An observer-based model-free parameter robust predictive current control method for permanent magnet synchronous motor

Abstract

The application provides an observer-based model-free parameter permanent magnet synchronous motor robust predictive current control method, mainly including design of a current difference observer, determination of current difference observer coefficients and determination of optimal voltage vector and current control. The application uses rotor angle and current switching state of a two-level inverter to calculate current difference between adjacent sampling points, and uses estimation error of the current difference to online self-adaptively update observer gain parameters; the observer gain is online updated based on the estimation error. Due to the filtering effect and online adaptability of the observer, the control method does not use motor parameters for current difference prediction, and the observer parameters have filtering characteristics and are robust to measurement noise and motor parameter changes.

Description

Technical Field

[0001] This invention belongs to the field of motor motion control technology, specifically relating to a robust predictive current control method for a model-free permanent magnet synchronous motor based on an observer. Background Technology

[0002] Permanent magnet synchronous motors (PMSMs) are widely used in various industrial fields due to their high power density and good efficiency. In the field of motor drives, model predictive control (MMC) has received extensive attention. However, traditional MMC current control is sensitive to motor parameter mismatch, and the prediction of relevant variables relies on an accurate system model, resulting in relatively poor parameter robustness. To address this issue, various model-parameter-free predictive current control methods have been proposed. These methods typically rely on measured current to update current predictions. For stable operation and satisfactory performance, immunity to induced noise is crucial; however, current measurements always contain noise due to electromagnetic interference caused by high-speed switching, especially for inverter-powered motor drives. Therefore, it is necessary to design an observer-based model-parameter-free robust predictive current control method for PMSMs. Summary of the Invention

[0003] (a) Technical problems to be solved This invention proposes a robust predictive current control method for model-parameterless permanent magnet synchronous motors based on observers, in order to solve the technical problem of traditional model predictive current control being sensitive to motor parameter mismatch and improve the robustness of model predictive current control methods to measurement noise and changes in motor parameters.

[0004] (II) Technical Solution To address the aforementioned technical problems, this invention proposes a robust predictive current control method for a model-free permanent magnet synchronous motor based on an observer. This control method includes the following steps: S1. Design a current difference observer The current difference prediction equation for the observer is designed as follows: in, For two adjacent sampling points k and k The stator current vector difference between -1; Sampling points k The stator current vector; and These are the two gain coefficients of the observer; Sampling points k -1 represents the motor rotor position; Sampling points k The stator voltage vector of -1 is expressed as: in, Indicates the angular displacement between the three phases; Indicates sampling point k -1 abc The switching state of the three phases; The equation for the current difference observer is as follows: in, For the current difference observer gain, variable Indicates an estimated value; S2. Determine the current difference observer coefficients Two gain coefficients are determined using an online adaptive update method. and The estimated value; S3. Determine the optimal voltage vector and current control. Assumption It remains constant over the next two sampling periods; Based on the following formula, calculate k Reference current value at +2 : in, and These are the d-axis and q-axis current references defined in the rotor flux orientation reference frame, respectively. The sampling period; Based on the current difference observer equation, the following formula is used to calculate... k Stator current at +1 time: Based on the following formula, the first voltage vector is calculated for all candidate voltage vectors. k Stator current at +2: in, Represents the candidate voltage vector; Based on the following formula, the loss function for each candidate voltage vector is calculated: The candidate voltage vector corresponding to the minimum value of the loss function is used as the optimal voltage vector, and robust predictive current control of the permanent magnet synchronous motor is achieved by using the optimal voltage vector.

[0005] Furthermore, in step S1, for a two-level inverter, there are eight different switching states, generating eight different voltage vectors.

[0006] Further, in step S3, the first voltage vector is calculated for all candidate voltage vectors based on the following formula. k Stator current at +2: in, This represents eight candidate voltage vectors.

[0007] Furthermore, in step S2, an online adaptive update method is used to determine the two gain coefficients. and The estimated value is shown in the following formula: in, This represents the estimation error of the current difference; , , which is the integral gain.

[0008] Furthermore, in step S3, the current reference... It is generated by the speed outer loop PI controller.

