis flux linkages and the q-axis flux linkages in each interval, taking the average values as interval flux linkages where the d-axis flux linkages and the q-axis flux linkages are located, and taking interval intermediate currents as interval currents.
And constructing a flux linkage self-saturation reference curve by utilizing all the interval flux linkages and the corresponding interval currents, and completing the flux linkage self-saturation offline self-learning as shown in fig. 5.
The mutual saturation part is as follows:
the d-axis and q-axis of the synchronous reluctance motor are injected with given currents at the same time in the form of sine signals, the injection current waveforms are the double-current injection waveforms marked in fig. 4, the injection coverage range is shown in fig. 3, and it can be seen that the current track at the moment almost covers the whole mutual saturation learning current plane. N-th given current simultaneously injected to d-axis and q-axis of synchronous reluctance motor
And->
The expression is as follows:
wherein k represents the period of current and k is less than or equal to f c [f d ，N＝1,2,...,N c ，N c Is a frequency division coefficient and is a positive odd number greater than 1, f d And f q The injection frequency of the current is respectively given to the d axis and the q axis, f c For controlling frequency of synchronous reluctance motor, I amp For a given magnitude of the current flow,
is the phase shift angle at the time of q-axis current injection, and +.>
When a given current is injected, a current controller is utilized to control the synchronous reluctance motor, and the specific process is as follows:
when injecting a given current, the current is given to the current moment i by using the proportional controller * And observing current
Proportional amplification is carried out on the difference value of the two to obtain a linear state error feedback result u 0 。
Line state error feedback result u using the observation of the extended state observer and the known term of the synchronous reluctance motor voltage equation 0 Compensating, and multiplying the compensation result by d-axis inductance L d The control amount u is obtained.
The synchronous reluctance motor is controlled by the control quantity u such that the actual current of the synchronous reluctance motor can follow the given current.
In the above process, d and q axis flux linkages and d and q axis actual currents of the synchronous reluctance motor in each period are obtained. D, q axis flux linkage ψ of kth period when current is simultaneously injected to two axes of synchronous reluctance motor d_double (k)、ψ q_double (k) The expression of (2) is:
ψ d_double (k)＝ψ d_double (k-1)+[u d,dq (k-1)-R s i d,dq (k)]/f c
ψ q_double (k)＝ψ q_double (k-1)+[u q,dq (k-1)-R s i q,dq (k)]/f c ，
wherein, psi is d_double (k-1) and ψ q_double (k-1) are d-axis flux linkages and q-axis flux linkages of the kth-1 th period of the synchronous reluctance motor when current is injected simultaneously, u d,dq (k-1) and u q,dq (k-1) d and q axis voltages, i, respectively, of the kth-1 th cycle of the synchronous reluctance motor when current is simultaneously injected d,dq (k) And i q,dq (k) D and q-axis actual currents of the kth period when the synchronous reluctance motor simultaneously injects current, f c Is the control frequency of the synchronous reluctance motor.
The d-axis flux linkage and the q-axis flux linkage of each period are respectively marked in a d-axis three-dimensional coordinate system and a q-axis three-dimensional coordinate system, wherein the three axes of the d-axis three-dimensional coordinate system are d-axis actual current and q-axis actual current and d-axis flux linkage respectively, and the three axes of the q-axis three-dimensional coordinate system are d-axis actual current and q-axis flux linkage respectively.
The d and q axes are respectively taken as two axes of a plane coordinate system, and a current data plane is established, and delta I is used 2 For a section length, respectively segmenting d-axis and q-axis given currents in a given current range to obtain a current data grid in a current data plane, wherein nodes in the current data grid are intersections of end points of each section of the d-axis and the q-axis, and the delta I is calculated by the current data grid 2 ∈(0，5％I N ]，I N The rated current is marked on the nameplate of the synchronous reluctance motor.
