Dead-beat predictive control method

Abstract

The invention relates to the technical field of motor control, and particularly provides a dead-beat prediction control method, which comprises the following steps of: establishing a voltage equation of a permanent magnet synchronous motor; based on a voltage equation and current loop disturbance of the permanent magnet synchronous motor, establishing a dead-beat prediction control model; establishing a generalized self-adaptive super-spiral observer according to a self-adaptive theory and a super-spiral technology; obtaining a current loop controller through a dead-beat predictive control model and a generalized self-adaptive super-spiral observer; establishing a quasi-resonance controller by adopting a condition switching resonance mode, and enabling the quasi-resonance controller and the dead-beat prediction control model to jointly act on the current loop controller to obtain a final current loop controller; in combination with the generalized self-adaptive super-spiral observer and the improved quasi-resonance controller, satisfactory disturbance estimation precision and convergence speed are provided, and periodic disturbance can be effectively suppressed, so that the dynamic performance and the steady-state performance of the permanent magnet synchronous motor system are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of motor control, and particularly provides a deadbeat predictive control method based on a generalized adaptive super-twisting observer and improved quasi-resonant control. Background Art

[0002] In the field of electric drive, the demand for high-precision and high-dynamic current control of permanent magnet synchronous motors is increasing day by day, especially in high-performance servo systems such as numerical control machine tools and robot servo control. However, there are various disturbances in the actual permanent magnet synchronous motor system. Aperiodic disturbances such as cross-coupling disturbances, parameter perturbations, and external disturbances in the current loop reduce the stability of the system, and periodic disturbances such as flux harmonics and voltage harmonics reduce the control accuracy of the system. At present, for the aperiodic disturbances in permanent magnet synchronous motors, widely used control strategies include active disturbance rejection control, sliding mode control, and predictive control, etc.; for periodic disturbances, methods such as resonant controllers, repetitive controllers, and iterative learning are mostly used.

[0003] Among these methods, deadbeat predictive control has become a potential solution for high-performance current control due to its high dynamic response and compatibility with digital control systems. However, the performance of this algorithm depends to a large extent on the accuracy of the parameters used in the prediction model. Factors such as temperature changes, magnetic saturation, cross-coupling, and unmodeled disturbances may cause inevitable disturbances and parameter mismatches during operation, resulting in a sharp decline in the control performance of the algorithm.

[0004] Therefore, in order to reduce the decline in algorithm performance caused by parameter mismatch and compensate for disturbances such as cross-coupling disturbances and external disturbances in the current loop, an observer with high precision and high dynamic characteristics needs to be designed. For a linear extended state observer, this makes it have a high-gain form. However, the high-gain extended state observer is very sensitive to measurement noise, and the measurement noise will be amplified by the high-gain extended state observer and then applied to the system, which will cause the system to vibrate or even become unstable. Therefore, the fast and accurate estimation of disturbances and the suppression of measurement noise are an inevitable contradiction for the extended state observer.

[0005] For periodic disturbances, the quasi-resonant controller not only solves the performance decline caused by resonant frequency offset, but also is simple to implement and convenient for parameter tuning. However, in the dynamic process of the motor, the resonant term may have side effects. At steady state, the resonant frequency is consistent with the disturbance frequency, thus effectively suppressing the disturbance; when the motor is in a dynamic condition, due to the significant deviation between the motor speed and the target value, this suppression ability will be lost. Worse still, the control input signal in the motor dynamic contains a large amount of AC components, which causes the output of the resonant term to be amplified sharply, thus weakening the overall control performance.

[0006] Therefore, to solve the above problems, the present application proposes a deadbeat predictive control method. Summary of the Invention

[0007] The present invention provides a deadbeat predictive control method for the non-periodic and periodic disturbances existing in the current loop of a permanent magnet synchronous motor, realizing high-performance current control of the permanent magnet synchronous motor.

