Improved permanent magnet synchronous motor speed loop control method of deadbeat predictive speed control

Abstract

This invention belongs to the field of motor control technology, and particularly relates to an improved deadbeat predictive speed control method for permanent magnet synchronous motors (PMSMs). It includes: S1: Constructing a composite error correction cascaded extended state observer based on the PMSM speed loop model; S2: Calculating a state identification index based on the real-time speed of the PMSM, and inputting the state identification index to a state-aware discrete iterative learning controller, so that it outputs periodic disturbance compensation only in steady state; introducing a state-aware bandwidth switching mechanism into the composite error correction cascaded extended state observer, which adaptively adjusts the bandwidth parameter based on the state identification index and outputs an aperiodic disturbance estimate; S3: Inputting the periodic disturbance compensation and the aperiodic disturbance estimate to a deadbeat predictive speed control law to achieve smooth speed control of the PMSM. This invention can improve the overall disturbance rejection capability of the PMSM speed control system.

Description

Technical Field

[0001] This invention belongs to the field of motor control technology, and in particular relates to an improved speed loop control method for permanent magnet synchronous motors with deadbeat predictive speed control. Background Technology

[0002] Permanent magnet synchronous motors (PMSMs) occupy a core position in high-performance drive fields such as transportation electrification and industrial servo systems due to their high power density, high efficiency, and excellent dynamic response characteristics. Deadbeat predictive speed control (DPSC) is widely used due to its fast dynamic response and accurate tracking capabilities, but its fault tolerance to model errors and external disturbances is weak. Unmodeled factors such as parameter mismatch and load disturbances can lead to a decrease in control performance, a deterioration in tracking accuracy, and even system instability. Summary of the Invention

[0003] In view of this, the present invention aims to provide an improved speed loop control method for permanent magnet synchronous motors with deadbeat predictive speed control, to address the problem that existing control methods have weak fault tolerance to model errors and external disturbances. Unmodeled factors such as parameter mismatch and load disturbances can lead to decreased control performance, deteriorated tracking accuracy, and even system instability. This invention improves the structure of traditional DPSC and traditional Cascade Extended State Observer (CESO), proposing an enhanced DPSC strategy based on Compound Error Correction-Cascade Extended State Observer (CEC-CESO) and a state-aware discrete iterative learning control compensation mechanism based on state recognition. It optimizes the observation performance of non-periodic disturbances through independent zero-pole control and suppresses periodic speed fluctuations through steady-state triggered discrete iterative learning compensation, thereby improving the DPSC's ability to suppress the above two types of disturbances.

[0004] To achieve the above objectives, the technical solution created by this invention is implemented as follows: An improved speed loop control method for permanent magnet synchronous motors with deadbeat predictive speed control specifically includes the following steps: S1: Establish a speed loop model of a permanent magnet synchronous motor containing disturbances, and construct a composite error correction cascaded extended state observer based on the speed loop model of the permanent magnet synchronous motor; S2: Calculate the state identification index based on the real-time speed of the permanent magnet synchronous motor, and input the state identification index into the state-aware discrete iterative learning controller so that the state-aware discrete iterative learning controller outputs the periodic disturbance compensation amount in steady state. A state-aware bandwidth switching mechanism is introduced into the composite error correction cascaded extended state observer. The state-aware bandwidth switching mechanism adaptively adjusts the bandwidth parameter based on the state identification index and outputs an aperiodic disturbance estimate. S3: Input the periodic disturbance compensation and the aperiodic disturbance estimate into the deadbeat predictive speed control law to achieve smooth speed control of the permanent magnet synchronous motor.

[0005] Furthermore, in step S1, the expression for the speed loop model of the permanent magnet synchronous motor containing disturbances is: ; in, The mechanical angular velocity at the (k+1)th sampling time. Let be the mechanical angular velocity at the k-th sampling time. The sampling period of the velocity loop. This refers to the nominal control gain of the speed loop. Let q be the reference current at the k-th sampling time. This represents the lumped disturbance at the k-th sampling time.

