Search

CN-121977834-A - Detection method of low-speed servo bearing and on-machine self-checking system of servo bearing

CN121977834ACN 121977834 ACN121977834 ACN 121977834ACN-121977834-A

Abstract

The invention provides a detection method of a low-speed servo bearing and an on-machine self-checking system of the servo bearing, which are characterized in that a servo system is constructed, a system identification model is established based on a system identification means, a moment excitation condition is determined according to the using working condition of the bearing, and a control signal is applied Collecting and storing angular velocity data under corresponding excitation conditions Dui Processing, extracting time domain features and frequency domain features, judging whether the bearing meets the requirements of a servo system or not through simulation evaluation, experience model evaluation or quantitative index evaluation based on the extracted features, and storing the features and evaluation results into a data set for optimizing evaluation accuracy. According to the technical scheme, the on-machine detection of the bearing without disassembling the bearing can be realized by modifying a software code layer and modifying a small amount of hardware of the servo system with the servo bearing, and technical support is provided for maintenance self-checking of long-time operation of high-precision equipment.

Inventors

  • TU BIAO
  • Cui Qunling
  • ZHANG QIJIE
  • LI ZHIQIANG
  • ZHANG RUIQIAN
  • ZHANG ZHIJIE

Assignees

  • 华中光电技术研究所(中国船舶集团有限公司第七一七研究所)

Dates

Publication Date
20260505
Application Date
20251229

Claims (10)

  1. 1. A method of detecting a low speed servo bearing, the method comprising: S101, a signal acquisition step of constructing a servo system and establishing a system identification model based on a system identification means, wherein the transfer function of the system identification model is as follows: Where s is the Laplace operator, y(s) is the Laplace transformation for measuring angular velocity, u(s) is the Laplace transformation for control signal, b and a are transfer function coefficients, τ is delay time; Determining moment excitation condition according to bearing use condition, applying control signal The expression is as follows: Wherein i is the experimental order, y 1 rms is the root mean square value of an actual working condition angular velocity time domain curve y 1 (t), ω i is the excitation frequency, l i is the physical constraint limiting value, and m i is the offset; Collecting and storing angular velocity data under corresponding excitation conditions S102, signal processing step of Processing is carried out, and time domain features and frequency domain features are extracted; s103, data evaluation, namely judging whether the bearing meets the requirements of a servo system or not through simulation evaluation, empirical model evaluation or quantitative index evaluation based on the extracted characteristics; And S104, a data set updating step, namely saving the characteristics and the evaluation result into the data set for optimizing the evaluation accuracy.
  2. 2. The method for detecting a low-speed servo bearing according to claim 1, wherein the time domain feature extraction in the signal processing step comprises: Calculating intermediate variable z i (k): Wherein k and m represent k and m sampling values corresponding to sampling or time sequence, and Q is a filter length parameter of sliding window mean value filtering; Extracting time domain features ζ i : The rotation speed signal of the ith experiment of N i Is used for the number of sampling points of (a), Is an intermediate variable;
  3. 3. the method for detecting a low-speed servo bearing according to claim 1, wherein the frequency domain feature extraction in the signal processing step includes: For a pair of N i -point discrete Fourier transform is performed on the N i -point equal-period sampling sequence to obtain a discrete spectrum Y i (N) Wherein Y i (n) represents The conversion result of the N i point discrete Fourier transform of the equal period sampling sequence of N i points at the nth frequency point of the frequency domain is that Y i (N) is complex number, N is serial number, and the value range of N is an integer in the range of 0 to (N i -1); Extracting frequencies corresponding to the first lambda i maximum points And amplitude value And recording the amplitude at the nearest excitation frequency omega i as
  4. 4. The method for detecting a low-speed servo bearing according to claim 1, wherein the simulation evaluation in the data evaluation step includes: constructing a servo closed-loop simulation model based on the system identification model G(s); constructing a rate disturbance signal d (t): Wherein N is the total number of frequency points of the excitation signal applied in the first step, namely the total number of omega i and the total experimental times, and h i is the discrete Fourier transform distance of the actual working condition rotating speed time domain curve y 1 (t) Amplitude at the nearest frequency point, when When the frequency resolution of h i is the same as or different from omega i and smaller than that of the corresponding discrete Fourier transform, the value of h i is 0;p i,n , which is a real number with the value range of [0,2 pi ], and the value of h i is randomly selected in simulation according to the difference of i and n. And judging whether the bearing meets the requirements or not through simulation analysis of servo control errors.
  5. 5. The method for detecting a low-speed servo bearing according to claim 1, wherein the empirical model evaluation in the data evaluation step includes: the extracted characteristics comprise omega i 、ζ i , And Comparing with the historical characteristic values in the data set; and fitting a mapping function of the characteristics and the degree of the bearing through a neural network.
  6. 6. The method for detecting a low-speed servo bearing according to claim 1, wherein the quantization index evaluation in the data evaluation step includes: the method directly uses the numerical value of the time domain characteristic zeta i to judge, the bearing is qualified when zeta i is less than 0.01, is unqualified when zeta i is more than 0.5, and combines other evaluation modes in between.
  7. 7. A bi-directional manual integral stopper according to claim 1, wherein said signal acquisition step, the servo system constructed comprises: The motor drives the servo bearing to rotate, and the measuring element is a gyroscope or an angle measuring device; The excitation frequency omega i takes a value of 2n pi, wherein n is an integer between 30 and 50.
  8. 8. An on-machine self-checking system for servo bearings, based on a method for detecting a low-speed servo bearing according to any one of claims 1 to 7, characterized in that it comprises: A bearing excitation identification system for generating the control signal The servo bearing system comprises a servo bearing, a motor and a rotating load; the signal acquisition system is used for acquiring angular velocity data The servo bearing evaluation system is used for executing a data evaluation step; And the data set is used for storing the characteristics and the evaluation results.
  9. 9. An on-machine self-checking system of a servo bearing, characterized in that the system comprises; a memory storing a computer program; A processor for implementing a method for detecting a low-speed servo bearing according to any one of claims 1 to 7 when executing said computer program.
  10. 10. A computer readable storage medium having stored thereon a computer program, which when executed by a processor implements a method of detecting a low speed servo bearing according to any one of claims 1 to 7.

