Circular inductosyn angle error compensation method based on convolution feature modeling

Abstract

The invention discloses a circular inductosyn angle error compensation method based on convolution feature modeling, and relates to the technical field of angle measurement and servo control, an angle error sequence of a circular inductosyn is constructed by calculating an angle error between angle measurement data and a reference angle value, sliding window segmentation processing is performed on the angle error sequence, and a circular inductosyn angle error compensation result is obtained. The method comprises the following steps: obtaining an error data fragment containing continuous angle interval information, carrying out feature extraction processing on the error data fragment to obtain a low-frequency change feature and a high-frequency oscillation feature in an angle error, and generating an angle error compensation value corresponding to the error data fragment through nonlinear mapping based on the two features, and compensating an angle measurement result of the circular inductosyn according to the angle error compensation value, and outputting a compensated angle value. According to the method, unified modeling is carried out on the angle error features, cooperative compensation of the zero error and the subdivision error is achieved, and the precision and consistency of angle measurement of the circular inductosyn are improved.

Description

Technical Field

[0001] This invention relates to the field of angle measurement and servo control technology, and in particular to a method for compensating for angle errors in a circular induction synchronizer based on convolution feature modeling. Background Technology

[0002] Circular inductive synchros are widely used in aerospace servo systems, solar array drive mechanisms, and high-precision angle measurement applications due to their advantages such as simple structure, strong environmental adaptability, and strong resistance to electromagnetic interference. However, in practical applications, the angle output accuracy of circular inductive synchros is easily affected by manufacturing errors, assembly deviations, and inconsistencies in demodulation circuits, resulting in zero-position errors, subdivision errors, and periodic nonlinear errors caused by spatial harmonics. Existing technologies often employ harmonic fitting, lookup table correction, or compensation methods based on analytical models to correct these errors. However, these methods rely on prior assumptions of the error model, making it difficult to simultaneously account for low-frequency system errors and high-frequency subdivision errors. Furthermore, they have poor adaptability to changes in assembly and adjustment conditions, limiting further improvements in the measurement accuracy of circular inductive synchros.

[0003] Therefore, there is an urgent need for a technical solution that can uniformly model and compensate for the angle error of the circular induction synchronizer in order to improve the angle measurement accuracy and system stability. Summary of the Invention

[0004] To address the aforementioned issues, this application proposes a circular induction synchronizer angle error compensation technique based on convolutional feature modeling. This technique is implemented through a combination of method flow and system structure, encompassing data acquisition, error construction, sliding window segmentation, convolutional feature modeling, compensation value generation, and compensation output. The entire process can be executed by a processor using a program stored in memory. The specific details are as follows: S1. Obtain angle measurement data of the circular induction synchronizer during the rotation process, and synchronously obtain the reference angle value corresponding to the angle measurement data; S2. Calculate the angle error between the angle measurement data and the reference angle value, and then construct the angle error sequence of the circular induction synchronizer throughout the entire rotation cycle; S3. Perform sliding window segmentation on the angle error sequence to obtain error data segments containing continuous angle interval information; S4. Perform feature extraction processing on the error data segments to obtain low-frequency variation features reflecting systematic deviations and high-frequency oscillation features reflecting periodic disturbances; S5. Based on the feature extraction results, generate an angle error compensation value corresponding to the error data segment through nonlinear mapping. ; S6. Compensate the angle measurement results of the circular induction synchronizer according to the angle error compensation value, and output the compensated angle value.

[0005] Preferably, the angle error is defined as the difference between the measured angle and the reference angle, expressed as: ; in, This indicates the measured angle value output by the circular induction synchronizer. This indicates the corresponding reference angle value. This indicates the angular error.

[0006] Preferably, the first error data segment Each error data segment can be represented as: ; in, Indicates the length of the sliding window. Indicates the first A data segment with error.

[0007] Preferably, in S4, feature extraction processing is performed on the error data segment based on convolutional feature modeling. The feature extraction process is expressed as follows: ; Where w represents the convolution kernel parameter, This represents the convolution operation. Represents a nonlinear mapping function. This represents the extracted error feature vector.

[0008] Preferably, the expression for the compensated angle value is: ; in, This represents the predicted angle error compensation value. This indicates the angle output after compensation.

