Current prediction control method and computer equipment based on variable gain deadbeat motor

By using variable gain-free beat-free motor current prediction control method and Longberg observer without model parameters in permanent magnet synchronous motor, the problem of complex current loop parameter setting and insufficient system dynamic response capability is solved, and the motor current is realized 和力矩的精准控制和系统稳定性提升。

CN118300467BActive Publication Date: 2024-11-26ANHUI UNIV +1
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202410419840.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-04-09
Publication Date
2024-11-26
Estimated Expiration
2044-04-09

AI Technical Summary

Technical Problem

In traditional control frameworks, permanent magnet synchronous motors have problems such as complex current ring parameter setting, coupling function and dynamic response capabilities affected by interference, especially in the case of system parameter mismatch and disturbance, which affects control performance and stability.

Method used

Using the current prediction control method of the non-difference motor based on variable gain, the mathematical model of the permanent magnet synchronous motor is discretized, and a Longberg observer without model parameters is constructed for disturbance compensation, and the self-immune disturbance control algorithm is used to adapt. Adjust the gain matrix of the Longberg observer and select nonlinear functions to improve observation accuracy and system stability.

Benefits of technology

It realizes precise control of motor current and torque, enhances the robustness of the control system, reduces the complexity of parameter setting, and improves dynamic and steady-state performance, ensuring the stability of the system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118300467B_ABST
    Figure CN118300467B_ABST
Patent Text Reader

Abstract

The present invention discloses a method for predicting and controlling a current of a motor with no deadbeat based on variable gain, including: S1: establishing a mathematical model of the motor, obtaining motor current and voltage prediction equations through forward difference discretization, performing deadbeat current prediction control on the current loop, ensuring that the output current of the control system at the next moment keeps up with the given current, and taking the given current value of the control system as the output current value at the next moment; S2: constructing a Lumberg observer without model parameters based on parameter mismatch, compensating the deadbeat current prediction controller with the disturbance caused, and obtaining the control signal of the motor current loop; S3: selecting a nonlinear function for the gain matrix of the Lumberg observer based on the error value between the actual current and the observed current during the operation of the motor, and adaptively adjusting the value of the Lumberg observer gain according to the change of the error value. The present invention effectively improves the dynamic and steady-state performance of the system.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of motor motion control, and in particular to a variable gain deadbeat motor current prediction control method and computer equipment. Background Art

[0002] Permanent magnet synchronous motors have the advantages of high torque density, fast dynamic performance, and good reliability, and have been widely used in transportation, manufacturing equipment, household appliances and other industries. At present, vector control is increasingly used in permanent magnet synchronous motors. Its basic principle is to decouple torque and flux by transforming the stationary phase current into a rotating coordinate system, which has better steady-state performance than other control strategies.

[0003] The traditional control framework of permanent magnet synchronous motor generally uses field oriented control (FOC) technology to control the motor by converting three-phase current into direct-axis current and quadrature-axis current, and using two proportional integral (PI) controllers. However, the parameter setting of the controller is relatively complex and will also produce mutual coupling. In the prior art, it has been proposed to improve the above problems by using a deadbeat current prediction control algorithm. However, the control performance of the deadbeat current prediction control algorithm depends on the accuracy of the motor parameters used to establish the prediction model. In actual operation, due to factors such as temperature change, magnetic saturation, cross-coupling, such as internal unmodeled and external unknown interference, there are inevitable interference and motor parameter mismatch problems during the operation of the motor, which will affect the steady-state static error and dynamic rapid response capability of the control algorithm; in addition, when adjusting the observer gain, a fixed gain matrix is ​​generally selected, but with a series of disturbances generated during the operation of the system, the pole position of the observer will also change from time to time, affecting the dynamic and steady-state performance of the system, and seriously affecting the stability of the system. Summary of the invention

[0004] The present invention provides a variable gain deadbeat motor current prediction control method and computer equipment that are helpful to improve the dynamic and steady-state performance of the system and can solve at least one of the above technical problems.