[0006] Furthermore, in step S1, the composite error correction cascaded extended state observer includes a first-stage discrete observer and a second-stage discrete observer, wherein the expression for the first-stage discrete observer is: ; The expression for the second-level discrete observer is: ; in, For the first-level discrete observer, For the second-level discrete observer, and These are the rotational speed estimates for the first-stage discrete observer and the second-stage discrete observer, respectively. and These represent the rotational speed estimation errors of the first-level discrete observer and the second-level discrete observer, respectively. Let be the mechanical angular velocity at the k-th sampling time. The sampling period of the velocity loop. This refers to the nominal control gain of the speed loop. Let q be the reference current at the k-th sampling time. and These are the aperiodic perturbations estimated by the first-level discrete observer and the second-level discrete observer, respectively. This is a composite error correction term. Here, 'a' represents the auxiliary state variable of the first-stage discrete observer, 'b' represents the cross-gain, and 'a' represents the correction gain. , and , These are the gain coefficients of the first-stage discrete observer and the second-stage discrete observer, respectively. The dynamic bandwidth parameter for the composite error correction cascaded extended state observer.

[0007] Furthermore, in step S2, the specific process for calculating the state identification index is as follows: ; ; Where J is the state identification index, When the permanent magnet synchronous motor is in steady state, The time indicates that the permanent magnet synchronous motor is operating dynamically, and sgn() represents the sign function. For speed tracking error, This is the preset error tolerance band. For speed tracking error The discrete-domain expression obtained after the transformation and These are the reference mechanical angular velocity and the actual mechanical angular velocity, respectively. Domain expression, It is a positive integer with extremely small values.

[0008] Furthermore, in step S2, the specific process of inputting the state recognition index into the state-aware discrete iterative learning controller for processing is as follows: ; in, For the state-aware discrete iterative learning controller in the first... The periodic perturbation compensation amount obtained in the next iteration Domain expression, For Q-filters, Let J be the learning filter, and J be the state recognition metric. For the first Speed ​​tracking error during the next iteration Domain expression.

[0009] Furthermore, in step S2, the process by which the state-aware bandwidth switching mechanism adaptively adjusts the bandwidth parameters based on state recognition indicators is as follows: ; ; ; ; ; in, This is the bandwidth expansion factor. For the perturbation state variable, Gain is adjusted for bandwidth. For observer bandwidth parameters, The nominal bandwidth parameter of the observer. This is a reset signal. Here, k is the sampling time index, and A is the state change indicator. For a positive constant with extremely small values, sgn() denotes the sign function. The rate of change of the speed command. The preset instruction change rate threshold, For reference mechanical angular velocity Domain expression, This is the sampling period for the velocity loop.

[0010] Furthermore, the specific process of inputting the periodic disturbance compensation and the aperiodic disturbance estimate into the deadbeat predictive speed control law for processing is as follows: ; ; in, This is the q-axis reference current at the (k+1)th sampling time. This refers to the nominal control gain of the speed loop. The sampling period of the velocity loop. For reference mechanical angular velocity Domain expression, This represents the rotational speed estimate of the first-level discrete observer at the (k+1)th sampling time. and These are the aperiodic perturbations estimated by the first-level discrete observer and the second-level discrete observer, respectively. This is the compensation amount for periodic disturbances. This is an estimate of the non-periodic disturbance.

[0011] Compared with the prior art, the present invention can achieve the following beneficial effects: This invention presents an improved deadbeat predictive speed control method for permanent magnet synchronous motors, where the zeros and poles of the traditional CESO (Cyclic Emission Predictive Speed ​​Control) vary with the bandwidth parameter. The overall movement of the zeros cannot be independently controlled. This invention introduces a cross-gain 'a' to reconstruct the zero distribution. As the cross-gain 'a' increases, the previously overlapping double zeros separate, with one zero moving closer to the unit circle boundary. This strengthens the observer's quasi-integral effect and reduces the amplitude of disturbance estimation errors in the low-frequency band. However, this inevitably increases the risk of raising the intermediate frequency (IF) sensitive peak. To suppress the risk of IF sensitive peak rise caused by the cross-gain 'a', this invention performs secondary optimization of the pole distribution through a correction gain 'b'. The traditional CESO's poles are densely distributed near the unit circle, which is the core reason for the prominent IF gain peak. After introducing the correction gain 'b', the heavy poles separate on both sides along the real axis, which can improve the bandwidth parameter... Under the premise of not changing, the damping characteristics of the observer are independently adjusted to reduce the mid-frequency gain peak and improve the overall anti-disturbance capability of the observer. This invention improves the non-periodic disturbance observation performance through independent zero-pole control and compensates for periodic disturbances through state-aware discrete iterative learning control, thereby improving the overall anti-disturbance capability of the permanent magnet synchronous motor speed control system. Attached Figure Description