And (3) corresponding the nodes in the current data grid to a d-axis three-dimensional coordinate system and a q-axis three-dimensional coordinate system, and calculating the flux linkage corresponding to each node in the current data grid by adopting a two-dimensional interpolation mode. Specific:
calculating a node Q (i) in the current data grid according to d ,i q ) Corresponding d-axis flux linkage ψ d,dq (i d ,i q )：
Wherein, psi is d,dq (i d ,i q1 ) Sum phi d,dq (i d ,i q2 ) Data points Q (i) d ,i q1 ) And Q (i) d ,i q2 ) Corresponding d-axis flux linkage, Q (i) d ,i q1 ) And Q (i) d ,i q2 ) As an intermediate interpolation point of the interpolation points,
ψ d_double (i d1 ,i q1 )、ψ d_double (i d2 ,i q2 )、ψ d_double (i d2 ,i q1 ) Sum phi d_double (i d1 ,i q2 ) Data points Q (i) marked in the current data plane respectively d1 ,i q1 )、Q(i d2 ,i q2 )、Q(i d2 ,i q1 ) And Q (i) d1 ,i q2 ) Corresponding d-axis flux linkage, data point Q (i d1 ,i q1 )、Q(i d2 ,i q2 )、Q(i d2 ,i q1 ) And Q (i) d1 ,i q2 ) Respectively, four different directions from the node Q (i d ,i q ) The four nearest data points, i d And i q Respectively nodes Q (i d ,i q ) Corresponding d, q axis currents, i d1 And i q1 Respectively nodes Q (i d1 ,i q1 ) Corresponding d, q axis currents, i d2 And i q2 Respectively nodes Q (i d2 ,i q2 ) Corresponding d, q-axis currents. i.e d1 And i d2 Distance i in two different directions d The two most recent values, i q1 And i q2 Respectively are distances i in two different directions q The last two values.
And constructing a flux linkage mutual saturation reference surface by using flux linkages corresponding to all nodes, and completing flux linkage mutual saturation offline self-learning as shown in fig. 6.
Therefore, equidistant flux linkage data points considering mutual saturation can be obtained and used for looking up a motor flux linkage parameter table in the synchronous reluctance motor control process, and motor control performance is optimized.
The interpolation expression result of the flux linkage self-learning result is obtained by adopting the interpolation method, namely an equidistant flux linkage mutual saturation plane obtained by interpolation as shown in fig. 6. It can be seen that the interpolation plane of the mutual saturation test result is uniform and smooth, and the coincidence degree of the plane and the mutual saturation test point is high. The experimental result shows that the current injection self-learning method can enhance the universality of the self-learning method and effectively improve the accuracy of self-learning.
Although the invention herein has been described with reference to particular embodiments, it is to be understood that these embodiments are merely illustrative of the principles and applications of the present invention. It is therefore to be understood that numerous modifications may be made to the illustrative embodiments and that other arrangements may be devised without departing from the spirit and scope of the present invention as defined by the appended claims. It should be understood that the different dependent claims and the features described herein may be combined in ways other than as described in the original claims. It is also to be understood that features described in connection with separate embodiments may be used in other described embodiments.
Claims (8)
1. A synchronous reluctance motor flux linkage offline self-learning method based on current injection is characterized by comprising a self-saturation part and a mutual saturation part,
the self-saturation part is as follows:
injecting given current into d and q axes of synchronous reluctance motor separately in sine signal form to obtain actual d and q axes current and d and q axes flux linkage of each period of synchronous reluctance motor,
at DeltaI 1 For a section length, the d-axis and q-axis currents are segmented within a given current range, and the delta I is calculated by 1 ∈[5％I N ，15％I N ]，I N For rated current marked on the nameplate of the synchronous reluctance motor,
partitioning d-axis flux linkage and q-axis flux linkage according to the interval where the d-axis and q-axis actual current of each period is located, calculating the average value of the d-axis flux linkage and the q-axis flux linkage in each interval, taking the average value as the interval flux linkage where the d-axis flux linkage and the q-axis flux linkage is located, taking the interval intermediate current as the interval current,
constructing a flux linkage self-saturation reference curve by utilizing all interval flux linkages and corresponding interval currents, and completing flux linkage self-saturation offline self-learning;
the mutual saturation part is as follows:
injecting given currents into d axis and q axis of synchronous reluctance motor in sine signal form to obtain d and q axis flux linkage and d and q axis actual currents of synchronous reluctance motor in each period,
the d and q axis flux linkages of each period are marked in a d and q axis three-dimensional coordinate system respectively, the three axes of the d axis three-dimensional coordinate system are d and q axis actual currents and d axis flux linkages respectively, the three axes of the q axis three-dimensional coordinate system are d and q axis actual currents and q axis flux linkages respectively,