[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0006] A method for predicting and controlling a current of a deadbeat motor based on variable gain comprises the following steps:

[0007] S1: Establish a mathematical model of the permanent magnet synchronous motor, and discretize the mathematical model through forward difference to obtain the current and voltage prediction equations of the motor. According to the prediction equation, perform deadbeat current prediction control on the current loop of the motor to ensure that the output current of the control system at the next moment keeps up with the given current, and use the given current value of the control system as the output current value at the next moment. Generate the required switching signal through space vector pulse width modulation and act on the inverter;

[0008] S2: Based on the parameter mismatch during the operation of the S1 motor, a Romberg observer without model parameters is constructed, and the disturbance caused by the observer is compensated for the deadbeat current prediction controller to obtain the control signal of the motor current loop;

[0009] S3: Based on the error value between the actual current and the observed current during the operation of the S2 motor, according to the anti-disturbance control algorithm, using the "large error, small gain, small error, large gain" adjustment method, a nonlinear function is selected for the gain matrix of the Lumberg observer, and the value of the Lumberg observer gain is adaptively adjusted according to the change of the error value.

[0010] Furthermore, the S1 further comprises:

[0011] S11: Establish the stator voltage equation of the permanent magnet synchronous motor in the rotating coordinate system:

[0012] (1)

[0013] (2)

[0014] in: and are the d-axis component and q-axis component of the stator voltage in the dq coordinate system respectively; and are the d-axis component and q-axis component of the stator current in the dq coordinate system respectively; R is the stator resistance; and are the stator inductances of the d-axis and q-axis respectively; is the angular velocity of the motor; is the magnetic flux of the rotor permanent magnet; for general label type permanent magnet synchronous motor, there are ;

[0015] S12: The current state equation is discretized using the forward Euler equation, and the sampling period is selected as T. The current prediction model of the permanent magnet synchronous motor is obtained as follows:

[0016] (3)

[0017] in:

[0018] , , (4)

[0019] in: and is the current component value of the dq axis at the k+1th moment;

[0020] S13: Set the current command value As the current prediction value of the next cycle, it is compared with the current sampling value of the motor under the current running state. Substitute them into equation (3), and the voltage at the kth moment is:

[0021] (5)

[0022] (6)

[0023] in: and are the voltage values ​​at the kth moment respectively; and They are the predicted current values ​​of the dq axis at time k+1 respectively.

[0024] Further, in S13, the control voltage at this time is:

[0025] (7)

[0026] Since the sampling time is short enough, the current prediction value at this moment is equal to the current prediction value at the next moment, that is, and Use and replace;

[0027] Therefore, the voltage control equation at the kth moment is:

[0028] (8)

[0029] (9).

[0030] Furthermore, the S2 further comprises:

[0031] S21: During the operation of the motor, the deadbeat current prediction controller is very sensitive to the changes in the motor parameters, and there are disturbances with the operation of the motor. At this time, the mathematical model of the motor is rewritten as:

[0032] (10)

[0033] (11)

[0034] in: , , Parameters for the normal working state of the motor; , L. is the value that changes when the motor parameters are mismatched;

[0035] S22: Establish a model-free Lumberg observer to reconstruct the state variables of the motor input and output, and compensate for the disturbance of the deadbeat current predictive controller in the case of motor parameter mismatch.

[0036] Furthermore, in S21, the disturbance in the case of system parameter mismatch is placed in a separate item, and the disturbance is set to:

[0037] - (12)

[0038] (13)

[0039] in: and are the disturbances caused by parameter changes on the dq axis respectively;

[0040] For the Motorless Model Parameters, select:

[0041] (14)

[0042] ( (15)

[0043] According to the above disturbance division, the motor equation without model parameters is obtained by bringing it into the mathematical model of the motor. At this time, the motor equation is:

[0044] (16)

[0045] (17).

[0046] Further, in S22, the state variable is selected as , the matrix expression of the mathematical model of the motor is:

[0047] (18)

[0048] (19)

[0049] in:

[0050] , ,

[0051] Set the state variables and observation variables of the Lumberg observer. In order to facilitate the calculation of the eigenvalues ​​of the Lumberg observer, the observer gain matrix is ​​selected as:

[0052] (20)

[0053] At this time, the state variable of the Romberg observer , at this time the state equation of the motor system is expressed as:

[0054] (twenty one)

[0055] (twenty two)

[0056] By discretizing the state variables of the Lumberg observer, the state equation in the discrete domain is obtained. Short enough, that is , , so the above items are omitted, and then we get:

[0057] (twenty three)

[0058] The stability discriminant is:

[0059] (twenty four)

[0060] According to the Julie stability criterion in the discrete domain, we can obtain:

[0061] (25)

[0062] Sure and The value range of is brought into the deadbeat current prediction control system to obtain the reference voltage vector, which is a voltage vector with fixed gain.

