CN-121979104-A - Circular induction synchronizer angle error compensation method based on convolution feature modeling

CN121979104ACN 121979104 ACN121979104 ACN 121979104ACN-121979104-A

Abstract

The application discloses a convolution feature modeling-based angle error compensation method of a circular induction synchronizer, which relates to the technical field of angle measurement and servo control, and comprises the steps of constructing an angle error sequence of the circular induction synchronizer by calculating angle errors of angle measurement data and reference angle values, carrying out sliding window segmentation processing on the angle error sequence to obtain error data fragments containing continuous angle interval information, carrying out feature extraction processing on the error data fragments to obtain low-frequency change features and high-frequency oscillation features in the angle errors, generating angle error compensation values corresponding to the error data fragments through nonlinear mapping based on the two features, compensating the angle measurement results of the circular induction synchronizer according to the angle error compensation values, and outputting the compensated angle values. The method is used for uniformly modeling the angle error characteristics, realizing the cooperative compensation of zero errors and subdivision errors, and improving the precision and consistency of the angle measurement of the circular induction synchronizer.

Inventors

WANG JIANYUAN
WANG JIAN
YANG JINGYUAN
CHEN JINBAO
Shao Yuankui
Luo Yuqiao

Assignees

安徽空天装备科技有限公司
南京航空航天大学

Dates

Publication Date: 20260505
Application Date: 20260205

Claims (7)

1. The angle error compensation method of the circular induction synchronizer based on convolution feature modeling is characterized by comprising the following steps of: S1, acquiring angle measurement data of a circular induction synchronizer in a rotating process, and synchronously acquiring a reference angle value corresponding to the angle measurement data; S2, calculating an angle error between the angle measurement data and a reference angle value so as to construct an angle error sequence of the circular induction synchronizer in the whole rotation period; s3, carrying out sliding window segmentation processing on the angle error sequence to obtain an error data segment containing continuous angle interval information; S4, performing feature extraction processing on the error data segment to obtain a low-frequency change feature and a high-frequency oscillation feature in the angle error; S5, generating an angle error compensation value corresponding to the error data segment through nonlinear mapping based on the feature extraction result ; And S6, compensating the angle measurement result of the circular induction synchronizer according to the angle error compensation value, and outputting the compensated angle value.
2. The method for compensating angle error of a circular induction synchronizer based on convolution feature modeling according to claim 1, wherein the angle error is defined as a difference relation between a measured angle and a reference angle, and the expression is: ; Wherein, the Representing the measured angle value output by the circular induction synchronizer, Representing the value of the corresponding reference angle, Indicating an angle error.
3. The method for compensating angle errors of a circular induction synchronizer based on convolution feature modeling as claimed in claim 2, wherein the first of the error data segments is The individual error data segments can be expressed as: ; Wherein, the Indicating the length of the sliding window, Represent the first The error data segments.
4. The method for compensating angle errors of a circular induction synchronizer based on convolution feature modeling according to claim 3, wherein in S4, feature extraction processing is performed on the error data segment based on a convolution feature modeling mode, and a feature extraction process is expressed as: ; wherein w represents the convolution kernel parameters, A convolution operation is represented and is performed, Representing the non-linear mapping function, Representing the extracted error feature vector.
5. The method for compensating angle errors of a circular induction synchronizer based on convolution feature modeling according to claim 4, wherein the expression of the compensated angle value is: ; Wherein, the Representing the predicted angle error compensation value, Representing the compensated angular output.
6. The method for compensating the angle error of the circular induction synchronizer based on the convolution feature modeling according to claim 5 is characterized in that the method for compensating the angle error of the circular induction synchronizer based on the convolution feature modeling is provided with a permanent magnet synchronous motor driving device based on STM32 and FPGA, and comprises an STM32 series microcontroller, an Altera series FPGA, a grid driving plate and an induction synchronizer; the STM32 microcontroller and the FPGA are provided with an FSMC bus interface; the STM32 microcontroller is responsible for a position PI and speed PI control method and an M/T speed observation method; The FPGA utilizes an internal RAM to realize an FSMC interface, writes a PWM wave with a dead zone output by a current loop IP core by using Verilog language, receives a feedback value from an induction synchronizer, converts the feedback value into a mechanical angle position value, and converts the obtained mechanical angle into an electrical angle value required by the current loop SVPWM; the grid driving plate converts PWM signals into three-phase inversion signals through a grid driving chip to drive a motor, and phase currents are collected to provide input for the FPGA current loop module; The sensing synchronizer is used as a position feedback module to provide accurate position feedback values.
7. A convolution feature modeling-based angle error compensation system for a circular induction synchronizer, configured to implement the convolution feature modeling-based angle error compensation method for a circular induction synchronizer according to claims 1-5, comprising: the data acquisition module is used for acquiring angle measurement data of the circular induction synchronizer; The reference angle measurement unit is used for acquiring a high-precision reference angle true value of the circular induction synchronizer in the calibration or test process, performing time synchronization and angle alignment with the measured data of the synchronizer, and providing a reference angle sequence for the error construction module; The error construction module is used for constructing an angle error sequence; The sliding window segmentation module is used for carrying out sliding window segmentation and sampling processing on the angle error sequence obtained by construction, generating continuous subsequence samples according to the set window length and step length, and completing normalization/formatting to be used as the input of the convolution characteristic modeling module The convolution feature modeling module is used for extracting features of the angle error sequence based on a convolution feature modeling mode; The compensation output module is used for nonlinear mapping to generate an angle error compensation value; And the compensation output module is used for outputting a compensated angle result according to the angle error compensation value.