[0008] To achieve the above object, the technical solution of the present invention is realized as follows: A deadbeat predictive control method, comprising: Establishing a voltage equation of a permanent magnet synchronous motor; Based on the voltage equation of the permanent magnet synchronous motor and current loop disturbances, establishing a deadbeat predictive control model; According to the adaptive theory and the super-twisting technique, establishing a generalized adaptive super-twisting observer; Obtaining a current loop controller through the deadbeat predictive control model and the generalized adaptive super-twisting observer; Adopting a conditional switching resonance mode to establish a quasi-resonant controller, and jointly applying the quasi-resonant controller and the deadbeat predictive control model to the current loop controller to obtain a final current loop controller.

[0009] Furthermore, the expressions of the non-periodic disturbances in the current loop disturbances are respectively: , , where and are respectively axis and axis non-periodic disturbances; and are respectively axis and axis cross-coupling disturbances; and are respectively axis and axis known disturbances; and are respectively axis and axis disturbances caused by parameter perturbations; and are respectively axis and axis external uncertain disturbances; and are respectively axis and axis stator currents; is Axis and The nominal value of the axis stator inductance; is the nominal value of the stator resistance, is the nominal value of the rotor flux; , , ; The expressions of the periodic disturbances in the current loop are respectively: , , where, and are respectively Axis and the flux harmonics of the axis, and are respectively Axis and the amplitudes of the 6n-th flux harmonics of the axis; is the electrical angular velocity; and are respectively Axis and the voltage harmonics of the axis; and are respectively the dead time and the sampling period; is the DC bus voltage; Then the periodic disturbance of the permanent magnet synchronous motor Axis is expressed as: , where, and are respectively Axis and the periodic disturbances of the axis; and are respectively Axis and the stator inductances of the axis; The first-order super-twisting mathematical model is expressed as: , where, and are respectively the output and input variables of the model; is the set of known and unknown disturbances; is a scaling factor; Then the first-order super-twisting mathematical model in the current loop is finally expressed as: , where, and are respectively the total disturbances of the d-axis and the q-axis; It is for controlling the gain.

[0010] Further, in the step of establishing the deadbeat predictive control model based on the voltage equation and current loop disturbance of the permanent magnet synchronous motor, the following specific steps are included: The voltage equation of the permanent magnet synchronous motor is expressed as: , , Among them, and are respectively axis and axis at time; the reference voltage. and are respectively axis and axis at time; the reference current value. and are respectively axis and axis at time; the stator current. Stator inductance; is the sampling period; is the stator resistance; is time; the electrical angular velocity, is the rotor flux; According to the voltage equation and current loop disturbance of the permanent magnet synchronous motor, a deadbeat predictive control model of the permanent magnet synchronous motor based on the super-local model is obtained, expressed as: , Among them, and are respectively the total disturbances of the d-axis and q-axis; is the control gain.

[0011] Further, in the step of establishing the generalized adaptive super-twisting observer according to the adaptive theory and super-twisting technology, the following specific steps are included: An extended state observation model of the d-q axis current equation of the permanent magnet synchronous motor is established as: , Among them, , , ; ; , is the control gain; ; ; ; is the first derivative of the total disturbance of the shaft; Based on the extended state observation model, as well as the adaptive theory and the super-twisting technique, the generalized adaptive super-twisting observer is: , The error correction term is designed as: , where is the estimated value of the state vector; is the time; is the sampling period; is the control gain; is the stator voltage at time ; and are the gain values of the observer; is the variable power exponent term, and ; is the adaptive gain term; and are the error correction terms.

[0012] Furthermore, in the step of obtaining the current loop controller through the deadbeat predictive control model and the generalized adaptive super-twisting observer, the control law of the current loop controller is: .

[0013] Furthermore, in the step of establishing a quasi-resonant controller by adopting a conditional switching resonant mode and jointly applying the quasi-resonant controller and the deadbeat predictive control model to the current loop controller to obtain the final current loop controller, the following steps are specifically included: The transfer function of the quasi-resonant controller is: , where is the step function that switches at ; is the adjustable switching point; is the current tracking error, is the reference current, is the stator current; is the resonant gain; is the damping frequency; is the Laplace variable; is the resonant frequency; Discretize the quasi-resonant controller, and the discretized transfer function is: , Among them, is the complex frequency domain of the discrete-time signal; .