[0063] Furthermore, the S3 further includes:

[0064] S31: According to the adjustment mode of “large error and small gain, small error and large gain”, a suitable nonlinear function is selected to improve the observation accuracy of the Lumberg observer by changing the fitting degree of the error;

[0065] S32: The selected nonlinear function is introduced into the system equation of the discretized Lumberg observer, and the stability of the system equation is judged. At the same time, the complexity of parameter setting is reduced by changing the parameters of the nonlinear function.

[0066] Further, in S31, according to the fal function:

[0067] (26)

[0068] Where: e is the error between the actual value and the estimated value; is the error threshold, α is a constant;

[0069] The nonlinear function of the Romberg observer gain matrix is ​​chosen to be the inverse tangent function:

[0070] (27)

[0071] Where: e is the error between the actual value and the estimated value, α is a constant used to adjust the amplitude of the inverse tangent function, and the positive or negative value of α determines the stability of the system, β is a constant used to adjust the shape of the inverse tangent function;

[0072] The slope Fun' obtained by differentiating the function is configured as the gain of the Romberg observer:

[0073] (28)

[0074] When α<0, Fun'<0, the system is constant stable and satisfies the fixed gain K 2 The value requirements of

[0075] When e tends to be infinitely small, Fun' tends to be equal to α. By adjusting α and β, Fun' can be configured to be consistent with α, and then Fun' can be configured to be consistent with K 2 Consistent, completing the adaptive adjustment of variable gain.

[0076] Furthermore, in S32, the Fun function is used to replace K in the Lumberg observer gain matrix. 2 The gain of α and β is determined according to Julie's stability criterion and the range of error variation, and the range of α and β is brought into the deadbeat current prediction control system to obtain the voltage vector without model parameter compensation, which is the voltage vector with variable gain.

[0077] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the variable gain deadbeat motor current prediction control method.

[0078] The beneficial effects of the present invention are embodied in:

[0079] 1. The present invention establishes a mathematical model of a permanent magnet synchronous motor, discretizes the mathematical model using forward difference, obtains prediction equations for the motor's current and voltage, and implements deadbeat current prediction control, thereby solving the problems of complex current loop parameter setting, large measurement, and mutual coupling of permanent magnet synchronous motors in existing FOC technology.

[0080] 2. The present invention establishes a Luneberg observer without model parameters in the dq-axis coordinate system, classifies all external disturbances such as parameter changes into one category, compensates the voltage vector of the motor's zero-beat current predictive control, avoids mutual influence of parameters between the motor's current loop and speed loop, solves the parameter mismatch problem, realizes precise control of the motor's current and torque, and greatly enhances the robustness of the control system.

[0081] 3. Based on the anti-disturbance control algorithm, the present invention aims at the situation that the pole position of the fixed gain observer changes with the system disturbance. According to the error value between the actual current and the observed current of the dq axis, the nonlinear function Fun is selected to replace K in the original observer gain matrix by using the adjustment method of "large error and small gain, small error and large gain". 2 The gain of the Lumberg observer is adaptively adjusted to achieve variable gain control, which not only reduces the complexity of parameter setting, but also improves the observation accuracy of the observer, expands the range of gain values, and further ensures the dynamic and steady-state performance of the system, thus better maintaining the stability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] The drawings described herein are used to provide further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute improper limitations on the present application.

[0083] Figure 1 It is a block diagram of current prediction control based on variable gain and model-free parameters according to an embodiment of the present invention.

[0084] Figure 2 It is a schematic diagram of an embodiment of the present invention based on different nonlinear function selections.

[0085] Figure 3 3 is a schematic diagram comparing the d-axis current of the adaptive gain and the fixed gain according to an embodiment of the present invention.

[0086] Figure 4 1 is a schematic diagram comparing the q-axis current of the adaptive gain and the fixed gain according to an embodiment of the present invention.

[0087] Figure 5 It is a schematic diagram of the error between the actual value and the observed value of the adaptive gain d-axis current in an embodiment of the present invention.

[0088] Figure 6 It is a schematic diagram of the error between the actual value and the observed value of the fixed gain d-axis current.

[0089] Figure 7 It is a schematic diagram of the error between the actual value and the observed value of the adaptive gain q-axis current in an embodiment of the present invention.