the d and q axes are respectively taken as two axes of a plane coordinate system, and a current data plane is established, and delta I is used 2 For a section length, respectively segmenting d-axis and q-axis given currents in a given current range to obtain a current data grid in a current data plane, wherein nodes in the current data grid are intersections of end points of each section of the d-axis and the q-axis, and the delta I is calculated by the current data grid 2 ∈(0，5％I N ]，I N For rated current marked on the nameplate of the synchronous reluctance motor,
corresponding nodes in the current data grid to d and q axis three-dimensional coordinate systems, calculating flux linkage corresponding to each node in the current data grid by adopting a two-dimensional interpolation mode, and constructing flux linkage mutual saturation reference surfaces by using flux linkage corresponding to all nodes to finish flux linkage mutual saturation offline self-learning;
in the self-saturation portion of the liquid crystal,
a given current independently injected to the d-axis of the synchronous reluctance motor in the kth period
The expression of (2) is:
a given current individually injected to the q-axis of the synchronous reluctance motor in the kth period
The expression of (2) is:
wherein I is amp For a given current amplitude, N d And N q The injection period of the current is respectively given to the d axis and the q axis, f d And f q The injection frequency of the current is respectively given to the d axis and the q axis, f c The control frequency of the synchronous reluctance motor;
in the mutual saturation part, the Nth given current is injected into the d and q axes of the synchronous reluctance motor simultaneously
And->
The expression is as follows:
wherein k represents the period of current and k is less than or equal to f c /f d ，N＝1,2,...,N c ，N c Is a frequency division coefficient and is a positive odd number greater than 1, f d And f q The injection frequency of the current is respectively given to the d axis and the q axis, f c For controlling frequency of synchronous reluctance motor, I amp For a given magnitude of the current flow,
is the phase shift angle at the time of q-axis current injection, and +.>
2. The method for offline self-learning of flux linkage of synchronous reluctance motor based on current injection according to claim 1, wherein the injection frequencies of the d-axis and q-axis given currents are f respectively d ＝f/N c And f q =f, the amplitude range of the given current is 110% i max ～120％I max ，
Wherein I is max For a given current peak value and
f is the rated frequency marked on the nameplate of the synchronous reluctance motor, N c Is a division factor and is a positive odd number greater than 1.
3. The off-line self-learning method of synchronous reluctance motor flux linkage based on current injection according to claim 1, wherein the current controller is designed so that the actual current of the synchronous reluctance motor can follow a given current under the condition of inaccurate inductance parameter, the current controller comprises a proportional controller and an extended state observer, and the specific process of controlling the synchronous reluctance motor by using the current controller is as follows:
the current i is given to the current moment by using the proportional controller * And observing current
Proportional amplification is carried out on the difference value of the two to obtain a linear state error feedback result u 0 ，
Line state error feedback result u using the observation of the extended state observer and the known term of the synchronous reluctance motor voltage equation 0 Compensating, and multiplying the compensation result by d-axis inductance L d The control quantity u is obtained and the control quantity u,
the synchronous reluctance motor is controlled by the control quantity u such that the actual current of the synchronous reluctance motor can follow the given current.
4. A current injection based synchronous reluctance machine flux linkage offline self-learning method according to claim 3, characterized in that the extended state observer model is as follows:
wherein ε r For r-axis observed current to actual current error, r-axis represents d-axis or q-axis,
observing the current for the r axis, i r For r-axis actual current, +.>
For the observation value of the expansion state of the r axis, p is a differential operator, u r For r-axis voltage, f r Known term for the r-axis of the voltage equation, beta 1 And beta 2 The first and second gain coefficients of the extended state observer, respectively.
5. The off-line self-learning method of synchronous reluctance motor flux linkage based on current injection according to claim 1, wherein the specific method for obtaining d and q axis actual currents of the kth period is as follows:
collecting abc three-phase current i of kth period of synchronous reluctance motor a (k)、i b (k)、i c (k)，
Obtaining d-axis and q-axis actual current i according to the following d (k) And i q (k)：
Wherein θ e (k) Rotor position angle for the kth cycle of the synchronous reluctance motor.