[0014] Furthermore, the discretized transfer function is added to the current loop controller, and combined with the deadbeat predictive control model, the final current loop controller is expressed as: .

[0015] Furthermore, the generalized adaptive super-twisting observer is used to estimate and compensate the aperiodic disturbance in the current loop disturbance; The quasi-resonant controller and the deadbeat predictive control model are implemented in parallel to estimate and compensate the periodic disturbance in the current loop disturbance.

[0016] Compared with the prior art, the present invention can achieve the following beneficial effects: (1) The present application proposes a deadbeat predictive control method for the problems of aperiodic and periodic disturbances existing in the current loop of a permanent magnet synchronous motor. By combining a generalized adaptive super-twisting observer and an improved quasi-resonant controller, it realizes fast and accurate estimation and compensation of the disturbances in the current loop; compared with the traditional deadbeat predictive control method, this method has significantly improved disturbance estimation accuracy and convergence speed, thus effectively improving the dynamic performance and steady-state performance of the permanent magnet synchronous motor system; (2) The generalized adaptive super-twisting observer is adopted. Based on the adaptive theory and super-twisting technology, a new nonlinear function is constructed to replace the linear term in the traditional extended state observer; through the adaptive gain, further according to the error dynamic feedback intensity, the observer can more quickly and accurately estimate and compensate the aperiodic disturbance existing in the current loop within a finite time; (3) By designing an improved quasi-resonant controller to be implemented in parallel with the deadbeat controller, it effectively estimates and compensates the periodic disturbance existing in the current loop, and avoids the side effects of the traditional quasi-resonant controller during the dynamic process, improving the dynamic performance and steady-state performance of the permanent magnet synchronous motor system. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings constituting a part of the present invention are used to provide a further understanding of the present invention. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation to the present invention. In the drawings: Figure 1 is a schematic flow chart of a deadbeat predictive control method according to an embodiment of the present invention; Figure 2 is a control structure diagram of a deadbeat predictive control method according to an embodiment of the present invention; Figure 3 It is the model structure diagram of the permanent magnet synchronous motor servo control system described in the embodiments of the present invention; Figure 4 It is a schematic diagram of the experimental results of the comparison of the d-axis and a-phase current curves; Figure 5 It is a schematic diagram of the experimental results of the comparison of the q-axis current curves; Figure 6 It is a schematic diagram of the experimental results of the comparison of the Fourier analysis of the q-axis current; Figure 7 It is for the inductance When there is a mismatch axis and schematic diagram of the experimental results of the comparison of the phase current; Figure 8 It is for the inductance When there is a mismatch Experimental results of the axis current comparison. Specific implementation manners

[0018] In order to make the purpose, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and do not constitute a limitation to the present invention. Similar elements in different embodiments are labeled with related similar element numbers. In the following embodiments, many detailed descriptions are for the better understanding of the present invention. However, those skilled in the art can easily recognize that some of the features can be omitted in different situations, or can be replaced by other elements, materials, methods. In some cases, some operations related to the present invention are not shown or described in the specification, which is to avoid the core part of the present invention being overwhelmed by excessive description. For those skilled in the art, it is not necessary to describe these related operations in detail, and they can fully understand the related operations according to the description in the specification and the general technical knowledge in the art.

[0019] It should be noted that, without conflict, the embodiments and features in the embodiments of the present invention can be combined with each other to form various implementation manners. At the same time, the steps or actions in the method description can also be adjusted in the order that can be obviously seen by those skilled in the art. Therefore, the various orders in the specification and the drawings are only for clearly describing a certain embodiment, and do not mean that they are the necessary orders, unless it is stated that a certain order must be followed.