[0090] Figure 8 It is a schematic diagram of the error between the actual value and the observed value of the fixed gain q-axis current.

[0091] Fig. 9 Schematic diagram of three-phase current with adaptive gain inductance mismatch according to an embodiment of the present invention.

[0092] Fig.10 It is a structural block diagram of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0093] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. In the absence of conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. It should be noted that a person of ordinary skill in the art can understand that all or part of the steps in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a program. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes. In addition, the "permanent magnet synchronous motor" in the present application can be replaced by "motor".

[0094] See also Figure 1 The embodiment of the present invention provides a method for predicting and controlling a current of a motor without a beat based on variable gain, comprising the following steps:

[0095] S1: Establish a mathematical model of the permanent magnet synchronous motor, and discretize the mathematical model through forward difference to obtain the current and voltage prediction equations of the motor. According to the prediction equation, perform zero-beat current prediction control on the current loop of the motor to ensure that the output current of the control system at the next moment keeps up with the given current, and use the given current value of the control system as the output current value at the next moment. Through space vector pulse width modulation, generate the required switching signal and act on the inverter.

[0096] In this embodiment, the S1 further includes:

[0097] S11: Establish the stator voltage equation of the permanent magnet synchronous motor in the rotating coordinate system:

[0098] (1)

[0099] (2)

[0100] in: and are the d-axis component and q-axis component of the stator voltage in the dq coordinate system respectively; and are the d-axis component and q-axis component of the stator current in the dq coordinate system respectively; R is the stator resistance; and are the stator inductances of the d-axis and q-axis respectively; is the angular velocity of the motor; is the magnetic flux of the rotor permanent magnet; for general label type permanent magnet synchronous motor, there are ;

[0101] S12: The current state equation is discretized using the forward Euler equation, and the sampling period is selected as T. The current prediction model of the permanent magnet synchronous motor is obtained as follows:

[0102] (3)

[0103] in:

[0104] , , (4)

[0105] in: and is the current component value of the dq axis at the k+1th moment;

[0106] S13: Set the current command value As the current prediction value of the next cycle, it is compared with the current sampling value of the motor under the current running state. Substitute them into equation (3), and the voltage at the kth moment is:

[0107] (5)

[0108] (6)

[0109] in: and are the voltage values ​​at the kth moment respectively; and They are the predicted current values ​​of the dq axis at time k+1 respectively.

[0110] In S13, the control voltage at this time is:

[0111] (7)

[0112] Under normal circumstances, since the sampling time is short enough, the current prediction value at this moment is equal to the current prediction value at the next moment, that is, and Can be used respectively and replace;

[0113] Therefore, the voltage control equation at the kth moment is:

[0114] (8)

[0115] (9)

[0116] S2: Based on the parameter mismatch during motor operation, a model-free parameter-free Lumberg observer is constructed, and the disturbance caused by the observer is used to compensate the deadbeat current prediction controller to obtain the control signal of the motor current loop.

[0117] In this embodiment, the S2 further includes:

[0118] S21: During the operation of the motor, the deadbeat current prediction controller is very sensitive to the changes in the motor parameters, and there are disturbances with the operation of the motor. At this time, the mathematical model of the motor is rewritten as:

[0119] (10)

[0120] (11)

[0121] in: , , Parameters for the normal working state of the motor; , L. is the value that changes when the motor parameters are mismatched;

[0122] In S21, the disturbance in the case of system parameter mismatch is placed in a separate item, and the disturbance is set to:

[0123] - (12)

[0124] (13)

[0125] in: and are the disturbances caused by parameter changes on the dq axis respectively;

[0126] For the Motorless Model Parameters, select:

[0127] (14)

[0128] ( (15)

[0129] According to the above disturbance division, by bringing it into the mathematical model of the motor, we can get the motor equation without model parameters. At this time, the motor equation is:

[0130] (16)

[0131] (17)

[0132] S22: Establish a model-free Lumberg observer to reconstruct the state variables of the motor input and output, and compensate for the disturbance of the deadbeat current predictive controller in the case of motor parameter mismatch;

[0133] In S22, the state variable is selected as , the matrix expression of the mathematical model of the motor is:

[0134] (18)

[0135] (19)

[0136] in:

[0137] , ,

[0138] Set the state variables and observation variables of the Lumberg observer. In order to facilitate the calculation of the eigenvalues ​​of the Lumberg observer, the observer gain matrix is ​​selected as:

