Synchronous reluctance motor full-speed-domain speed sensorless control method combined with parameter identification

By combining the fundamental wave model and the convex pole model, the filterless high-frequency square wave injection method and the voltage-current model method are used, combined with online parameter identification and weighting calculation, the full-speed sensor control of the synchronous reluctance motor is realized, and the solution is to The problems of inaccurate estimation and high manufacturing cost in traditional methods improve the reliability and estimation accuracy of the motor.

CN118971708APending Publication Date: 2024-11-15CHINA UNIV OF MINING & TECH
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202411377529.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-30
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

The prior art is difficult to realize the full-speed sensorless control of synchronous magnetoresistive motors. The traditional method is inaccurate in the medium and high speed areas, and the manufacturing cost is high, and the sensor reliability is low.

Method used

Combining the fundamental wave model and convex pole model, the speed is estimated in the low-speed domain by filterless high-frequency square wave injection method, the voltage-current model method is estimated in the high-speed domain, and smooth transition is achieved in the transition zone through online parameter identification and weighting calculation. Avoid filter use and improve response time and system stability.

Benefits of technology

The synchronous magnetoresistive motor has been realized with no speed sensor control in the full-speed domain, which reduces manufacturing costs, improves the reliability and estimation accuracy of motor operation, is suitable for medium and high-speed areas, and reduces system jitter and estimation deviation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure QLYQS_1
    Figure QLYQS_1
  • Figure QLYQS_3
    Figure QLYQS_3
  • Figure QLYQS_5
    Figure QLYQS_5
Patent Text Reader

Abstract

The invention discloses a synchronous reluctance motor full-speed-domain speed-sensorless control method combined with parameter identification, and the method aims at the problem that the calculation precision of a speed-sensorless sensor of a high-speed-domain model method is reduced due to the parameter change during the operation of a synchronous reluctance motor, and employs an online parameter identification method to update inductance parameters in real time. The calculation precision of the speed sensorless model method is improved; a speed sensorless control scheme of a non-filtering high-frequency square wave injection method which does not depend on motor parameters in a low-speed domain is combined, two control methods are fused in a transition region, smooth transition is realized by adjusting weights of the two control methods through system speed, and the method can realize rotation speed estimation of a synchronous reluctance motor in a full-speed domain. The speed sensorless estimation precision of the high-speed domain model method is improved through online parameter identification, full-speed domain speed sensorless control of the synchronous reluctance motor is achieved in combination with the filtering-free high-frequency square wave injection method, and the operation efficiency of speed sensorless control of the synchronous reluctance motor is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of synchronous reluctance motor rotor position and parameter estimation, and in particular to a full-speed range speed sensorless control method for a synchronous reluctance motor combined with parameter identification. Background Art

[0002] In order to cope with the fluctuations in rare earth prices and the shortcomings of traditional motors in recent years, synchronous reluctance motors have been widely used in many fields due to their advantages such as high power density, good high temperature resistance and low cost. Although synchronous reluctance motors are also synchronous motors in essence, the principle of their torque generation is quite different from that of traditional synchronous motors.

[0003] In high-performance synchronous reluctance motor control, accurate rotor position information needs to be obtained. The common method is to introduce a rotor position sensor, but this will increase the manufacturing cost and volume of the motor and reduce the reliability of the motor operation due to the presence of the sensor. Therefore, in order to improve the reliability of the motor operation and reduce the manufacturing cost, it is particularly urgent to realize the speed sensorless control of the synchronous reluctance motor.

[0004] The position estimation methods of synchronous reluctance motors can be roughly divided into two categories: fundamental wave models that use the fundamental wave equations of voltage and current when the motor is working to estimate the rotor position and speed information, and salient pole models that use the salient pole effect of the synchronous motor's rotor to estimate the rotor position and speed information. Among them, fundamental wave models such as flux observation method, model reference adaptive method, extended electromotive force method and sliding film observer method are mainly suitable for medium and high speed operation ranges; salient pole models such as rotating high frequency injection method and pulsating high frequency injection method are mainly suitable for low speed operation ranges. Therefore, it is difficult to achieve full-speed sensorless control of synchronous reluctance motors through a single method, so a composite observer is designed and used to solve this problem. Summary of the invention

[0005] The present invention aims to solve the full-speed range speed sensorless control scheme of the synchronous reluctance motor, combines two different estimation methods to estimate the rotational speed in the low speed range and the medium and high speed range respectively, and introduces a parameter identification method to solve the problem of inaccurate estimation caused by large changes in motor parameters at high speed.