[0012] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments and descriptions of the invention are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings: Figure 1 A schematic flowchart of the improved deadbeat predictive speed control method for permanent magnet synchronous motors described in the embodiments of the present invention; Figure 2 The improved DPSC permanent magnet synchronous motor speed control block diagram described in the embodiment of the present invention; Figure 3 The Bode plot and zero-pole distribution plot of the perturbation estimation error transfer function under different cross-gain a are shown in the embodiments of the present invention. (a) is the Bode plot of the perturbation estimation error transfer function under different cross-gain a, and (b) is the zero-pole distribution plot under different cross-gain a. Figure 4 The Bode plot and zero-pole distribution plot of the perturbation estimation error transfer function under different correction gains b are shown in the embodiments of the present invention. (a) is the Bode plot of the perturbation estimation error transfer function under different correction gains b, and (b) is the zero-pole distribution plot under different correction gains b. Figure 5 The Bode plots of the perturbation estimation error transfer functions of CESO and CEC-CESO described in the embodiments of the present invention are shown in the following comparison diagrams: (a) is the Bode plot of the perturbation estimation error transfer function of the two observers, and (b) is the Bode plot of the perturbation estimation transfer function of the two observers. Figure 6Comparison of dynamic speed response under sudden load increase and decrease conditions using two different control methods described in the embodiments of the present invention; Figure 7 A comparison diagram of the dynamic response of speed step tracking under two different control methods described in the embodiments of the present invention; Figure 8 A comparison diagram of steady-state speed waveforms for two different control methods described in the embodiments of the present invention. Detailed Implementation

[0013] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not constitute a limitation thereof.

[0014] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0015] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on this invention. Furthermore, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0016] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0017] The invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0018] like Figure 1As shown, this invention proposes an improved speed loop control method for permanent magnet synchronous motors with deadbeat predictive speed control, specifically including the following steps: S1: Establish a speed loop model of a permanent magnet synchronous motor containing disturbances, and construct a composite error correction cascaded extended state observer based on the speed loop model of the permanent magnet synchronous motor; S2: Calculate the state identification index based on the real-time speed of the permanent magnet synchronous motor, and input the state identification index into the state-aware discrete iterative learning controller so that the state-aware discrete iterative learning controller outputs the periodic disturbance compensation amount in steady state. A state-aware bandwidth switching mechanism is introduced into the composite error correction cascaded extended state observer. The state-aware bandwidth switching mechanism adaptively adjusts the bandwidth parameter based on the state identification index and outputs an aperiodic disturbance estimate. S3: Input the periodic disturbance compensation and the aperiodic disturbance estimate into the deadbeat predictive speed control law to achieve smooth speed control of the permanent magnet synchronous motor.

[0019] It should be noted that this invention designs a composite error correction cascaded extended state observer to achieve independent control of zeros and poles, used to estimate the aperiodic disturbances of the speed loop of a permanent magnet synchronous motor (DPSC), thereby improving the robustness of the DPSC system under parameter perturbations and external load disturbances. This invention also introduces a state-aware bandwidth switching mechanism to simultaneously achieve adaptive switching of the observer bandwidth parameter and steady-state activation of discrete iterative learning control, thus realizing the coordinated suppression of periodic and aperiodic disturbances.

[0020] Furthermore, this invention first constructs a Composite Error Correction Cascaded Extended State Observer (CEC-CESO), adjusting the zero and pole distributions of CEC-CESO through cross-gain and correction gain respectively to improve the observation performance of aperiodic disturbances. Second, a state-aware bandwidth switching mechanism is introduced into CEC-CESO, adjusting the CEC-CESO bandwidth parameters according to the identified operating state to form an adaptive composite error correction cascaded extended state observer A-CEC-CESO. Further, a state-aware discrete iterative learning controller SAILC is set and its activation is limited to the steady-state period to compensate for periodic disturbances. The aperiodic disturbance estimate output by A-CEC-CESO and the periodic disturbance compensation output by SAILC are jointly introduced into a deadbeat predictive speed control law, achieving synergistic suppression of combined periodic and aperiodic disturbances.