Description

Detection method of low-speed servo bearing and on-machine self-checking system of servo bearing Technical Field The invention belongs to the technical field of bearing detection, and particularly relates to a detection method of a low-speed servo bearing and an on-machine self-detection system of the servo bearing. Background When the bearing is part of a servo system, it is necessary to detect and evaluate the effect of the bearing on the servo system. The current patents or technologies for bearing detection are many, but almost all focus on the detection of end runout, radial runout, surface damage and characteristic frequency of the bearing, and fail to explain whether the control requirement of a servo system can be met. The existing and the previous methods are all used for bearing detection from aspects of end jump, radial jump, shaft swaying, damage position positioning and the like, and whether the corresponding bearing can meet the requirement of servo control cannot be clearly given. Therefore, how to provide a detection method of a low-speed servo bearing and an on-machine self-checking system of the servo bearing, on-machine detection of the bearing without disassembling the bearing can be realized by modifying a software code layer and modifying a small amount of hardware of the servo system with the servo bearing, and technical support is provided for maintenance self-checking of long-time operation of high-precision equipment, so that the method and the system have become a technical problem to be solved urgently. Disclosure of Invention The embodiment of the invention provides a detection method of a low-speed servo bearing and an on-machine self-checking system of the servo bearing, which can realize on-machine detection of the bearing without dismantling the bearing by modifying a software code layer and modifying a small amount of hardware of the servo system with the servo bearing, and provide technical support for maintenance self-checking of long-time operation of high-precision equipment. In one embodiment of the present invention, a method for detecting a low-speed servo bearing is provided, including: S101, a signal acquisition step of constructing a servo system and establishing a system identification model based on a system identification means, wherein the transfer function of the system identification model is as follows: Where s is the Laplace operator, y(s) is the Laplace transformation for measuring angular velocity, u(s) is the Laplace transformation for control signal, b and a are transfer function coefficients, τ is delay time; Determining moment excitation condition according to bearing use condition, applying control signal The expression is as follows: Wherein i is the experimental order, y 1rms is the root mean square value of an actual working condition angular velocity time domain curve y 1 (t), ω i is the excitation frequency, l i is the physical constraint limiting value, and m i is the offset; Collecting and storing angular velocity data under corresponding excitation conditions S102, signal processing step ofProcessing is carried out, and time domain features and frequency domain features are extracted; s103, data evaluation, namely judging whether the bearing meets the requirements of a servo system or not through simulation evaluation, empirical model evaluation or quantitative index evaluation based on the extracted characteristics; And S104, a data set updating step, namely saving the characteristics and the evaluation result into the data set for optimizing the evaluation accuracy. Further, the time domain feature extraction in the signal processing step includes: Calculating intermediate variable z i (k): Wherein k and m represent k and m sampling values corresponding to sampling or time sequence, and Q is a filter length parameter of sliding window mean value filtering; Extracting time domain features ζ i: The rotation speed signal of the ith experiment of N iIs used for the number of sampling points of (a),Is an intermediate variable; Further, the frequency domain feature extraction in the signal processing step includes: For a pair of N i -point discrete Fourier transform is performed on the N i -point equal-period sampling sequence to obtain a discrete spectrum Y i (N) Wherein Y i (n) representsThe conversion result of the N i point discrete Fourier transform of the equal period sampling sequence of N i points at the nth frequency point of the frequency domain is that Y i (N) is complex number, N is serial number, and the value range of N is an integer in the range of 0 to (N i -1); Extracting frequencies corresponding to the first lambda i maximum points And amplitude valueAnd recording the amplitude at the nearest excitation frequency omega i as Further, the simulation evaluation in the data evaluation step includes: constructing a servo closed-loop simulation model based on the system identification model G(s); constructing a rate disturbance sign