[0009] Preferably, the circular induction synchronizer angle error compensation method based on convolution feature modeling is configured with a permanent magnet synchronous motor drive device based on STM32 and FPGA, including an STM32 series microcontroller, an Altera series FPGA, a gate driver board, and an induction synchronizer. The STM32 microcontroller and FPGA are equipped with an FSMC bus interface; The STM32 microcontroller is responsible for the position PI and velocity PI control methods, as well as the M / T velocity observation method. The FPGA uses internal RAM to implement the FSMC interface and uses Verilog language to write the current loop IP core to output a PWM wave with dead time. It receives feedback values ​​from the inductive synchronizer, converts them into mechanical angle position values, and converts the obtained mechanical angles into electrical angle values ​​required by the current loop SVPWM. The gate driver board converts the PWM signal into a three-phase inverter signal to drive the motor through the gate driver chip, and collects the phase current to provide input for the FPGA current loop module; The inductive synchronizer serves as a position feedback module, providing accurate position feedback values.

[0010] An angle error compensation system for a circular sensor synchronizer based on convolutional feature modeling is provided to implement an angle error compensation method for a circular sensor synchronizer based on convolutional feature modeling, comprising: Data acquisition module: used to acquire angle measurement data from the circular induction synchronizer; Reference angle measurement unit: used to obtain the high-precision reference angle true value of the circular induction synchronizer during the calibration / testing process, and to synchronize and align the time and angle with the synchronizer measurement data, providing a reference angle sequence for the error construction module; Error construction module: used to construct the angle error sequence; Sliding window segmentation module: This module is used to segment and sample the constructed angle error sequence using a sliding window. It generates continuous subsequence samples according to the set window length and stride, and performs normalization / formatting to serve as input for the convolutional feature modeling module. Convolutional Feature Modeling Module: Used to extract features from angle error sequences based on convolutional feature modeling. Compensation output module: used for nonlinear mapping to generate angle error compensation values; Compensation output module: Used to output the compensated angle result based on the angle error compensation value.

[0011] In summary, the circular sensor synchronizer angle error compensation method based on convolutional feature modeling of the present invention has the following advantages compared with traditional technologies: 1. By using convolutional feature modeling, we can effectively represent the multi-scale features in the angle error of the circular induction synchronizer, and simultaneously compensate for the zero-position error and the subdivision error. 2. Reduce reliance on fixed analytical error models and improve the adaptability of error compensation methods to assembly deviations, winding inconsistencies, and non-idealities in demodulation circuits; 3. Effectively suppresses periodic errors and high-frequency oscillation errors in angle measurement results, significantly improving angle measurement accuracy and output consistency; 4. No changes are needed to the mechanical structure and signal demodulation circuit of the circular induction synchro; error compensation can be achieved solely through data processing, resulting in low engineering implementation costs and easy system integration. 5. The compensated angle output can be directly used as the feedback signal of the servo control system, which helps to improve the system's operational stability and control accuracy, and has good engineering application value.

[0012] The technical method of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0013] Figure 1 The flowchart shows the angle error compensation method for a circular induction synchronizer based on convolution feature modeling.

[0014] Figure 2 A schematic diagram of the sliding window segmentation of the angle error sequence; Figure 3 A schematic diagram of the functional modules for modeling convolutional features and generating compensation values; Figure 4 Network structure diagram of functional modules for modeling convolutional features and generating compensation values; Figure 5 This is a diagram illustrating the compensation output process; Figure 6 A block diagram of an angle error compensation system for a circular sensor synchronizer based on convolutional feature modeling; Figure 7 A comparison chart of the circular induction synchronizer error before and after compensation; Figure 8 A comparison chart of the full-cycle positions before and after compensation; Figure 9 Enlarged view of the acceleration segment A before and after compensation; Figure 10 A magnified view of position B during the full-cycle acceleration phase before and after compensation. Detailed Implementation

[0015] The technical method of the present invention will be further described below with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of this application.

[0016] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the scope of this application and its application or use.

[0017] Techniques, systems, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, they should be considered part of the instruction manual.

[0018] In all the examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0019] Unless otherwise defined, the technical or scientific terms used in this invention shall have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0020] The purpose of this invention is to provide a method for modeling and compensating for the angle error of a circular induction synchronizer, such as... Figure 1 As shown, by uniformly modeling the angular error characteristics, coordinated compensation for zero-position error and subdivision error is achieved, thereby improving the accuracy and consistency of angle measurement by the circular induction synchronizer. Furthermore, this invention also provides a system and apparatus for implementing the above method, facilitating engineering applications in existing servo systems.

[0021] Example 1 In this embodiment, an experimental system for modeling and compensating for the angle error of a circular induction synchronizer was constructed.

[0022] The system includes: a synchronizer for the circle under test, a rotation drive unit, a reference angle measurement unit, a signal demodulation and data acquisition unit, and a data processing and error compensation unit.