[0021] Figure 2 In this context, DPCC stands for Deadbeat Predictive Current Controller, SVPWM stands for Space Vector Pulse Width Modulation, and VSI stands for Voltage Source Inverter. This indicates that the d-axis reference current setting is zero. For the first The actual current vector along the dq axis at each sampling time. The q-axis component of the actual current vector along the dq axis; stationary Stator voltage reference vector in coordinate system The electrical angle of the rotor of the permanent magnet synchronous motor. , and These are the stator currents of phases a, b, and c of the motor, respectively. This refers to the DC bus voltage of a voltage source inverter. Unit delay operator; This indicates that the coordinate system dq is rotated to rest. Coordinate transformation of a coordinate system P represents the coordinate transformation from the three-phase abc coordinate system to the dq rotating coordinate system. n This represents the number of pole pairs in a permanent magnet synchronous motor.

[0022] The specific implementation method of the present invention is as follows: To improve the estimation accuracy of deadbeat predictive speed control systems for aperiodic disturbances and address the inherent limitations of traditional cascaded extended state observers, such as zero-pole coupling and prominent mid-frequency gain peaks, a speed loop model of a permanent magnet synchronous motor (PMSM) considering the disturbance effect is established. Based on this model, a composite error-corrected cascaded extended state observer is constructed. The mechanical dynamics equations of the PMSM can be expressed as follows: (1); In the formula, This refers to the mechanical angular velocity of the permanent magnet synchronous motor. The nominal control gain of the speed loop can be expressed as: , , , These represent the number of pole pairs, nominal value of permanent magnet flux linkage, and nominal value of moment of inertia of a permanent magnet synchronous motor, respectively. This is the q-axis reference current; This is the lumped interference in the motor speed loop.

[0023] After forward Euler discretization, equation (1) can be written as: (2); in, , and They represent the first The mechanical angular velocity measured at each sampling time, the q-axis reference current at the k-th sampling time, and the velocity loop sampling period. Lumped disturbance. Due to non-periodic disturbances and periodic disturbances Composition, namely: (3); in, This represents the non-periodic disturbance caused by factors such as parameter mismatch of permanent magnet synchronous motor, load changes, current tracking error, and unmodeled dynamics. This indicates the periodic disturbance caused by factors such as current sampling error, inverter nonlinearity, cogging torque, and flux harmonics.

[0024] The composite error correction cascaded extended state observer uses a two-stage cascaded architecture to achieve decoupled control of zeros and poles. The expression of its first-stage discrete observer is as follows: (4); Its second-level discrete observer expression is as follows: (5); In the formula, Let be the mechanical angular velocity at the k-th sampling time. , These are the rotational speed estimates for the first-stage discrete observer and the second-stage discrete observer, respectively. , These are the rotational speed estimation errors of the first-stage discrete observer and the second-stage discrete observer, respectively. , These are the non-periodic perturbations estimated by the first-level discrete observer and the second-level discrete observer, respectively. These are auxiliary state variables for the first-level discrete observer, used for iterative updates of perturbations; This is a composite error correction term, used to introduce the speed difference feedback between the two-stage observers to improve the disturbance estimation performance; , and , , respectively, are the gain coefficients of the first-stage discrete observer and the second-stage discrete observer; a is the cross gain, used to adjust the zero distribution; b is the correction gain, used to adjust the pole distribution and thus improve the damping of the observer system; The dynamic bandwidth parameter for the composite error correction cascaded extended state observer; This is the sampling period for the velocity loop.