[0020] The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

[0021] Please refer to Figure 1, is a schematic flow diagram of a deadbeat predictive control method described in this embodiment; aiming at the non-periodic and periodic disturbances existing in the current loop of a permanent magnet synchronous motor, by combining a generalized adaptive super-twisting observer and an improved quasi-resonant controller, high-performance current control of the permanent magnet synchronous motor is achieved. The specific steps include: S1: Establish the voltage equation of the permanent magnet synchronous motor.

[0022] In this embodiment, ignoring the hysteresis loss, eddy current and iron core saturation of the surface-mounted permanent magnet synchronous motor, the axis voltage equation of the permanent magnet synchronous motor in the synchronous rotating coordinate system is: , (1) Among them, and are respectively axis and axis stator voltages; and are respectively axis and axis stator currents; is the stator resistance; and are respectively axis and axis stator inductances; is the electrical angular velocity; is the rotor flux.

[0023] For the surface-mounted permanent magnet synchronous motor, it is expressed as: (2) Discretize the axis voltage equation through first-order forward difference to obtain the discrete voltage equation as: , (3) To ensure that the actual current can accurately track the reference current in the next cycle under the action of the predicted control voltage, and are replaced with the reference currents and , and the predicted voltage equation at time is obtained as: , (4) Among them, and are respectively axis and The stator voltage at the moment; is the sampling period; is the reference current value of the axis at the moment; is the axis at the moment; is the stator current of the axis at the moment; is the stator current of the axis at the moment;

[0024] Due to the delay of electronic components, the control voltage calculated at the moment cannot be applied until the moment. Therefore, a one-step delay compensation needs to be performed on the voltage prediction equation and it is modeled as: (5) where represents the control voltage at the moment, represents the reference voltage at the

[0025] Considering one-step delay compensation, rewrite equation (4) as: , (6) Since the sampling period is very short, then , , . Therefore, the traditional deadbeat voltage control equation considering one-step delay compensation is: , (7) where and are respectively the axis and the axis at the moment; and are respectively the axis and the axis at the moment; and are respectively the axis and the The stator current at ; Stator inductance; Sampling period; Stator resistance; is the electrical angular velocity at and

[0026] S2: Based on the voltage equation and current loop disturbance of the permanent magnet synchronous motor, establish a deadbeat predictive control model.

[0027] In this embodiment, the aperiodic disturbances in the current loop of the permanent magnet synchronous motor mainly include d-q axis cross-coupling disturbance, parameter perturbation, and external uncertain disturbance; expressed as: (8) (9) Where and are axis and axis aperiodic disturbances respectively; and are axis and axis cross-coupling disturbances respectively; and are axis and axis known disturbances respectively; and are axis and axis disturbances caused by parameter perturbation respectively; and are axis and axis external uncertain disturbances respectively; and are axis and axis stator currents respectively; is axis and axis nominal value of stator inductance; is the nominal value of stator resistance, is the nominal value of rotor flux; , , .

[0028] The periodic disturbances in the current loop of the permanent magnet synchronous motor mainly include flux harmonics caused by motor manufacturing processes and voltage harmonics caused by inverter nonlinearities. The specific analysis processes of flux harmonics and voltage harmonics are as follows: Ideally, the magnetic flux of a permanent magnet synchronous motor is sinusoidally distributed. However, in an actual permanent magnet synchronous motor system, due to manufacturing limitations and magnetic saturation, it is difficult to obtain an ideal sinusoidal magnetic flux distribution, which results in harmonics in the actual magnetic flux. The permanent magnet synchronous motor The magnetic flux harmonics of the axis are expressed as: where and are the magnetic flux harmonics of the axis and the axis respectively; and are the amplitudes of the 6n - th magnetic flux harmonics of the axis and the axis respectively; is the electrical angular velocity.

