Search

CN-122026761-A - High-precision position control method for high-overload permanent magnet synchronous motor of cooperative robot

CN122026761ACN 122026761 ACN122026761 ACN 122026761ACN-122026761-A

Abstract

The invention discloses a high-precision position control method and a high-precision position control system for a high-overload permanent magnet synchronous motor of a cooperative robot, and belongs to the technical field of motor control. The method comprises the steps of establishing a high-precision permanent magnet synchronous motor model considering eddy current effect and inductance variation characteristics to represent magnetic linkage phase lag and dynamic coupling characteristics under high overload working conditions, designing a double-power-law expansion state observer to perform online estimation and feedforward compensation on total disturbance, combining a three-section saturated approach-law sliding mode controller to achieve position high-precision tracking control under sudden load, further taking controller parameters as input variables, constructing a nonlinear mapping relation between control parameters, recovery time and maximum position tracking error, and performing multi-objective optimization based on a hiking optimization algorithm to obtain global optimal control parameters.

Inventors

  • ZHAO JIWEN
  • LI MEN

Assignees

  • 合肥工业大学

Dates

Publication Date
20260512
Application Date
20260410

Claims (7)

  1. 1. The high-precision position control method for the high-overload permanent magnet synchronous motor of the cooperative robot is characterized by comprising the following steps of: S100, establishing a high-precision permanent magnet synchronous motor model considering the eddy current effect and the inductance change characteristic, wherein the magnetic linkage phase lag and the dynamic coupling effect caused by the eddy current are represented based on a vector magnetic circuit model, and the d-axis dynamic inductance and the q-axis dynamic inductance are obtained by adopting a frozen permeability method; s200, designing a double power law extended state observer to perform total disturbance estimation and feedforward compensation, and combining a sliding mode controller of three sections of saturated approach laws to realize high-precision position tracking control under sudden load; s300, taking a controller parameter as an input variable, applying sudden load disturbance under a sinusoidal position tracking working condition to perform system simulation, extracting recovery time and maximum position tracking error as performance evaluation indexes, and generating a multidimensional input and output data sample set; s400, constructing a proxy model by utilizing a long-term memory network and a gradient lifting tree regressive to learn a nonlinear mapping relation between controller parameters and the performance evaluation indexes; S500, performing multi-objective optimization on the controller parameters based on a hiking optimization algorithm, and simultaneously minimizing recovery time and maximum position tracking error on the premise of meeting closed-loop stability and physical constraint to obtain global optimal control parameters; the obtained global optimal control parameters are used for implementing high-precision position control on the permanent magnet synchronous motor.
  2. 2. The control method according to claim 1, wherein in the step S100, the total disturbance term includes motor parameter disturbance, friction disturbance, external load disturbance and unmodeled dynamics, wherein the actual output characteristic and the nominal model output result are compared and analyzed by collecting experimental data of the motor under the real working condition, and the value ranges or equivalent values of the uncertainty factors Δe, Δl and Δb are obtained by identification or fitting.
  3. 3. The control method according to claim 1, wherein in step S100, the high-precision permanent magnet synchronous motor model taking into account eddy current effect and inductance variation characteristics includes: establishing an equivalent closed coil model based on an internal dominant eddy current loop of the conductive magnetic material to obtain reverse magnetomotive force caused by eddy current; constructing a vector magnetic circuit model under a three-phase static coordinate system, and obtaining d-axis and q-axis magnetic circuit equations through Park transformation; calculating flux linkage distribution according to finite element simulation results under given current excitation conditions, and obtaining d-axis and q-axis dynamic inductances according to the ratio of flux linkage to current; substituting the dynamic inductance into a voltage equation to obtain the voltage equation considering the eddy current effect.
  4. 4. The control method as claimed in claim 1, wherein in step S200, the double power law extended state observer uses the mechanical angular velocity of the motor and the total disturbance term as the observation object, and an observer error correction law including a linear feedback term and different order power terms is constructed after the observation error is defined, so as to enhance the correction capability of a small error region while ensuring stable convergence of the error, and reduce high frequency jitter caused by noise sensitivity.
  5. 5. The control method according to claim 1, wherein in step S200, the three-segment saturated approach law sliding mode controller defines a sliding mode surface, and introduces a saturation switching term, a linear damping term and a state-dependent gain term into a velocity sliding mode control law, so that the sliding mode surface converges to exhibit the piecewise characteristics of rapid arrival of a large error region, stable convergence of a medium error region and smooth vibration suppression of a small error region.
  6. 6. The control method according to claim 1, wherein in step S400, the long-short term memory network performs serialization encoding on the normalized controller parameter vector to output a feature vector, the gradient-lifting tree regressor performs joint estimation on the recovery time and the maximum position tracking error based on the feature vector, and the gradient-lifting tree regressor is composed of a plurality of regression decision trees, and outputs the accumulated value of the output result of each regression decision tree.
  7. 7. A high-precision position control system for a high-overload permanent magnet synchronous motor of a cooperative robot, characterized in that the control system comprises a memory, a processor and a control program stored on the memory and capable of running on the processor, and the control program realizes the steps of the high-precision position control method for the high-overload permanent magnet synchronous motor of the cooperative robot according to any one of claims 1 to 6 when the control program is executed by the processor.