[0025] To achieve independent zero-pole adjustment, the gain of the composite error-corrected cascaded extended state observer is parametrically designed using the pole placement method. The characteristic poles of both discrete observer stages are placed at positions corresponding to the nominal bandwidth. Finally, the gain coefficient of the composite error-corrected cascaded extended state observer can be expressed as: (6); Through this gain mapping relationship, the zero-point distribution of the composite error correction cascaded extended state observer is determined by the cross gain a and... The pole distribution is determined by the correction gain b and The decision removes the traditional CESO zero-pole dual-bandwidth parameter. The limitations of coupled control increase the degree of freedom in adjusting the observer's performance.

[0026] To perform performance analysis on the enhanced DPSC system based on CEC-CESO proposed in this invention, the perturbation estimation error transfer function of CEC-CESO is derived by combining equations (2), (4), and (5). Its expression is: (7); in, , , , , , .

[0027] In the formula, , The zero-point coefficients of the perturbation estimation error transfer function of CEC-CESO are given by the cross-gain a and Joint decision, , The first-level pole coefficients are determined by the correction gain b and... Joint decision; , These are the second-order pole coefficients, determined solely by the bandwidth parameter. The decision was made. This transfer function verifies that the composite error-corrected cascaded extended state observer achieves a decoupled zero-pole configuration under a fixed bandwidth parameter. The observer performance can be adjusted by independently adjusting the zero and pole positions using two parameters, a and b.

[0028] The zeros and poles of a traditional CESO will vary with the bandwidth parameter. The overall movement cannot be independently controlled; however, this invention introduces a cross-gain 'a' to reconstruct the zero-point distribution. As 'a' increases, the originally overlapping double zeros separate, with one of the zeros moving closer to the unit circle boundary. This strengthens the quasi-integral effect of the composite error correction cascaded extended state observer, reducing the amplitude of disturbance estimation errors in the low-frequency band. However, this inevitably increases the risk of raising the intermediate frequency sensitive peak of the composite error correction cascaded extended state observer.

[0029] To mitigate the risk of increased mid-frequency sensitive peaks caused by cross-gain 'a', this invention performs secondary optimization of the pole distribution through a correction gain 'b'. Traditional CESO poles are densely distributed near the unit circle, which is the core reason for the prominent mid-frequency gain peaks. By introducing the correction gain 'b', the heavy poles are separated along the real axis, which can improve bandwidth parameters. Under the premise of not changing, the damping characteristics of the composite error correction cascaded expansion state observer are independently adjusted to reduce the mid-frequency gain peak and improve the overall anti-interference capability of the observer.

[0030] Figure 3 The characteristic curves of the perturbation estimation error transfer function are given for different cross-gain 'a' values, where... Figure 3 (a) in the diagram is a Bode diagram. Figure 3 (b) in the diagram shows the corresponding pole-zero distribution. From... Figure 3 It can be seen that as 'a' increases, the amplitude of the disturbance estimation error in the low-frequency band gradually decreases, verifying the enhanced effect of zero-point reconstruction on the ability to suppress low-frequency disturbances; at the same time, the zeros gradually separate, with one of them moving closer to the unit circle, which is consistent with the theoretical analysis.

[0031] Figure 4 The characteristic curves of the perturbation estimation error transfer function under different correction gains b are shown, where Figure 4 (a) is a Bode diagram. Figure 4 (b) shows the corresponding zero-pole distribution diagram. As can be seen from the figure, within the parameter range, increasing the correction gain b can reduce the gain peak in the mid-frequency band, verifying the effect of pole reconstruction on suppressing the mid-frequency sensitivity peak, thereby improving the overall anti-interference performance of the observer; at the same time, the poles move separately to both sides along the real axis, and the damping characteristics are optimized.

[0032] To further verify the observer's perturbation estimation performance, the frequency domain characteristics of CEC-CESO and traditional CESO were compared, and the results are as follows: Figure 5 As shown, under the same bandwidth parameter configuration, the low-frequency disturbance estimation error amplitude of CEC-CESO is much lower than that of traditional CESO, confirming the enhanced effect of zero-point reconstruction on low-frequency disturbance suppression capability; at the same time, the correction gain b effectively weakens the gain peak in the mid-frequency band, avoiding excessive amplification of the mid-frequency signal.