[0029] In an IGBT voltage - source inverter, a certain dead - time needs to be set to prevent the upper and lower switching devices on the same side from closing simultaneously, which may cause an inverter short - circuit. However, the dead - time will cause voltage harmonics. The voltage harmonics of the d - q axes of the permanent magnet synchronous motor are expressed as Equation (11) where and are the voltage harmonics of the axis and the axis respectively; and are the dead - time and the sampling period respectively; is the DC bus voltage.

[0030] Combining the magnetic flux harmonics and the voltage harmonics, the periodic disturbance of the axis of the permanent magnet synchronous motor is expressed as: Equation (12) where and are the periodic disturbances of the axis and the axis respectively; and are the stator inductances of the axis and the axis respectively; Furthermore, the first - order super - local mathematical model is expressed as: Equation (13) where and are the output and input variables of the model respectively; is the set of known and unknown disturbances; is a proportionality factor; According to Equation (12), the final expression of the first-order superlocal mathematical model in the current loop is obtained as: (14) where and are the total disturbances on the d-axis and q-axis respectively; is the control gain.

[0031] Based on the voltage equation of the permanent magnet synchronous motor and the current loop disturbance, the deadbeat predictive control model of the permanent magnet synchronous motor based on the superlocal model is obtained, expressed as: (15).

[0032] S3: Based on the adaptive theory and the super-twisting technique, a generalized adaptive super-twisting observer is established.

[0033] The extended state observation model of the d-q axis current equation of the permanent magnet synchronous motor is established as: (16) where , , ; ; , is the control gain; ; ; ; is the first derivative of the total disturbance on the

[0034] Based on the extended state observation model (16), as well as the adaptive theory and the super-twisting technique, the generalized adaptive super-twisting observer is: (17) The error correction term is designed as: (18) where is the estimated value of the state vector; is the time; is the sampling period; is the control benefit; is the stator voltage at time ; and are the gain values of the observer; is the variable power exponent term, and ; is the adaptive gain term; and is the error correction term.

[0035] The designed adaptive gain term is (19) where and are adjustable parameters; is the gain, is the arctangent function.

[0036] According to the pole placement strategy of the observer, the gain value is adjusted to: (20) where is the bandwidth of the observer.

[0037] Then the error equation of the system is: (21) Judge the stability of the obtained generalized adaptive super-twisting observer, specifically including: Construct the Lyapunv function as: (22) where P is a positive definite symmetric matrix; , and its derivative with respect to time is: (23) where ; .

[0038] Let (24) where , and on this basis, by choosing appropriate matrices R and S, it can be ensured that .

[0039] Let there exist a positive definite symmetric matrix P and a positive constant satisfying the following matrix inequality: (25) Then it is proved that the permanent magnet synchronous motor system is finite-time stable.

[0040] At this time, the derivative of the Lyapunv function can be obtained as: (26) According to Equation (22), we can get: (27) where is the eigenvalue of P, and 。

[0041] Case 1: When holds, Equation (26) is written as: (28) From Equation (27), the inequality holds, so Equation (28) is written as (29) Case 2: When holds, Equation (26) is written as (30) From Equation (27), the inequality holds, so Equation (30) is written as (31) Lemma 1: For the dynamic system , and , , there exists a positive definite matrix satisfying , represents a dynamic system function, , and represent three gains, which are parameters without special significance, represents a positive definite matrix.

[0042] If and , then the system is finite-time stable, and the settling time , is the initial state; if and , then the system is finite-time stable, and the settling time , represents the settling time of the system.

[0043] According to Lemma 1, it can be seen that the second convergence condition is satisfied, and the convergence times for the two cases are:[[]] (32) (33) Compared with the existing super-twisting state observer, the present application realizes a dynamic adaptive gain that can be dynamically adjusted according to the change of the error input. Specifically, when holds, the power exponent is set to ; when holds, the power exponent is adjusted to , thus improving the convergence speed of the observer. In addition, by introducing a linear error term and combining it with the non-linear term, the ability of disturbance estimation is further improved. Further, an adaptive gain is introduced to further optimize the performance of the observer, improving the convergence speed and estimation accuracy while preventing overcompensation.