[0139] (20)

[0140] At this time, the state variable of the Romberg observer , at this time the state equation of the motor system can be expressed as:

[0141] (twenty one)

[0142] (twenty two)

[0143] By discretizing the state variables of the Lumberg observer, the state equation in the discrete domain is obtained. Under normal circumstances, due to the sampling time Short enough, that is , , so the above items can be omitted, and then we get:

[0144] (twenty three)

[0145] The stability discriminant is:

[0146] (twenty four)

[0147] According to the Julie stability criterion in the discrete domain, we can obtain:

[0148] (25)

[0149] Sure and The value range of is brought into the deadbeat current prediction control system to obtain the reference voltage vector, which is a voltage vector with fixed gain.

[0150] S3: Based on the error value between the actual current and the observed current during the operation of the motor, according to the anti-disturbance control algorithm, using the "large error, small gain, small error, large gain" adjustment method, select a nonlinear function for the gain matrix of the Lumberg observer, and adaptively adjust the value of the Lumberg observer gain according to the change of the error value.

[0151] In this embodiment, S3 further includes:

[0152] S31: According to the adjustment mode of “large error and small gain, small error and large gain”, a suitable nonlinear function is selected to improve the observation accuracy of the Lumberg observer by changing the fitting degree of the error;

[0153] In S31, according to the fal function:

[0154] (26)

[0155] Where: e is the error between the actual value and the estimated value; is the error threshold, α is a constant;

[0156] The nonlinear function of the Romberg observer gain matrix is ​​chosen to be the inverse tangent function:

[0157] (27)

[0158] Where: e is the error between the actual value and the estimated value, α is a constant used to adjust the amplitude of the inverse tangent function, and the positive or negative value of α determines the stability of the system, β is a constant used to adjust the shape of the inverse tangent function;

[0159] The slope Fun' obtained by differentiating the function is configured as the gain of the Romberg observer:

[0160] (28)

[0161] When α<0, Fun'<0, the system is constant stable and satisfies the fixed gain K 2 The value requirements of

[0162] When e tends to be infinitely small, Fun' tends to be equal to α. By adjusting α and β, Fun' can be configured to be consistent with α, and then Fun' can be configured to be consistent with K 2 Consistent, completing the adaptive adjustment of variable gain;

[0163] It can be seen that the positive and negative values ​​of α can determine the stability of the system, and the value of β can not only reduce the number of parameter settings, but also simplify the difficulty of setting;

[0164] S32: Bring the selected nonlinear function into the system equation of the discretized Lumberg observer, and judge the stability of the system equation. At the same time, by changing the parameters of the nonlinear function, the complexity of parameter setting is reduced;

[0165] In S32, the Fun function is used to replace K in the Lumberg observer gain matrix. 2 The gain of α and β is determined according to Julie's stability criterion and the range of error variation, and the range of α and β is brought into the deadbeat current prediction control system to obtain the voltage vector without model parameter compensation, which is the voltage vector with variable gain.

[0166] See also Figure 2 For the above embodiment, when the gain of the Lumberg observer is adjusted by the fal function, the eigenvalue of the observer system equation is difficult to obtain. Therefore, the present application selects the nonlinear function Fun. By adjusting the parameter values ​​of the Fun function, the function fitting curves of fun1 and fun2 are simulated. It can be seen that when the error approaches 0, the slopes of fun1 and fun2 are greater, that is, the gain is greater. Conversely, when the error is greater, the slopes of fun1 and fun2 are flatter, that is, the gain is smaller, which meets the adjustment method of "large error and small gain, small error and large gain" proposed by Professor Han Jingqing.

[0167] The present invention selects deadbeat current predictive control, establishes a Lumberg observer without model parameters, and compensates for the disturbance of the deadbeat controller by calculating the change of the disturbance in view of the mismatch problem of the motor current loop and the speed loop parameters. At the same time, by selecting a nonlinear function Fun for the gain matrix of the Lumberg observer, the gain of the observer is adaptively adjusted according to the error value between the actual current of the dq axis and the observed current, so as to realize variable gain control, improve the observation accuracy of the observer, expand the value range of the gain, solve the problems of complex parameter setting and poor dynamic and steady-state performance of the system during the operation of the motor, and help to further improve the accuracy, stability and robustness of the system control.

[0168] The system parameters of the present invention are shown in Table 1. The motor starts without load, accelerates during 0-0.05s, and suddenly adds load at 0.3s.