[0033] Based on the construction of CEC-CESO, a state-aware bandwidth switching mechanism is introduced to dynamically adjust the bandwidth parameters of CEC-CESO according to the operating state, thereby forming A-CEC-CESO; and SAILC is activated when the system is in steady state to handle aperiodic and periodic disturbances respectively.

[0034] First, define a state identification index J to distinguish between steady-state and transient operating states. Its expression is: (8); In the formula, sgn( ) represents a sign function. This is for speed tracking error; This is the preset error tolerance band; This is a very small positive constant. When the speed error satisfies... When J=1, it indicates that the system has entered a steady state; otherwise, J=0, the system is in a transient state.

[0035] The reasons for system deviations from steady state can be categorized into two types: changes in speed command and external load disturbances. Indiscriminately increasing the observer bandwidth parameter will amplify noise during command tracking. To decouple the disturbance logic, the rate of change of speed command needs to be calculated to shield against interference from command dynamics. The expression for the rate of change of speed command is: (9); Further define the set signal With reset signal , Used to mark when the system enters a load disturbance transient state. Used to mark the system's recovery to a steady state, in order to accurately identify transient processes caused by load disturbances, its expression is: (10); In the formula, A is the state change indicator, which is used to characterize the jump direction of the state identification index J; The preset command change rate threshold is used. When the speed command change rate exceeds this threshold, it is determined to be a command tracking transient, and bandwidth parameter enhancement is not triggered; max ( () is the function that takes the maximum value; It is triggered only when the system deviates from the steady-state boundary due to load disturbances. The change rate of speed command is used to mask the transient of command tracking, so as to improve the reliability of load disturbance identification. Triggered when the system returns to the error tolerance band, used to reset the disturbance state.

[0036] Update the perturbation state variable using an algebraic latch. To maintain the adaptive bandwidth mode and avoid frequent switching during disturbance identification, its update logic is as follows: (11); Disturbance state variables When a load disturbance is detected, the bit is set and latched high until the system re-enters steady state and is cleared by a reset signal. This is based on the disturbance state variable. The observer's dynamic bandwidth enables adaptive switching, and the bandwidth parameter adjusts the gain. With final dynamic bandwidth parameters The expressions are as follows: (12); In the formula, This represents the bandwidth expansion factor. This adaptive bandwidth parameter design allows the system to quickly switch to a high bandwidth parameter when load disturbances occur, enhancing disturbance response speed and suppression capability; during steady-state operation or command tracing, it reverts to the nominal bandwidth parameter. To avoid amplifying high-frequency measurement noise and degrading steady-state speed fluctuations, this invention defines the CEC-CESO, which combines the above-mentioned state-aware bandwidth switching mechanism, as an adaptive composite error correction cascaded extended state observer, abbreviated as A-CEC-CESO.

[0037] Based on the A-CEC-CESO estimation of aperiodic disturbances, a state-aware discrete iterative learning controller is designed as a feedforward compensation channel to suppress periodic disturbances in the velocity loop caused by factors such as current sampling errors. The learning law expression for discrete iterative learning control is: (13); In the formula, It is a Q-filter used to improve the convergence conditions of the learning process and suppress the accumulation of high-frequency noise; This is a learning filter used to adjust the learning gain and phase compensation. For the state-aware discrete iterative learning controller in the first... The periodic perturbation compensation amount obtained in the next iteration of The domain expression is used, and the periodic disturbance compensation amount obtained in the current iteration is used as the controller output, i.e. Furthermore, it is uniformly used in the beat-free predictive speed control law. The control input for the initial iteration period is defined as follows: .

[0038] In traditional ILC (Integrated Discrete Control) systems, the accumulation of historical errors during system transients can lead to control input overshoot, prolonging settling time and even causing oscillations. To address this issue, a state-aware switch is constructed using the aforementioned state identification index J to strictly limit the activation timing of discrete iterative learning control. Its triggering logic is as follows: (14); This state-aware bandwidth switching mechanism ensures that discrete iterative learning control is activated only when the system is in steady state (J=1), updating the periodic disturbance compensation; during the system transient process (J=0), learning and updating cease, and zero compensation value is output. This design suppresses periodic speed fluctuations while reducing the risk of overcompensation and speed overshoot caused by historical learning errors during transient processes.