[0044] S4: Obtain a current loop controller through the deadbeat predictive control model and the generalized adaptive super-twisting observer.

[0045] The control law of the current loop controller is: (34).

[0046] S5: Establish a quasi-resonant controller by adopting a conditional switching resonant mode, and jointly apply the quasi-resonant controller and the deadbeat predictive control model to the current loop controller to obtain a final current loop controller.

[0047] For the periodic disturbance in the permanent magnet synchronous motor system, this application designs an improved quasi-resonant controller by adopting a conditional switching resonant mode to suppress the harmonics caused by flux harmonics and inverter non-linearity. The transfer function of the quasi-resonant controller is: (35) Where, is a step function that switches at ; is an adjustable switching point; is the current tracking error, is the reference current, is the stator current; is the resonant gain; is the damping frequency; is the Laplace variable; is the resonant frequency; After discretizing the quasi-resonant controller, the discretized transfer function is: (36) Where, is the complex frequency domain of the discrete-time signal; .

[0048] It can be understood that when the permanent magnet synchronous motor system is in the dynamic process, if , the resonant term in the quasi-resonant controller is disabled and its state is reset to zero. When the permanent magnet synchronous motor system approaches the steady state, if , these resonant terms are enabled again. This application avoids the side effects of the traditional quasi-resonant controller during the dynamic operation process, while maintaining its performance under steady-state conditions, thus significantly improving its effectiveness.

[0049] Further, the discretized transfer function is added to the current loop controller, and combined with the deadbeat predictive control model, to obtain the final current loop controller expressed as: (37).

[0050] In the deadbeat predictive control method proposed in this application, the generalized adaptive super-twisting observer is used to estimate and compensate the non-periodic disturbance in the current loop disturbance; the quasi-resonant controller is implemented in parallel with the deadbeat predictive control model to estimate and compensate the periodic disturbance in the current loop disturbance. Compared with the traditional deadbeat predictive control method, this application can efficiently estimate and compensate the disturbance existing in the current loop, and realizes high-performance current control.

[0051] Please refer to Figure 3 , which is the model structure diagram of the permanent magnet synchronous motor servo control system described in the embodiment of the present invention. In order to verify the effectiveness of the permanent magnet synchronous motor current control strategy proposed in the present invention, the performance of the deadbeat predictive control method combining the generalized adaptive super-twisting observer and the improved quasi-resonant control is analyzed in combination with the experimental results. The structure diagram of the servo control system of the permanent magnet synchronous motor based on field-oriented control and the proposed current loop control scheme is as Figure 3 shown. In the experiment shown in Figure 3 , the PI control is adopted for the speed loop, and the deadbeat predictive control method combining the generalized adaptive super-twisting observer and the improved quasi-resonant control proposed in this application is adopted for the current loop. The specific process includes: First, the mechanical angular velocity reference input of the permanent magnet synchronous motor is given as , the electrical angle and the angular velocity of the motor are obtained by using an encoder. The electrical angle is used for coordinate transformation, and the angular velocity is used for speed closed-loop control to obtain the q-axis reference current for current loop control; then, the a-phase, b-phase and c-phase currents , and of the motor are obtained by using current sensors, and then the actual current on the axis of the motor is obtained through coordinate transformation for current control to obtain the current loop control input ; further, the switching signal state vector of the inverter is obtained through coordinate transformation and the SVPWM pulse modulation method; finally, the inverter outputs three-phase currents according to the DC bus voltage and the switching signals to drive the permanent magnet synchronous motor to operate.

[0052] Based on Figure 3 the model of the permanent magnet synchronous motor servo control system, the comparative experimental results of the traditional deadbeat predictive current control method and the deadbeat predictive control method proposed in this application are given.