[0006] The specific technical solutions are as follows:

[0007] A speed sensorless control method for a synchronous reluctance motor in full speed range combined with parameter identification comprises the following steps:

[0008] Step 1, before the motor starts running, a fixed voltage vector is applied to the motor to rotate the rotor to a preset initial position;

[0009] Step 2: In the low-speed domain, the rotation speed is estimated by using a high-frequency square wave injection method without filtering, the position relationship between the coordinate systems of the square wave injection method is defined, and the high-frequency signal model is derived; a high-frequency square wave voltage signal is injected into the estimated d-axis, and then a signal separation method without a filter is used to extract the high-frequency response signal. After the position error signal is decoupled, the rotor position is finally estimated by a position tracker;

[0010] Step 3: In the high-speed domain, the voltage-current model method is used to estimate the speed, the stator current and back electromotive force are combined to estimate the flux linkage, and the recursive least squares method and the least mean square algorithm are used to estimate the real-time parameters of the motor for use in the model method speed estimation to improve the estimation accuracy;

[0011] Step 4, combining the low speed and high speed methods in the transition zone, calculating the switching speed in the transition interval to determine the weighted coefficients of the two methods, and obtaining the estimated speed of the transition zone by the weighted calculation method to achieve a smooth transition between the two methods.

[0012] The present invention provides a solution for speed sensorless control of a synchronous reluctance motor in the full speed range, and introduces an online parameter identification solution in the medium and high speed range. Compared with the traditional control solution, the method optimizes the process of offline parameter measurement and tabulation of the motor in the front end, and is more universal. In the low speed range, a position and speed estimation method based on high-frequency square wave pulse injection is adopted, which avoids the extensive use of filters, improves the response time of the system, and reduces system jitter.

[0013] The present invention also provides an electronic device, including a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the full-speed range rotational speed and position estimation method and the full-parameter identification method as described above.

[0014] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the full-speed range rotation speed and position estimation method and the full-parameter identification method as described above. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required to be used in the embodiments are briefly introduced below.

[0016] Figure 1 is a structural block diagram of a composite observer combined with parameter identification in an embodiment of the present invention;

[0017] Figure 2 Flow chart of the online parameter identification scheme in this embodiment;

[0018] Figure 3It is the structural block diagram of the flux observer;

[0019] Figure 4 It is an orthogonal phase-locked loop structure;

[0020] Figure 5 This is the control block diagram of the synchronous reluctance motor without speed sensor in the full speed range;

[0021] Figure 6 The estimated speed and actual speed diagram obtained by the composite observer combined with parameter identification;

[0022] Figure 7 Estimated rotor position and actual position diagram obtained by the composite observer combined with parameter identification. DETAILED DESCRIPTION

[0023] The full-speed range speed sensorless control technical solution of the synchronous reluctance motor will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0024] The present invention combines two speed sensorless control strategies, including flux observer and high frequency injection method, and identifies the variable parameters of the motor and uses them in the flux observer. The main structural block diagram is as follows: Figure 1 shown.

[0025] The structure of the flux observer is as follows: Figure 3 As shown, specifically:

[0026] The voltage equation of the synchronous reluctance motor in the αβ coordinate system is:

[0027]

[0028] From formula (17), the magnetic flux in the αβ coordinate system can be deduced as:

[0029]

[0030] The flux linkage in the dq coordinate system can be obtained through the current model, that is:

[0031]

[0032] Since equation (18) is a pure integral link, the stator flux obtained in the αβ coordinate system will have DC bias error and initial phase error. αβ 、Current i αβ As an easily measurable input quantity, the motor current model is used as the basic model of the motor, the voltage model is used as the adjustable model, and the stator flux obtained by the two models is used as a correction link to adjust the voltage model. Figure 3The difference between the estimated flux and the actual flux is compensated to the voltage model through the gain G, so that the integral DC offset error can be compensated. The estimated flux can be expressed as:

[0033]

[0034] In order to obtain the rotor position information, an effective flux ψ is defined a =ψ s -L q i q , from which we can know:

[0035]

[0036] Then, through coordinate transformation, we transform from the dq coordinate system to the αβ coordinate system, and the equation is:

[0037]

[0038] Then we can get it by using the inverse tangent function:

[0039]

[0040] This model needs to go through two coordinate changes in the process of estimating the rotor position, and there is an error between the estimated angle and the actual angle, and this error will be fed back to the system through the current model, causing system stability problems. The rotor position angle information obtained by the inverse tangent calculation is unfiltered information, so a more accurate speed estimation algorithm is needed. Therefore, a phase-locked loop needs to be added to calculate the motor speed. The phase-locked loop structure is as follows: Figure 4 shown.