[0039] Finally, the periodic disturbance compensation amount output by SAILC Aperiodic disturbance estimates of A-CEC-CESO output Introduced into the deadbeat predictive control law, a collaborative compensation structure for aperiodic and periodic disturbances is formed to improve the system's suppression performance against these two types of disturbances. Based on this, the q-axis reference current for enhanced deadbeat predictive velocity control is obtained. The control law is expressed as follows: (15); In the formula, This is the reference speed command at the k-th moment; The aperiodic disturbance estimate of the A-CEC-CESO output satisfies ; This is the periodic disturbance compensation amount output by the discrete iterative learning controller; the calculated... As the input command of the current loop controller, after being regulated by the current loop closed loop and modulated by the space vector pulse width, the output drive voltage is sent to the permanent magnet synchronous motor.

[0040] To verify the effectiveness of the improved DPSC method proposed in this invention, namely its stronger ability to suppress periodic and aperiodic disturbances and its better velocity tracking performance, simulations were conducted under the same conditions using the traditional CESO-based DPSC method and the enhanced DPSC method proposed in this invention.

[0041] Figure 6 The dynamic speed response of the traditional CESO-type DPSC method and the present invention under sudden load increase and decrease conditions is demonstrated. Under the same simulation conditions, after a sudden change in load torque, the maximum speed deviation corresponding to the present invention is smaller than that of the traditional CESO-type DPSC method, and the time required for the speed to recover to the preset error range is shorter. This result shows that A-CEC-CESO switches the observer bandwidth parameter from the nominal bandwidth parameter to a higher dynamic bandwidth parameter after detecting load disturbance, thereby accelerating the estimation and compensation of aperiodic disturbances.

[0042] Figure 7 The dynamic responses of two control methods under step speed conditions are demonstrated. Under the same simulation conditions, the traditional method exhibits significant overshoot and a longer settling time; while the method proposed in this invention shows no overshoot in speed tracking, a shorter settling time, and superior dynamic response performance. These results indicate that when the speed command changes, the state-aware bandwidth switching mechanism does not trigger the observer's high-bandwidth parameter mode and pauses the discrete iterative learning control output, thereby reducing the impact of bandwidth parameter mis-switching and historical learning errors on the command tracking process.

[0043] Figure 8The speed waveforms of two control methods under steady-state conditions at rated load are shown. Under the same simulation conditions, the periodic speed fluctuation amplitude corresponding to this invention is smaller than that of the traditional CESO-type DPSC method, indicating that the steady-state activated discrete iterative learning control can compensate for periodic disturbances.

[0044] Depend on Figures 6 to 8 The results show that, under the simulated conditions of sudden load increase and decrease, speed step, and steady-state operation under rated load, compared with the traditional CESO-type DPSC method, the load disturbance speed deviation and periodic speed fluctuation corresponding to the present invention are smaller, and the recovery time and speed step adjustment time after disturbance are shorter.

[0045] It should be understood that the various forms of processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this invention disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this invention can be achieved, and this is not limited herein.

[0046] The specific embodiments described above do not constitute a limitation on the scope of protection of this 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 principles of this invention should be included within the scope of protection of this invention.

Claims

1. An improved speed loop control method for permanent magnet synchronous motors with deadbeat predictive speed control, characterized in that: Specifically, the steps include the following: S1: Establish a speed loop model of a permanent magnet synchronous motor containing disturbances, and construct a composite error correction cascaded extended state observer based on the speed loop model of the permanent magnet synchronous motor; S2: Calculate the state identification index based on the real-time speed of the permanent magnet synchronous motor, and input the state identification index into the state-aware discrete iterative learning controller so that the state-aware discrete iterative learning controller outputs the periodic disturbance compensation amount in steady state. A state-aware bandwidth switching mechanism is introduced into the composite error correction cascaded extended state observer. The state-aware bandwidth switching mechanism adaptively adjusts the bandwidth parameter based on the state identification index and outputs an aperiodic disturbance estimate. S3: Input the periodic disturbance compensation and the aperiodic disturbance estimate into the deadbeat predictive speed control law to achieve smooth speed control of the permanent magnet synchronous motor.