[0023] Among them, the rotary drive unit is used to drive the circular induction synchronizer to rotate at a constant speed; The reference angle measurement unit is used to provide a true reference value for the rotation angle of the circular induction synchronizer; The signal demodulation and data acquisition unit is used to demodulate the output signal of the circular induction synchronizer and acquire angle measurement data; The data processing and error compensation unit is used to execute the angle error modeling and compensation process and output the compensation results.

[0024] In the data acquisition stage, angle measurement data of the circular induction synchronizer during the rotation process is collected, and reference angle values ​​corresponding to the angle measurement data are obtained synchronously.

[0025] Angle error sequence construction: During the rotation of the circular induction synchronizer, the signal demodulation and data acquisition unit collects the angle measurement data output by the circular induction synchronizer in real time, and simultaneously obtains the reference angle value output by the reference angle measurement unit.

[0026] By aligning the angle measurement data with the reference angle value and calculating the difference between them, the angle error sequence of the circular induction synchronizer over the entire rotation cycle is obtained. The angle error is: ; Sliding window segmentation: To effectively model local features in the angle error, this embodiment performs sliding window segmentation on the angle error sequence, such as... Figure 2 As shown.

[0027] A fixed-length sliding window is used to slide point-by-point across the angle error sequence, extracting a continuous error data segment at each slide. The k-th error data segment is: ; The sliding window length L is set to cover multiple electrical angle periods to ensure that the error data segment simultaneously contains low-frequency variation trends and high-frequency periodic oscillation components.

[0028] Convolutional Feature Modeling and Error Compensation Value Generation: After normalizing the error data segment, it is input into the feature modeling module. The feature modeling module extracts features from the error data segment based on the convolutional feature modeling method, such as... Figure 3 and Figure 4 As shown, the feature extraction process is as follows: ; in, For convolution kernel parameters, It is a nonlinear mapping function. This is the extracted error feature vector.

[0029] Based on the error feature vector, an angle error compensation value corresponding to the error data segment is generated through nonlinear mapping. .

[0030] Compensation Output and Application: The angle error compensation value is applied to the original angle measurement result to obtain the compensated angle output, such as... Figure 5 As shown, the compensation relationship is as follows: ; The compensated angle output can be used as the final angle measurement result of the circular induction synchronizer and directly used as the angle feedback signal of the servo control system.

[0031] Example 2 This embodiment provides a circular sensor synchronizer angle error compensation system based on convolutional feature modeling, such as... Figure 6 As shown, it includes: Data acquisition module: used to acquire angle measurement data from the circular induction synchronizer; Error construction module: used to construct the angle error sequence; Feature modeling module: used to extract features and perform nonlinear mapping on the angle error sequence based on convolutional feature modeling to generate angle error compensation values; Compensation output module: Used to output the compensated angle result based on the angle error compensation value.

[0032] The aforementioned functional modules can be implemented by hardware circuits, programmable logic devices, general-purpose processors, or combinations thereof, and the method flow of the present invention can be implemented by storing a program in a storage medium and executing it by a processor.

[0033] Example 3 This invention discloses a permanent magnet synchronous motor drive device based on STM32 and FPGA, including an STM32 series microcontroller, an Altera series FPGA, a gate driver board, and an inductive synchronizer. The STM32 microcontroller and FPGA are equipped with an FSMC bus interface; The STM32 microcontroller is responsible for the position PI and velocity PI control methods, as well as the M / T velocity observation method. The FPGA uses internal RAM to implement the FSMC interface and uses Verilog language to write the current loop IP core to output a PWM wave with dead time. It receives feedback values ​​from the inductive synchronizer, converts them into mechanical angle position values, and converts the obtained mechanical angles into electrical angle values ​​required by the current loop SVPWM. The gate driver board converts the PWM signal into a three-phase inverter signal to drive the motor through the gate driver chip, and collects the phase current to provide input for the FPGA current loop module; The inductive synchronizer serves as a position feedback module, providing accurate position feedback values.

[0034] The permanent magnet synchronous motor drive system based on STM32 and FPGA provided by this invention can accurately control the magnitude and direction of the magnetic field, so that the permanent magnet synchronous motor has stable torque, fast dynamic response and low latency.

[0035] Example 4 A comparative analysis of the angle errors before and after compensation was conducted, such as... Figure 7 , Figure 8 , Figure 9 and Figure 10 As shown, Figure 7 This indicates that the long-period zero-point error prediction is accurate before and after compensation. Figure 8 This indicates that the compensation effect on the random and system components of the full-cycle position values ​​before and after compensation is significant, and the position value feedback accuracy is improved. Figure 9 and Figure 10 The prediction of the full-cycle position value and the local amplification of the measured data show that the periodic component of the compensated angle error is significantly reduced, and the residual error exhibits random distribution characteristics. This verifies the effectiveness and engineering applicability of the method of the present invention in angle error compensation of circular induction synchronizers.