[0041] The objects of online parameter identification include stator resistance and dq axis inductance. The identification process is as follows: Figure 2 As shown, the specific process is as follows:

[0042] Step 1: Initialize the identification parameters, including R s =R s0 , L d =L d0 , L q =L q0 , λ=λ0,μ=μ0,P(k)=P(0), set the system execution time T c =T0, set the time mark to T flag , set the time limit bit to T lim ;

[0043] Step 2: Update the inductance parameter L by recursive least squares method such as equations (6)-(8) and (12)-(14). d , L q , and let Tflag =T flag +T c ;

[0044] Step 3: Determine T flag Is it greater than T lim , if it is less than T lim Return to step 2 to update the inductance L d , L q If it exceeds T lim Then perform the operation of step 4;

[0045] Step 4: Update the stator resistance R by using the least mean square method as shown in equations (9)-(11) and (15)-(17). s , and let T flag =T lim ;

[0046] Step 5: Determine whether the motor is in a steady state. If the motor is not in a steady state, return to step 4 to update the stator resistance R. s , if it is in steady state, let T flag =0, and return to step 2 to update the inductance L d , L q .

[0047] The specific process of using the high-frequency square wave signal injection method in the low-speed domain is to inject a high-frequency square wave voltage signal into the estimated d-axis, and then extract the high-frequency response signal through a filter-free signal separation method. After the position error signal is decoupled, the rotor position is finally estimated through a position tracker. At this time, the injected signal frequency is much higher than the fundamental operating frequency of the synchronous reluctance motor, so the high-frequency voltage model of the motor in the two-phase rotating coordinate system can be written as:

[0048]

[0049] Converting it to the two-phase stationary αβ coordinate system yields:

[0050]

[0051] Among them, the high-frequency voltage in the two-phase rotating coordinate system can be expressed as:

[0052]

[0053] The purpose of injecting a high-frequency square wave signal into the axis of the estimated two-phase rotating coordinate system is to reduce the torque pulsation caused by the high-frequency signal injection method. Therefore, the injected high-frequency signal can be expressed as:

[0054]

[0055] Therefore, by substituting equations (26) and (27) into equation (25), the current response of the high-frequency voltage in the two-phase stationary coordinate system can be obtained as follows:

[0056]

[0057] It can be seen from equation (28) that when the observer converges, that is, Δθ e =0, formula (28) can be rewritten as:

[0058]

[0059] Therefore, according to formula (29), it can be directly seen that the rotor position angle can use the inverse tangent function to obtain the rotor position information, but the direct use of the inverse tangent function is easily affected by noise signals, so the following is used: Figure 4 Extract the rotor position error signal Δθ e .

[0060] The key to obtaining the position error signal is to separate the q-axis high-frequency response current signal stimulated by the high-frequency square wave signal injected from the d-axis. The present invention adopts a filter-free signal separation strategy to achieve the separation of the high-frequency response signal. According to the form of the injected high-frequency voltage, it can be known that the current in two adjacent cycles can be expressed as a superposition of the fundamental component and the high-frequency component:

[0061]

[0062] Since the frequency of the high-frequency signal is much greater than the fundamental signal frequency, it can be assumed that the fundamental component remains unchanged in two adjacent sampling periods. Since the injected high-frequency component is a square wave signal and is much higher than the system operating frequency, the electronic resistance can be ignored. In adjacent sampling periods, the response current is equal in magnitude and opposite in sign, which can be expressed as:

[0063]

[0064] Substituting equation (31) into equation (30) can obtain the fundamental current and high-frequency response current, thereby realizing the separation of the current signal:

[0065]