[0009] (III) Beneficial Effects This invention proposes a robust predictive current control method for a model-free permanent magnet synchronous motor based on an observer. The main components include designing a current difference observer, determining the observer coefficients, and determining the optimal voltage vector and current control. This invention calculates the current difference between adjacent sampling points using the rotor angle and the current switching state of the two-level inverter. The observer gain parameters are then updated online adaptively based on the estimation error of the current difference. Due to the observer's filtering effect and online adaptability, the control method of this invention does not use motor parameters for current difference prediction. The observer parameters possess filtering characteristics, making it robust to measurement noise and changes in motor parameters. Attached Figure Description

[0010] Figure 1 The simulation results are for the method of the present invention when the inductance suddenly changes to 150% of the true value under rated load and speed. Figure 2 The simulation results show the starting process of the motor from standstill to rated speed. Figure 3 shows a comparison of simulation results under parameter mismatch; where Figure 3a For traditional model predictive control, , Figure 3b The controller proposed in this invention, ; Figure 4 The results of the experiment show the starting process of the motor from standstill to rated speed; Figure 5The results are from experiments involving sudden loading and sudden unloading. Detailed Implementation

[0011] To make the objectives, contents, and advantages of the present invention clearer, the specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples.

[0012] This embodiment proposes a robust predictive current control method for a model-free permanent magnet synchronous motor based on an observer. The method specifically includes the following steps: S1. Design a current difference observer The current difference prediction equation for the observer is designed as follows: in, For two adjacent sampling points k and k The stator current vector difference between -1; Sampling points k The stator current vector; and These are the two gain coefficients of the observer; Sampling points k -1 represents the motor rotor position; Sampling points k The stator voltage vector of -1 is expressed as: in, Indicates the angular displacement between the three phases; Indicates sampling point k -1 abc For a two-level inverter, there are eight different switching states for three-phase switching, resulting in eight different voltage vectors.

[0013] The equation for the current difference observer is as follows: in, For the current difference observer gain, variable This represents an estimated value.

[0014] S2. Determine the current difference observer coefficients For model-parameter-free current prediction, the two gain coefficients of the current difference observer... and This is an unknown quantity. An online adaptive update method is used to determine it. and The estimated value is shown in the following formula: in, This represents the estimation error of the current difference; , , which is the integral gain.

[0015] S3. Determine the optimal voltage vector and current control. Derivation k Reference current and stator current at +2: Current reference This is generated by the speed outer loop PI controller. Considering that the speed loop bandwidth is typically much lower than the sampling rate, it is assumed... It remains constant over the next two sampling periods.

[0016] Based on the following formula, calculate k Reference current value at +2 : in, and These are the d-axis and q-axis current references defined in the rotor flux orientation reference frame, respectively. The sampling period.

[0017] Based on the current difference observer equation, the following formula is used to calculate... k Stator current at +1 time: Based on the following formula, the first voltage vector is calculated for all candidate voltage vectors. k Stator current at +2: in, This represents eight candidate voltage vectors.

[0018] Based on the following formula, the loss function for each of the eight candidate voltage vectors is calculated: The candidate voltage vector corresponding to the minimum value of the loss function is used as the optimal voltage vector, and robust predictive current control of the permanent magnet synchronous motor is achieved by using the optimal voltage vector.

[0019] In the simulation test, the dynamic response of the robust predictive current control method for model-parameterless permanent magnet synchronous motor based on observer and the traditional model predictive control method under motor parameter mismatch was compared.

[0020] Figure 1The figure shows the simulation results of the control method of this invention under rated load and speed when the inductance suddenly changes to 150% of the nominal value. As can be seen from the figure, when the motor inductance value changes abruptly, the d-axis and q-axis currents quickly recover to their tracking values, and the online adaptive update method of the observer parameters can rapidly update the two gain coefficients. and The value of is adjusted to match the actual motor parameters. Simulation results show that when the actual motor parameters deviate from their nominal values, the control method of the present invention can maintain accurate tracking control.

[0021] Figure 2 This figure shows the simulation results of the motor's startup process from standstill to rated speed. As can be seen from the figure, the motor speed accelerates smoothly to the rated speed. During startup, the d-axis and q-axis currents exhibit good tracking performance to the reference current, and the observer's two gain coefficients... and It can converge quickly without current surges or oscillations; simulation results show that the control method of the present invention can achieve fast and safe motor start-up without prior knowledge of motor model parameters.