[0053] Please refer to Figure 4 Figure for the comparative experimental results of the d-axis and a-phase current curves; Figure 5 Figure for the comparative experimental results of the q-axis current curve. Specifically, it includes: When the speed reference is rpm and a sudden 3 step disturbance is applied at 10 seconds, from Figure 4 and Figure 5 it can be obtained that the current fluctuations of the axis and axis using the method of this application are 0.129 A and 0.147 A respectively, which are much smaller than the current fluctuations of the axis and axis of 0.288 A and 0.246 A using the traditional method; in addition, the total harmonic contents of the phase and axis currents using the method of this application are 1.42% and 1.78% respectively, which are much smaller than the total harmonic contents of the axis and axis currents of 3.39% and 2.63% using the traditional method; and after a sudden load is applied at 10 seconds, the times for the axis currents using the method of this application and the traditional method to enter the steady state are 1.875 s and 1.884 s respectively. Based on this, it is verified that the method of this application has better dynamic and steady-state performance and can achieve more robust and smooth permanent magnet synchronous motor current control.

[0054] Please refer to Figure 6 Figure for the comparative experimental results of the Fourier analysis of the q-axis current. Using the method of this application, the 6th and 12th harmonics of the axis current are significantly reduced, verifying that the deadbeat predictive control method proposed in this application can effectively estimate and compensate for periodic disturbances.

[0055] Please refer to Figure 7 Figure for the comparative experimental results of the axis and phase currents when the inductor is mismatched; Figure 8 Figure for the comparative experimental results of the axis current when the inductor is mismatched. Specifically, it includes: Under the condition of inductor parameter mismatch, from Figure 7It can be seen that, compared with the traditional method, the method proposed in this application is significantly less sensitive to the inductor parameters, and the mismatch of the inductor parameters has almost negligible impact on the shaft current. From Figure 8 it can be seen that during the dynamic characteristics during inductor mismatch, the shaft current overshoot is the largest when using the traditional method. At and , the current overshoot values are 0.885 A and 0.5 A respectively, while the current overshoot using the method of this application is the smallest, which are 0.384 A and 0.206 A respectively; in addition, the recovery times using the method of this application at and are 26 ms and 16 ms respectively. In contrast, the recovery times using the traditional method are 29 ms and 23 ms respectively. Further, compared with the steady-state current after inductor mismatch, the method of this application shows the smallest change in current variation, and the total harmonic values are 1.78% (at ) and 1.68% (at ) respectively, while the change using the traditional method is the largest, and the total harmonic values are 3.33% and 2.34% respectively. And it can be seen from the current loop controller expression that the motor parameters resistance and the permanent magnet flux linkage will not affect the control performance of the proposed algorithm. It is verified that the deadbeat predictive control method proposed in this application has stronger robustness and can achieve more accurate current control.

[0056] It should be understood that various forms of the processes shown above can be used, steps can be reordered, added or deleted. For example, the steps recorded in the disclosure of the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution disclosed in the present invention can be achieved, and no limitation is made herein.

[0057] The above specific embodiments do not constitute a limitation to the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub - combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A deadbeat predictive control method, characterized in that, Including: Establish the voltage equation of a permanent magnet synchronous motor; Based on the voltage equation of the permanent magnet synchronous motor and current loop disturbances, establish a deadbeat predictive control model; According to the adaptive theory and super-twisting technique, establish a generalized adaptive super-twisting observer; Through the deadbeat predictive control model and the generalized adaptive super-twisting observer, obtain a current loop controller; Adopt a conditional switching resonant mode to establish a quasi-resonant controller, and jointly apply the quasi-resonant controller and the deadbeat predictive control model to the current loop controller to obtain the final current loop controller.