6. The method for offline self-learning of synchronous reluctance motor flux linkage based on current injection according to claim 1, wherein, in the self-saturation part, when a given current is individually injected to d and q axes of the synchronous reluctance motor,
the d-axis flux linkage ψ of the kth period obtained d_single (k) The method comprises the following steps:
ψ d_single (k)＝ψ d_single (k-1)+[u d,d (k-1)-R s i d_single (k)]/f c ，
the q-axis flux linkage ψ of the kth period obtained q_single (k) The method comprises the following steps:
ψ q_single (k)＝ψ q_single (k-1)+[u q,q (k-1)-R s i q_single (k)]/f c ，
wherein, psi is d_single (k-1) and ψ q_single (k-1) are d-axis flux linkages and q-axis flux linkages of the (k-1) th period when a given current is independently injected into two axes of the synchronous reluctance motor, u d,d (k-1) and u q,q (k-1) the d-axis voltage and the q-axis voltage of the kth-1 th period when the given current is independently injected into two axes of the synchronous reluctance motor, i d_single (k) And i q_single (k) Respectively, d and q-axis actual currents of the kth period when the given current is independently injected into two axes of the synchronous reluctance motor, f c For controlling frequency of synchronous reluctance motor, R s Is the stator resistance of a synchronous reluctance motor.
7. The off-line self-learning method of synchronous reluctance motor flux linkage based on current injection according to claim 1, wherein the d-axis flux linkage ψ and the q-axis flux linkage ψ of the kth period are obtained when current is simultaneously injected to two axes of the synchronous reluctance motor d_double (k)、ψ q_double (k) The expression of (2) is:
wherein, psi is d_double (k-1) and ψ q_double (k-1) are d and q axis flux linkages of the kth-1 th period when the synchronous reluctance motor is simultaneously injected with current,u d,dq (k-1) and u q,dq (k-1) d and q axis voltages, i, respectively, of the kth-1 th cycle of the synchronous reluctance motor when current is simultaneously injected d,dq (k) And i q,dq (k) Respectively the actual current of d and q axes of the kth-1 th period when the synchronous reluctance motor simultaneously injects current, f c For controlling frequency of synchronous reluctance motor, R s Is the stator resistance of a synchronous reluctance motor.
8. The method of off-line self-learning of current injection based synchronous reluctance motor flux linkage of claim 7 wherein the node Q (i d ,i q ) Corresponding d-axis flux linkage ψ d,dq (i d ,i q )：
Wherein, psi is d,dq (i d ,i q1 ) Sum phi d,dq (i d ,i q2 ) Data points Q (i) d ,i q1 ) And Q (i) d ,i q2 ) Corresponding d-axis flux linkage, Q (i) d ,i q1 ) And Q (i) d ,i q2 ) As an intermediate interpolation point of the interpolation points,
ψ d_double (i d1 ,i q1 )、ψ d_double (i d2 ,i q2 )、ψ d_double (i d2 ,i q1 ) Sum phi d_double (i d1 ,i q2 ) Data points Q (i) marked in the current data plane respectively d1 ,i q1 )、Q(i d2 ,i q2 )、Q(i d2 ,i q1 ) And Q (i) d1 ,i q2 ) Corresponding d-axis flux linkage, data point Q (i d1 ,i q1 )、Q(i d2 ,i q2 )、Q(i d2 ,i q1 ) And Q (i) d1 ,i q2 ) Respectively, four different directions from the node Q (i d ,i q ) The nearest data point, i d And i q Respectively nodes Q (i d ,i q ) Corresponding d, q axis currents, i d1 And i q1 Respectively nodes Q (i d1 ,i q1 ) Corresponding d, q axis currents, i d2 And i q2 Respectively nodes Q (i d2 ,i q2 ) Corresponding d, q-axis currents.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211635610.XA CN115833673B (en) | 2022-12-19 | 2022-12-19 | Current injection-based off-line self-learning method for flux linkage of synchronous reluctance motor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211635610.XA CN115833673B (en) | 2022-12-19 | 2022-12-19 | Current injection-based off-line self-learning method for flux linkage of synchronous reluctance motor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115833673A CN115833673A (en) | 2023-03-21 |
| CN115833673B true CN115833673B (en) | 2023-07-14 |
Family
ID=85516781
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211635610.XA Active CN115833673B (en) | 2022-12-19 | 2022-12-19 | Current injection-based off-line self-learning method for flux linkage of synchronous reluctance motor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115833673B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117424505B (en) * | 2023-12-19 | 2024-03-12 | 上海精泰技术有限公司 | Control method, equipment and medium of synchronous reluctance motor |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110995075A (en) * | 2019-10-24 | 2020-04-10 | 深圳市汇川技术股份有限公司 | Saturation model identification method, system, equipment and computer readable storage medium |