Description

Circular induction synchronizer angle error compensation method based on convolution feature modeling Technical Field The invention relates to the technical field of angle measurement and servo control, in particular to a method for compensating angle errors of a circular induction synchronizer based on convolution feature modeling. Background The circular induction synchronizer has the advantages of simple structure, strong environmental adaptability, strong electromagnetic interference resistance and the like, and is widely applied to space servo systems, solar wing driving mechanisms and high-precision angle measurement occasions. However, in practical applications, the angular output accuracy of the circular induction synchronizer is susceptible to manufacturing errors, assembly deviations, and demodulation circuit inconsistencies, producing zero errors, subdivision errors, and periodic nonlinear errors due to spatial harmonics. In the prior art, harmonic fitting, table lookup correction or compensation method based on analytical model is mostly adopted to correct the errors, but the methods depend on priori assumption of an error model, so that both low-frequency system errors and high-frequency subdivision errors are difficult to be considered, adaptability to change of adjustment conditions is poor, and further improvement of measurement accuracy of the circular induction synchronizer is limited. Therefore, a technical solution capable of uniformly modeling and compensating the angle error of the circular induction synchronizer is needed to improve the angle measurement accuracy and the system stability. Disclosure of Invention In order to solve the problems, the application provides a technical scheme for angle error compensation of a circular induction synchronizer based on convolution feature modeling, which is realized by a method flow and a system structure together, and integrally comprises links of data acquisition, error construction, sliding window segmentation, convolution feature modeling, compensation value generation, compensation output and the like, and can be realized by a processor executing a program stored in a memory, wherein the method comprises the following specific contents: S1, acquiring angle measurement data of a circular induction synchronizer in a rotating process, and synchronously acquiring a reference angle value corresponding to the angle measurement data; S2, calculating an angle error between the angle measurement data and a reference angle value so as to construct an angle error sequence of the circular induction synchronizer in the whole rotation period; s3, carrying out sliding window segmentation processing on the angle error sequence to obtain an error data segment containing continuous angle interval information; S4, performing feature extraction processing on the error data segment to obtain a low-frequency change feature reflecting systematic deviation and a high-frequency oscillation feature reflecting periodic disturbance; S5, generating an angle error compensation value corresponding to the error data segment through nonlinear mapping based on the feature extraction result ; And S6, compensating the angle measurement result of the circular induction synchronizer according to the angle error compensation value, and outputting the compensated angle value. Preferably, the angle error is defined as the difference between the measured angle and the reference angle, expressed as: ; Wherein, the Representing the measured angle value output by the circular induction synchronizer,Representing the value of the corresponding reference angle,Indicating an angle error. Preferably, the first of the error data segmentsThe individual error data segments can be expressed as: ; Wherein, the Indicating the length of the sliding window,Represent the firstThe error data segments. Preferably, in S4, feature extraction processing is performed on the error data segment based on a convolution feature modeling manner, where the feature extraction process is expressed as: ; wherein w represents the convolution kernel parameters, A convolution operation is represented and is performed,Representing the non-linear mapping function,Representing the extracted error feature vector. Preferably, the expression of the compensated angle value is: ; Wherein, the Representing the predicted angle error compensation value,Representing the compensated angular output. Preferably, the angle error compensation method of the circular induction synchronizer based on convolution feature modeling is provided with a permanent magnet synchronous motor driving device based on STM32 and FPGA, and comprises an STM32 series microcontroller, an Altera series FPGA, a grid driving plate and an induction synchronizer; the STM32 microcontroller and the FPGA are provided with an FSMC bus interface; the STM32 microcontroller is responsible for a position PI and speed PI control method and an M/T speed observation