2. The improved deadbeat predictive speed control method for permanent magnet synchronous motors according to claim 1, characterized in that: In step S1, the expression for the speed loop model of the permanent magnet synchronous motor containing disturbances is: ； in, The mechanical angular velocity at the (k+1)th sampling time. Let be the mechanical angular velocity at the k-th sampling time. The sampling period of the velocity loop. This refers to the nominal control gain of the speed loop. Let q be the reference current at the k-th sampling time. This represents the lumped disturbance at the k-th sampling time.

3. The improved deadbeat predictive speed control method for permanent magnet synchronous motors according to claim 1, characterized in that: In step S1, the composite error correction cascaded extended state observer includes a first-stage discrete observer and a second-stage discrete observer, wherein the expression for the first-stage discrete observer is: ； The expression for the second-level discrete observer is: ； in, For the first-level discrete observer, For the second-level discrete observer, and These are the rotational speed estimates for the first-stage discrete observer and the second-stage discrete observer, respectively. and These represent the rotational speed estimation errors of the first-level discrete observer and the second-level discrete observer, respectively. Let be the mechanical angular velocity at the k-th sampling time. The sampling period of the velocity loop. This refers to the nominal control gain of the speed loop. Let q be the reference current at the k-th sampling time. and These are the aperiodic perturbations estimated by the first-level discrete observer and the second-level discrete observer, respectively. This is a composite error correction term. Here, 'a' represents the auxiliary state variable of the first-stage discrete observer, 'b' represents the cross-gain, and 'a' represents the correction gain. , and , These are the gain coefficients of the first-stage discrete observer and the second-stage discrete observer, respectively. The dynamic bandwidth parameter for the composite error correction cascaded extended state observer.

4. The improved deadbeat predictive speed control method for permanent magnet synchronous motors according to claim 1, characterized in that: In step S2, the specific process of calculating the state identification index is as follows: ； ； Where J is the state recognition index. When the permanent magnet synchronous motor is in steady state, The time indicates that the permanent magnet synchronous motor is operating dynamically, and sgn() represents the sign function. For speed tracking error, This is the preset error tolerance band. For speed tracking error The discrete-domain expression obtained after the transformation and These are the reference mechanical angular velocity and the actual mechanical angular velocity, respectively. Domain expression, It is a positive integer with extremely small values.

5. The improved deadbeat predictive speed control method for permanent magnet synchronous motors according to claim 4, characterized in that: In step S2, the specific process of inputting the state recognition index into the state-aware discrete iterative learning controller for processing is as follows: ； in, For the state-aware discrete iterative learning controller in the first... The periodic perturbation compensation amount obtained in the next iteration Domain expression, For Q-filters, Let J be the learning filter, and J be the state recognition metric. For the first Speed ​​tracking error during the next iteration Domain expression.

6. The improved deadbeat predictive speed control method for permanent magnet synchronous motors according to claim 4, characterized in that: In step S2, the process by which the state-aware bandwidth switching mechanism adaptively adjusts the bandwidth parameters based on state recognition indicators is as follows: ； ； ； ； ； in, This is the bandwidth expansion factor. For the perturbation state variable, Gain is adjusted for bandwidth. For observer bandwidth parameters, The nominal bandwidth parameter of the observer. This is a reset signal. Here, k is the sampling time index, and A is the state change indicator. For a positive constant with extremely small values, sgn() denotes the sign function. The rate of change of the speed command. The preset instruction change rate threshold, For reference mechanical angular velocity Domain expression, This is the sampling period for the velocity loop.

7. The improved deadbeat predictive speed control method for permanent magnet synchronous motors according to claim 1, characterized in that: The specific process of inputting the periodic disturbance compensation and the aperiodic disturbance estimate into the deadbeat predictive speed control law is as follows: ； ； in, This is the q-axis reference current at the (k+1)th sampling time. This refers to the nominal control gain of the speed loop. The sampling period of the velocity loop. For reference mechanical angular velocity Domain expression, This represents the rotational speed estimate of the first-level discrete observer at the (k+1)th sampling time. and These are the aperiodic perturbations estimated by the first-level discrete observer and the second-level discrete observer, respectively. This is the compensation amount for periodic disturbances. This is an estimate of the non-periodic disturbance.

Images