[0066] The switching problem of the composite position observer adopts a weighted average solution. By setting the speed threshold, the speed is divided into three areas: low speed area, transition area and medium and high speed area. The high-frequency injection method is used to obtain the speed information at low speed. In the transition area, the weighted coefficient is used to make the two observers work at the same time. In the medium and high speed area, the flux observer is used to obtain the speed information, thereby realizing a smooth transition between the two observation methods. The formula of the weight factor is:

[0067]

[0068] Among them, ω1 and ω2 are the lower limit and upper limit of the speed transition zone respectively. In the low speed zone, the weight of the high-frequency signal injection method is 1, and the final estimated speed is completely determined by the high-frequency signal injection method; in the transition zone, the weight of the high-frequency signal injection method gradually decreases from 1 to 0, and the weight of the flux observer gradually increases from 0 to 1, thereby achieving smooth switching of the two position observers and preventing sudden changes in the estimated speed; in the high speed zone, the weight of the flux observer is 1, avoiding the influence of the high-frequency injection method on the estimated speed of the flux observer at high speed.

[0069] Based on the above technical solution, the following embodiments are provided.

[0070] The motor parameters are shown in Table 1. The bus voltage of the system is 700V and the switching frequency is 10kHz.

[0071] Table 1 Synchronous reluctance motor parameters

[0072]

[0073] In order to verify the feasibility of the full-speed range speed sensorless control method of the synchronous reluctance motor combined with parameter identification, the motor parameters in this example will change greatly with the operating state during operation. This example is intended to solve the problem of speed sensorless deviation caused by parameter changes. The complete system block diagram is shown in the figure. Figure 5 As shown in the figure, current and voltage are selected as easily measured input quantities, and after the operation of the above control algorithm, reasonable control output is obtained, thereby ensuring the accuracy of control.

[0074] Figure 6 and Figure 7 The speed comparison diagram and rotor position comparison diagram obtained by the composite observer combined with parameter identification are shown in the figure. It can be seen from the figure that the tracking delay is small when the improved estimation method is used, and the two estimation methods also achieve a smooth transition in the transition zone, without the system out of control caused by the large fluctuation of the estimation results.

[0075] It can be concluded that this method is feasible in the speed sensorless control of the synchronous reluctance motor, can cope with the problem of large parameter changes of the synchronous reluctance motor, and can ensure the accuracy of the output speed and rotor position information and has good ability to resist load disturbances.

[0076] The present invention also provides a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the speed sensorless control method for a synchronous reluctance motor as described above.

[0077] This article uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only used to help understand the core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A speed sensorless control method for a synchronous reluctance motor in full speed range combined with parameter identification, characterized in that: include: Step 1, before the motor starts running, a fixed voltage vector is applied to the motor to rotate the rotor to a preset initial position; Step 2: In the low-speed domain, the rotation speed is estimated by using a high-frequency square wave injection method without filtering, the position relationship between the coordinate systems of the square wave injection method is defined, and the high-frequency signal model is derived; a high-frequency square wave voltage signal is injected into the estimated d-axis, and then a signal separation method without a filter is used to extract the high-frequency response signal. After the position error signal is decoupled, the rotor position is finally estimated by a position tracker; Step 3: In the high-speed domain, the voltage-current model method is used to estimate the speed, the stator current and back electromotive force are combined to estimate the flux linkage, and the recursive least squares method and the least mean square algorithm are used to estimate the real-time parameters of the motor for use in the model method speed estimation to improve the estimation accuracy; Step 4, combining the low speed and high speed methods in the transition zone, calculating the switching speed in the transition interval to determine the weighted coefficients of the two methods, and obtaining the estimated speed of the transition zone by the weighted calculation method to achieve a smooth transition between the two methods.

2. A speed sensorless control method for a synchronous reluctance motor in full speed range combined with parameter identification according to claim 1, characterized in that The derivation of the low-speed domain high-frequency signal model in step 2 specifically includes: Since the injection signal frequency is much larger than the fundamental operating frequency, the influence of the stator resistance voltage drop and the back electromotive force can be ignored, and the synchronous reluctance motor can be equivalent to an inductive load, that is: In the formula, are the high-frequency voltage and current components under the d and q axes respectively; the subscript h represents the high-frequency component; the superscript r represents the rotor shaft system; L d , L q are the d-axis and q-axis inductances respectively; p is the differential operator. Transform equation (1) to the stationary αβ coordinate system and inject a symmetrical square wave voltage signal with a magnitude of one tenth of the DC bus current into the d-axis, and the high-frequency response current signal envelope can be obtained as: In the formula, I αh ,I βh They are respectively the high frequency response current envelope under the stationary α and β axes, is the high-frequency voltage component injected into the d-axis; ω h is the frequency of the injected square wave voltage signal; L avg =(L d +L q ) / 2 is the average inductance; L dif =(L d -L q ) / 2 is the inductance difference. When the position tracker converges, Δθ e =0, we can get: Will I αh ,I βh After extraction, the position and speed information can be obtained by inputting into the phase-locked loop.