[0169]

[0170] Table 1 System parameters

[0171] See also Figure 2-9 By comparison, it can be seen that the deadbeat current control without model parameters shows better steady-state and dynamic performance when the parameters are mismatched. At the same time, by comparing the gain matrix of the fixed gain and the variable gain Lumberg observer, the variable gain can reflect the accuracy of the observer, and see Fig. 9 It can be seen that in the case of parameter mismatch, the system can still maintain good stability.

[0172] It should be noted that in the present invention, in the accompanying drawings, L=2.5L 0 ; in the attached Figure 2 In the example, id is deadbeat, id* has no model parameters; Figure 4 In the equation, iq is deadbeat and iq* has no model parameters.

[0173] See also Fig.10 The embodiment of the present invention further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the variable gain deadbeat motor current prediction control method. The steps of the variable gain deadbeat motor current prediction control method here can be the steps of the variable gain deadbeat motor current prediction control method in the above-mentioned embodiments.

[0174] In summary, the present invention first establishes a mathematical model of a permanent magnet synchronous motor, discretizes the mathematical model using forward difference, obtains prediction equations for the current and voltage of the motor, realizes deadbeat current prediction control, and solves the problems of complex parameter setting, large measurement, and mutual coupling of the permanent magnet synchronous motor current loop in the existing FOC technology;

[0175] At the same time, by establishing a model-free parameter Lumberg observer in the dq axis coordinate system, all external disturbances such as parameter changes are classified into one category, and the voltage vector of the deadbeat current predictive control of the motor is compensated to avoid the mutual influence of parameters between the current loop and the speed loop of the motor, solve the parameter mismatch problem, realize the precise control of the current and torque of the motor, and greatly enhance the robustness of the control system;

[0176] At the same time, based on the anti-disturbance control algorithm, in view of the fact that the pole position of the fixed gain observer changes with the system disturbance, according to the error value between the actual current and the observed current of the dq axis, the adjustment method of "large error and small gain, small error and large gain" is used to select the nonlinear function Fun to replace K in the original observer gain matrix. 2The gain of the Lumberg observer is adaptively adjusted to achieve variable gain control, which not only reduces the complexity of parameter setting, but also improves the observation accuracy of the observer, expands the range of gain values, and further ensures the dynamic and steady-state performance of the system, thus better maintaining the stability of the system.

[0177] It should be understood that the examples and implementation modes described herein are for illustrative purposes only and are not intended to limit the present invention. Those skilled in the art may make various modifications or changes based on the examples and implementation modes. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for predictive control of motor current based on variable gain deadbeat, characterized in that: The following steps are involved: S1: Establish a mathematical model of the permanent magnet synchronous motor, and discretize the mathematical model through forward difference to obtain the current and voltage prediction equations of the motor. According to the prediction equation, perform deadbeat current prediction control on the current loop of the motor to ensure that the output current of the control system at the next moment keeps up with the given current, and use the given current value of the control system as the output current value at the next moment. Generate the required switching signal through space vector pulse width modulation and act on the inverter; S2: Based on the parameter mismatch during the operation of the S1 motor, a Romberg observer without model parameters is constructed, and the disturbance caused by the observer is compensated for the deadbeat current prediction controller to obtain the control signal of the motor current loop; S3: Based on the error value between the actual current and the observed current during the operation of the S2 motor, according to the active disturbance rejection control algorithm, a nonlinear function is selected for the gain matrix of the Lumberg observer using the adjustment method of "large error, small gain, small error, large gain", and the value of the Lumberg observer gain is adaptively adjusted according to the change of the error value; The S3 further comprises: S31: According to the adjustment mode of "large error and small gain, small error and large gain", a suitable nonlinear function is selected to improve the observation accuracy of the Lumberg observer by changing the error fitting degree; S32: Bring the selected nonlinear function into the system equation of the discretized Lumberg observer, and judge the stability of the system equation. At the same time, by changing the parameters of the nonlinear function, the complexity of parameter setting is reduced; In S31, according to the Fal function: (26) Where: e is the error between the actual value and the estimated value; is the error threshold, α is a constant; The nonlinear function of the Romberg observer gain matrix is ​​chosen to be the inverse tangent function: (27) Where: e is the error between the actual value and the estimated value, α is a constant used to adjust the amplitude of the inverse tangent function, and the positive or negative value of α determines the stability of the system, β is a constant used to adjust the shape of the inverse tangent function; The slope Fun' obtained by differentiating the function is configured as the gain of the Romberg observer: (28) When α<0, Fun'<0, the system is constantly stable and meets the value requirement of fixed gain K2; When e tends to be infinitely small, Fun' tends to be equal to α. By adjusting α and β, Fun' can be configured to be consistent with α, and then Fun' can be configured to be consistent with K2, completing the adaptive adjustment of variable gain. In S32, Fun' is used to replace K2 in the Lumberg observer gain matrix, and the value ranges of α and β are determined according to the Julie stability criterion and the error variation range, and the value ranges of α and β are brought into the zero-beat current prediction control system, so that the voltage vector without model parameter compensation can be obtained, which is the voltage vector with variable gain.