2. The deadbeat predictive control method according to claim 1, characterized in that The expressions of the non-periodic disturbances in the current loop disturbances are respectively: ， ， Among them, and are respectively the aperiodic perturbations of the axis and the axis; and are respectively the cross-coupling perturbations of the axis and the axis; and are respectively the known perturbations of the axis and the axis; and are respectively the perturbations of the axis and the axis caused by parameter perturbations; and are respectively the external uncertain perturbations of the axis and the axis; and are respectively the stator currents of the axis and the axis; is the nominal value of the stator inductance of the axis and the axis; is the nominal value of the stator resistance, is the nominal value of the rotor flux; is ; The expressions of the periodic disturbances in the current loop disturbances are respectively: ， ， Among them, and are respectively the flux harmonics of the axis and the and are respectively the amplitudes of the 6n-th flux harmonics of the axis; is the electrical angular velocity; and are respectively the voltage harmonics of the axis and the and are respectively the dead time and the sampling period; is the DC bus voltage; Then the permanent magnet synchronous motor The periodic disturbance of the shaft is expressed as: ， Among them, and are respectively the periodic perturbations of the axis and the and are respectively the stator inductances of the axis and the The first-order super-twisting mathematical model is expressed as: ， Among them, and are the output and input variables of the model respectively; is the set of known and unknown perturbations; is a scaling factor; Then the first-order super-twisting mathematical model in the current loop is finally expressed as: ， Among them, and are the total disturbances on the d-axis and q-axis, respectively; is the control gain.

3. The deadbeat predictive control method according to claim 2, characterized in that, In the step of establishing a deadbeat predictive control model based on the voltage equation of the permanent magnet synchronous motor and current loop disturbances, it specifically includes the following steps: The voltage equation of the permanent magnet synchronous motor is expressed as: ， ， Wherein, and are respectively axis and axis at reference voltage at the moment; and are respectively axis and axis at reference current value at the moment; and are respectively axis and axis at stator current at the moment; stator inductance; is the sampling period; is the stator resistance; is electrical angular velocity at the moment, is the rotor flux; According to the voltage equation of the permanent magnet synchronous motor and current loop disturbances, obtain the deadbeat predictive control model of the permanent magnet synchronous motor based on the super-twisting model, expressed as: ， Among them, and are the total disturbances on the d-axis and q-axis respectively; is the control gain.

4. The deadbeat predictive control method according to claim 3, wherein In the step of establishing a generalized adaptive super-twisting observer according to the adaptive theory and super-twisting technique, it specifically includes the following steps: Establish the extended state observation model of the d-q axis current equation of the permanent magnet synchronous motor as: ， Among them, , , ; ; , is the control gain; ; ; ; is the first derivative of the total disturbance of the Based on the extended state observation model, as well as the adaptive theory and super-twisting technique, the generalized adaptive super-twisting observer is: ， The error correction term is designed as: ， Among them, is the estimated value of the state vector; is the time; is the sampling period; is the control benefit; is the stator voltage at time; ; and are the gain values of the observer; is the variable power exponent term, and ; is the adaptive gain term; and are the error correction terms.

5. A deadbeat predictive control method according to claim 4, wherein In the step of obtaining the current loop controller through the deadbeat predictive control model and the generalized adaptive super-twisting observer, the control law of the current loop controller is: 。 6. The deadbeat predictive control method according to claim 5, wherein In the step of adopting a conditional switching resonant mode to establish a quasi-resonant controller, and jointly applying the quasi-resonant controller and the deadbeat predictive control model to the current loop controller to obtain the final current loop controller, it specifically includes the following steps: The transfer function of the quasi-resonant controller is: ， wherein, is a step function that switches at ; is an adjustable switching point; is the current tracking error, is the reference current, is the stator current; is the resonance gain; is the damping frequency; is the Laplace variable; is the resonance frequency; Discretize the quasi-resonant controller to obtain the discretized transfer function as: ， Among them, is the complex frequency domain of the discrete-time signal; .

7. A deadbeat predictive control method according to claim 6, characterized in that Add the discretized transfer function to the current loop controller, and combine it with the deadbeat predictive control model to obtain the final current loop controller expressed as: 。 8. A deadbeat predictive control method according to claim 1, characterized in that Use the generalized adaptive super-twisting observer to estimate and compensate the non-periodic disturbances in the current loop disturbances; Use the quasi-resonant controller and the deadbeat predictive control model to be implemented in parallel to estimate and compensate the periodic disturbances in the current loop disturbances.

Images