Description

High-precision position control method for high-overload permanent magnet synchronous motor of cooperative robot Technical Field The invention belongs to the technical field of motor control, and particularly relates to a high-precision position control method for a high-overload permanent magnet synchronous motor of a cooperative robot. Background The permanent magnet synchronous motor has the advantages of high power density, high response speed, high control precision and the like, and is widely applied to a servo driving system of a cooperative robot. In the prior art, a vector control method based on a d-q axis model is mostly adopted, and the robustness of the system is improved by combining a disturbance observer or a sliding mode controller. However, for a high overload permanent magnet synchronous motor, due to remarkable eddy current effect and nonlinear inductance parameter variation along with working conditions, the traditional simplified model is difficult to accurately represent actual dynamic characteristics, and position errors are increased, recovery speed is slow and control performance is reduced under the action of sudden load. Therefore, a high-precision position control scheme facing high overload conditions is needed. Referring to the related published technology, the technical scheme with the publication number CN111130410A obtains high-frequency voltage through sine waves with random periods, samples the high-frequency voltage to obtain high-frequency current, and obtains the rotor angular speed of the permanent magnet synchronous motor according to the high-frequency current to obtain the rotor position of the permanent magnet synchronous motor. The technical scheme disclosed in the publication number WO2021082476A1 obtains the stator resistance through the motor stator temperature by using an online lookup table, obtains the stator inductance through the motor stator temperature T, the current amplitude and the current vector angle by using an online lookup table, calculates the flux linkage value in real time by using a flux linkage observation model, thereby improving the accuracy of motor control and decoupling, redistributes the given stator current by using the output result of a torque closed loop, maintains the permanent magnet synchronous motor to operate according to a better control track, and reduces the heating and the loss of the motor. The technical scheme with the publication number of US08786221B2 carries out compensation calculation on the rotating speed based on the absolute angular position, the rotating speed, the torque command and the battery voltage of the motor, generates dq axis current and voltage command according to the rotating speed, and converts the dq axis current and the voltage command into three-phase voltage command to realize closed-loop driving control of the permanent magnet synchronous motor. The above technical solutions all provide a plurality of control solutions for implementing the permanent magnet synchronous motor, but for the control of the permanent magnet motor applied to the precise action of the cooperative robot, a more optimized solution is still needed. The foregoing discussion of the background art is intended to facilitate an understanding of the present invention only. This discussion is not an admission or admission that any of the material referred to was common general knowledge. Disclosure of Invention The invention aims to provide a high-precision position control method and a high-precision position control system for a high-overload permanent magnet synchronous motor of a cooperative robot, wherein the control method comprises the steps of establishing a high-precision permanent magnet synchronous motor model considering eddy current effect and inductance variation characteristics so as to represent magnetic linkage phase lag and dynamic coupling characteristics under the high-overload working condition; the method comprises the steps of designing a double-power-law expanded state observer to perform online estimation and feedforward compensation on total disturbance, combining a three-section saturated approach law sliding mode controller to realize position high-precision tracking control under sudden load, further taking controller parameters as input variables, constructing a nonlinear mapping relation between control parameters, recovery time and maximum position tracking error, and performing multi-objective optimization based on a hiking optimization algorithm to obtain global optimal control parameters. The invention adopts the following technical scheme that the high-precision position control method of the high-overload permanent magnet synchronous motor of the cooperative robot is shown in the attached figure 1, and comprises the following steps of: S100, a high-precision permanent magnet synchronous motor model considering the eddy current effect and the inductance change characteristic is established, w