[0022] Figure 3a and Figure 3b The figures show the simulation results of the traditional model predictive control algorithm and the control method of this invention under parameter mismatch. As can be seen from the figures, the traditional model predictive control algorithm exhibits significant current tracking error when motor parameters are mismatched, while the control method of this invention does not have current tracking error and demonstrates good steady-state performance.

[0023] To further verify the control performance of the control method of this invention, experimental tests were conducted based on a 2.4kW surface-mount permanent magnet synchronous motor speed control platform. The stator three-phase current was directly measured by a current probe, while the motor speed, d-axis and q-axis currents, and other internal variable waveforms were obtained through an onboard digital-to-analog converter chip. To confirm the superior control performance of the control method of this invention, the dynamic and static responses of a traditional model predictive control method were compared under the same conditions.

[0024] Figure 4 The figure shows the experimental test waveform of the motor starting from a standstill to its rated speed. As can be seen from the figure, similar to the simulation results, using the control method of this invention, the motor speed is smoothly accelerated to the rated speed. There is no current surge or oscillation during the acceleration process, and the motor is safely and quickly started without any motor parameters.

[0025] Figure 5 The figures show experimental test waveforms for sudden loading and sudden unloading. As can be seen from the figures, using the control method of this invention, the motor speed deviates from the reference value instantaneously during sudden loading and sudden unloading, and then quickly recovers to the rated speed, indicating that the control method of this invention has good robustness to load changes.

[0026] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A robust predictive current control method for a model-free permanent magnet synchronous motor based on an observer, characterized in that, The control method includes the following steps: S1. Design a current difference observer The current difference prediction equation for the observer is designed as follows: in, For two adjacent sampling points k and k The stator current vector difference between -1; Sampling points k The stator current vector; and These are the two gain coefficients of the observer; Sampling points k -1 represents the motor rotor position; Sampling points k The stator voltage vector of -1 is expressed as: in, Indicates the angular displacement between the three phases; Indicates sampling point k -1 abc The switching state of the three phases; The equation for the current difference observer is as follows: in, For the current difference observer gain, variable Indicates an estimated value; S2. Determine the current difference observer coefficients Two gain coefficients are determined using an online adaptive update method. and The estimated value; S3. Determine the optimal voltage vector and current control. Assumption It remains constant over the next two sampling periods; Based on the following formula, calculate k Reference current value at +2 : in, and These are the d-axis and q-axis current references defined in the rotor flux orientation reference frame, respectively. The sampling period; Based on the current difference observer equation, the following formula is used to calculate... k Stator current at +1 time: Based on the following formula, the first voltage vector is calculated for all candidate voltage vectors. k Stator current at +2: in, Represents the candidate voltage vector; Based on the following formula, the loss function for each candidate voltage vector is calculated: The candidate voltage vector corresponding to the minimum value of the loss function is used as the optimal voltage vector, and robust predictive current control of the permanent magnet synchronous motor is achieved by using the optimal voltage vector.

2. The robust predictive current control method for a model-free permanent magnet synchronous motor based on an observer, as described in claim 1, is characterized in that... In step S1, for a two-level inverter, there are eight different switching states, resulting in eight different voltage vectors.

3. The robust predictive current control method for a model-free permanent magnet synchronous motor based on an observer, as described in claim 2, is characterized in that... In step S3, the first voltage vector is calculated for all candidate voltage vectors based on the following formula. k Stator current at +2: in, This represents eight candidate voltage vectors.

4. The robust predictive current control method for a model-free permanent magnet synchronous motor based on an observer, as described in claim 1, is characterized in that... In step S2, an online adaptive update method is used to determine the two gain coefficients. and The estimated value is shown in the following formula: in, This represents the estimation error of the current difference; , , which is the integral gain.

5. The robust predictive current control method for a model-free permanent magnet synchronous motor based on an observer, as described in claim 1, is characterized in that... In step S3, the current reference It is generated by the speed outer loop PI controller.

Images