[0036] Finally, it should be noted that the above embodiments are only used to illustrate the technical methods of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical methods of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical methods to deviate from the spirit and scope of the technical methods of the present invention.

Claims

1. A method for compensating for angle errors in a circular induction synchronizer based on convolutional feature modeling, characterized in that, Includes the following steps: S1. Obtain angle measurement data of the circular induction synchronizer during the rotation process, and synchronously obtain the reference angle value corresponding to the angle measurement data; S2. Calculate the angle error between the angle measurement data and the reference angle value, and then construct the angle error sequence of the circular induction synchronizer throughout the entire rotation cycle; S3. Perform sliding window segmentation on the angle error sequence to obtain error data segments containing continuous angle interval information; S4. Perform feature extraction processing on the error data segments to obtain the low-frequency variation features and high-frequency oscillation features of the angle error; S5. Based on the feature extraction results, generate an angle error compensation value corresponding to the error data segment through nonlinear mapping. ; S6. Compensate the angle measurement results of the circular induction synchronizer according to the angle error compensation value, and output the compensated angle value.

2. The method for compensating for angle errors in a circular induction synchronizer based on convolutional feature modeling as described in claim 1, characterized in that, Angular error is defined as the difference between the measured angle and the reference angle, expressed as: ； in, This indicates the measured angle value output by the circular induction synchronizer. This indicates the corresponding reference angle value. This indicates the angular error.

3. The method for compensating for angle errors in a circular induction synchronizer based on convolutional feature modeling as described in claim 2, characterized in that, The first error data segment Each error data segment can be represented as: ； in, Indicates the length of the sliding window. Indicates the first A data segment with error.

4. The method for compensating for angle errors in a circular induction synchronizer based on convolutional feature modeling according to claim 3, characterized in that, In S4, feature extraction is performed on the error data segment based on convolutional feature modeling. The feature extraction process is as follows: ； Where w represents the convolution kernel parameter, This represents the convolution operation. Represents a nonlinear mapping function. This represents the extracted error feature vector.

5. The method for compensating for angle errors in a circular induction synchronizer based on convolutional feature modeling according to claim 4, characterized in that, The expression for the compensated angle value is: ； in, This represents the predicted angle error compensation value. This indicates the angle output after compensation.

6. The method for compensating for angle errors in a circular induction synchronizer based on convolutional feature modeling according to claim 5, characterized in that, The circular inductive synchronizer angle error compensation method based on convolution feature modeling is equipped with a permanent magnet synchronous motor drive device based on STM32 and FPGA, including an STM32 series microcontroller, an Altera series FPGA, a gate driver board, and an inductive synchronizer. The STM32 microcontroller and FPGA are equipped with an FSMC bus interface; The STM32 microcontroller is responsible for the position PI and velocity PI control methods, as well as the M / T velocity observation method. The FPGA uses internal RAM to implement the FSMC interface and uses Verilog language to write the current loop IP core to output a PWM wave with dead time. It receives feedback values ​​from the inductive synchronizer, converts them into mechanical angle position values, and converts the obtained mechanical angles into electrical angle values ​​required by the current loop SVPWM. The gate driver board converts the PWM signal into a three-phase inverter signal to drive the motor through the gate driver chip, and collects the phase current to provide input for the FPGA current loop module; The inductive synchronizer serves as a position feedback module, providing accurate position feedback values.

7. A circular induction synchronizer angle error compensation system based on convolutional feature modeling, used to implement the circular induction synchronizer angle error compensation method based on convolutional feature modeling as described in claims 1-5, characterized in that, include: Data acquisition module: used to acquire angle measurement data from the circular induction synchronizer; Reference angle measurement unit: used to obtain the high-precision reference angle true value of the circular induction synchronizer during calibration or testing, and to synchronize and align the time and angle with the synchronizer measurement data, providing a reference angle sequence for the error construction module; Error construction module: used to construct the angle error sequence; The sliding window segmentation module is used to segment and sample the constructed angle error sequence using a sliding window. It generates continuous subsequence samples according to the set window length and stride, and performs normalization / formatting to serve as input for the convolutional feature modeling module. Convolutional Feature Modeling Module: Used to extract features from angle error sequences based on convolutional feature modeling. Compensation output module: used for nonlinear mapping to generate angle error compensation values; Compensation output module: Used to output the compensated angle result based on the angle error compensation value.

Images