2. The variable gain deadbeat motor current prediction control method according to claim 1, characterized in that: The S1 further comprises: S11: Establish the stator voltage equation of the permanent magnet synchronous motor in the rotating coordinate system: (1) (2) in: and are the d-axis component and q-axis component of the stator voltage in the dq coordinate system respectively; and are the d-axis component and q-axis component of the stator current in the dq coordinate system respectively; R is the stator resistance; and are the stator inductances of the d-axis and q-axis respectively; is the angular velocity of the motor; is the magnetic flux of the rotor permanent magnet; for general label-type permanent magnet synchronous motors, there are ; S12: The current state equation is discretized using the forward Euler equation, and the sampling period is selected as T S , the current prediction model of permanent magnet synchronous motor is obtained as: (3) in: , , (4) in: and is the current component value of the dq axis at the k+1th moment; S13: Set the current command value As the current prediction value of the next cycle, it is compared with the current sampling value of the motor under the current running state. Substitute them into equation (3), and the voltage at the kth moment is: (5) (6) in: and are the voltage values ​​at the kth moment respectively; and They are the predicted current values ​​of the dq axis at time k+1 respectively.

3. The variable gain deadbeat motor current prediction control method according to claim 2, characterized in that: In S13, the control voltage at this time is: (7) Since the sampling time is short enough, the current prediction value at this moment is equal to the current prediction value at the next moment, that is, and Use and replace; Therefore, the voltage control equation at the kth moment is: (8) (9)。 4. The variable gain deadbeat motor current prediction control method according to claim 1, characterized in that: The S2 further comprises: S21: During the operation of the motor, the deadbeat current prediction controller is very sensitive to the changes in the motor parameters, and there are disturbances with the operation of the motor. At this time, the mathematical model of the motor is rewritten as: (10) (11) in: , , Parameters for the normal working state of the motor; , L. is the value that changes when the motor parameters are mismatched; S22: Establish a model-free Lumberg observer to reconstruct the state variables of the motor input and output, and compensate for the disturbance of the deadbeat current predictive controller in the case of motor parameter mismatch.

5. The variable gain deadbeat motor current prediction control method according to claim 4, characterized in that: In S21, the disturbance in the case of system parameter mismatch is placed in a separate item, and the disturbance is set to: (12) (13) in: and are the disturbances caused by parameter changes on the dq axis respectively; For the Motorless Model Parameters, select: (14) ( (15) According to the above disturbance division, the motor equation without model parameters is obtained by bringing it into the mathematical model of the motor. At this time, the motor equation is: (16) (17)。 6. The variable gain deadbeat motor current prediction control method according to claim 5, characterized in that: In S22, the state variable is selected as , the matrix expression of the mathematical model of the motor is: (18) (19) in: , , Set the state variables and observation variables of the Lumberg observer. In order to facilitate the calculation of the eigenvalues ​​of the Lumberg observer, the observer gain matrix is ​​selected as: (20) At this time, the state variable of the Romberg observer , at this time the state equation of the motor system is expressed as: (21) (22) By discretizing the state variables of the Lumberg observer, the state equation in the discrete domain is obtained. Short enough, that is , , so the relevant terms are omitted, and then we get: (23) The stability discriminant is: (24) According to the Julie stability criterion in the discrete domain, we can obtain: (25) Sure and The value range of is brought into the deadbeat current prediction control system to obtain the reference voltage vector, which is a voltage vector with fixed gain.

7. A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method according to any one of claims 1 to 6.

Citation Information

Cited By

  • Servo motor disturbance suppression method based on variable gain control

    CN121356393A