| CN111917350A (en) * | 2020-06-22 | 2020-11-10 | 东南大学 | Multi-parameter identification method for flux linkage adjustable permanent magnet auxiliary synchronous reluctance motor |
| CN113783488A (en) * | 2021-08-04 | 2021-12-10 | 华中科技大学 | Permanent magnet synchronous motor full-parameter identification method and permanent magnet synchronous motor system |
| CN115411991A (en) * | 2022-10-09 | 2022-11-29 | 哈尔滨工业大学 | Inverter nonlinear self-learning method of synchronous reluctance motor driver |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8742767B2 (en) * | 2009-11-03 | 2014-06-03 | Rockwell Automation Technologies, Inc. | Parameter estimation system and method for an induction motor |
-
2022
- 2022-12-19 CN CN202211635610.XA patent/CN115833673B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110995075A (en) * | 2019-10-24 | 2020-04-10 | 深圳市汇川技术股份有限公司 | Saturation model identification method, system, equipment and computer readable storage medium |
| CN111917350A (en) * | 2020-06-22 | 2020-11-10 | 东南大学 | Multi-parameter identification method for flux linkage adjustable permanent magnet auxiliary synchronous reluctance motor |
| CN113783488A (en) * | 2021-08-04 | 2021-12-10 | 华中科技大学 | Permanent magnet synchronous motor full-parameter identification method and permanent magnet synchronous motor system |
| CN115411991A (en) * | 2022-10-09 | 2022-11-29 | 哈尔滨工业大学 | Inverter nonlinear self-learning method of synchronous reluctance motor driver |
Non-Patent Citations (1)
| Title |
|---|
| 永磁同步电机离线参数辨识方法研究;张瑞峰等;《机车电传动》(第3期);第18-23页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115833673A (en) | 2023-03-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107359837B (en) | Permanent magnet synchronous motor torque control method based on sliding mode observer and active disturbance rejection | |
| CN109787524B (en) | Permanent magnet synchronous motor parameter online identification method | |
| CN110048655B (en) | Counter potential fundamental wave extraction permanent magnet synchronous motor position sensorless control system | |
| CN103701386B (en) | Based on the acquisition methods of the full rank flux observer of the Speed Sensorless Induction Motor of observation magnetic linkage error | |
| CN109150029B (en) | Permanent magnet synchronous motor position sensorless control method based on smooth nonsingular terminal sliding-mode observer | |
| CN109768753B (en) | Novel sliding-mode observer position-sensorless permanent magnet synchronous motor model prediction control method | |
| CN106911280A (en) | Permanent-magnetism linear motor method for controlling position-less sensor based on new disturbance observer | |
| CN115833673B (en) | Current injection-based off-line self-learning method for flux linkage of synchronous reluctance motor | |
| CN111371362B (en) | Compensation method for rotor position estimation of permanent magnet linear motor by high-frequency injection method | |
| CN108347207B (en) | Position and speed estimation method of permanent magnet synchronous motor based on complex PI controller | |
| CN110995102A (en) | Direct torque control method and system for permanent magnet synchronous motor | |
| CN113328665B (en) | Synchronous reluctance motor position sensorless control method based on inductance identification | |
| CN109245648B (en) | Online compensation method for periodic error in output signal of rotary transformer | |
| CN112054730A (en) | Permanent magnet synchronous motor online parameter identification method | |
| CN115173774A (en) | Permanent magnet synchronous motor position sensorless control method and system | |
| CN109510539B (en) | Model prediction flux linkage control system and method based on gain matrix | |
| CN109067276B (en) | High-dynamic robust prediction current control method for permanent magnet synchronous motor | |
| Scalcon et al. | Sensorless control of a SynRM drive based on a Luenberger observer with an extended EMF model | |
| CN114785228B (en) | Permanent magnet synchronous motor inductance parameter online identification method based on virtual shafting injection | |
| Pengcheng et al. | Offline parameter identification strategy of permanent magnet synchronous motor considering the inverter nonlinearities | |
| CN111130409B (en) | Ultra-low speed high-precision positioning control method of frameless torque motor | |
| Shen et al. | Seamless transition strategy for wide speed-range sensorless IPMSM drives with a virtual q-axis inductance | |
| Morandin et al. | Locked rotor characterization tests of IPM/REL synchronous machine for sensorless drives | |
| Ding et al. | Research on sensorless control of permanent magnet synchronous motor based on adaptive sliding mode observer | |
| Wang et al. | Initial rotor position and inductance estimation of PMSMs utilizing zero-current-clamping effect |
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 |