3. A speed sensorless control method for a synchronous reluctance motor in full speed range combined with parameter identification according to claim 1, characterized in that In step 2, there is no filtering high-frequency signal separation solution. The specific process is as follows: The high frequency response current can be expressed as: In the formula, i dq (k), i dqs (k), i dqf (k) are the high-frequency response current vector, sampling current vector and fundamental current vector at sampling time k respectively. Therefore, filter-free signal separation can be achieved through simple algebraic operations, and we can get:

4. A speed sensorless control method for a synchronous reluctance motor in full speed range combined with parameter identification according to claim 1, characterized in that The specific design process of the voltage-current flux observer in step 3 is as follows: The motor voltage u αβ 、Current i αβ As the easily measurable input quantity, the current model of the motor is taken as the basic model of the motor, and the voltage model is taken as the dynamic link. Then the flux obtained by observing the two models is subtracted as the correction link. The voltage model is compensated by the gain G to compensate for the DC bias in the voltage model. That is, the estimated motor αβ flux is first obtained by using the voltage model of the motor, and then the motor dq axis flux is obtained by using the current model, and finally the rotor position of the motor is obtained.

5. A speed sensorless control method for a synchronous reluctance motor in full speed range combined with parameter identification according to claim 1, characterized in that The specific process of parameter identification in step 3 is: In order to improve the accuracy of online parameter identification, the stator resistance and inductance parameters of the motor are identified step by step. According to the characteristics that the inductance parameter changes quickly with the current, while the stator electronics changes slowly with the winding problem, the recursive least squares algorithm with fast convergence speed is used to quickly identify the inductance. Then, based on the identification results of the recursive least squares algorithm, the least mean square algorithm with less calculation is used to update the stator resistance. The iterative formula of the recursive least squares algorithm is: Among them, θ est (k) is the parameter matrix to be identified; K(k) is the gain matrix; Y(k) is the system output; is the intermediate quantity; P(k) is the covariance matrix, P(0)=νI, v=10000; λ=0.

99. The least mean square algorithm can be used to identify resistor parameters because of its good robustness and low computational burden. The formula is: ε(k)=d(k)-x T (k)w(k) (9) w(k)=w(k-1)+2με(k-1)x(k-1) (10) Where x is the input signal vector; d is the desired signal vector; ε is the estimation error; w and μ are the weight parameter and convergence factor, respectively. According to the recursive least squares iterative formula and the motor mathematical model, the inductance identification model can be obtained as follows: Based on the least mean square algorithm, the resistor identification model can be obtained as follows: x(k)=t s i d (k) (14) d(k)=t s u d (k)+t s ω e (k)i q (k)L q -(i d (k)-i d (k-1))L d (15) w(k)=R s (k) (16)。 6. A speed sensorless control method for a synchronous reluctance motor in full speed range combined with parameter identification according to claim 5, characterized in that The specific process of the full parameter identification scheme is as follows: First, the identification parameters are initialized, the inductance is identified by the recursive least squares algorithm and a flag bit T is set. flag , when the inductance identification program reaches the set time T lim The upper limit stops identification and updates the stator resistance through the least mean square algorithm. When the motor is running stably, the identification position T is maintained. flag =T lim , and stop updating the motor parameters; when the motor's operating status changes, return to the inductance identification program to update the inductance value and repeat the above operations.

7. An electronic device, characterized in that: The electronic device comprises a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the steps of the control method of the synchronous reluctance motor according to claim 1.

8. A computer-readable storage medium, characterized in that: The computer program is stored therein, and when the computer program is executed by a processor, the steps of the synchronous reluctance motor control method as claimed in claim 1 are implemented.

Citation Information

Cited By

  • Static electronic speed measurement test method for generator

    CN119322183A

  • Stationary Electronic Speed Measurement Test Method for Generators

    CN119322183B