Method and device for evaluating and correcting the aging of a feed system based on a path position signal

Abstract

This invention relates to a method and apparatus for aging assessment and correction of a feed system based on path position signals. The apparatus includes a data acquisition module, a spectrum analysis module, an aging judgment module, a model identification module, and a model coefficient update module. By acquiring the position commands and feedback signals of the machine tool feed axis during actual reciprocating motion, the position error is calculated and then frequency domain analysis is performed to extract aging characteristics. A standard second-order transfer function model is built-in, and an adaptive differential evolution algorithm is used to automatically identify the feed system, which can determine the degree of aging of the mechanism. Furthermore, the control adjustment module is used to write back and update the model coefficients, providing a basis for feed axis compensation. This invention can complete aging assessment and correction using only the position commands and feedback signals in the daily machining of the machine tool.

Claims

1. A method for aging assessment and correction of a feed system based on path position signals, applied to a feed servo system of a machine tool, the method comprising the following steps: (A) Acquisition step: performing a reciprocating path test on one feed axis of the feed servo system, and acquiring the position command signal and the corresponding position feedback signal of the feed axis in the reciprocating path; (B) Signal processing step: calculating the difference between the position command and the position feedback signal, generating a position error signal, and performing spectral analysis on the error signal to obtain the amplitude characteristics corresponding to the system resonant frequency; (C) Model building step: establishing a standard second-order transfer function model; (D) Coefficient identification step: using an adaptive differential evolutionary algorithm (SaDE), searching and optimizing the model coefficients based on the spectral characteristics, and searching and identifying unknown coefficients in the model based on the characteristic frequency, so that the predicted response of the transfer function matches the measured error signal; (E) Correction step: updating the simulation model of the feed axis based on the identified model coefficients to correct the control parameters of the feed servo system.

2. The aging assessment and correction method for a feed system based on path position signals as described in claim 1, wherein, In the signal processing step (B), when the position error signal exceeds a preset aging threshold, an aging warning is output.

3. The aging assessment and correction method for a feed system based on path position signals as described in claim 1, wherein, The standard second-order transfer function model has the following mathematical form: where ωn​ is the natural frequency, ξ is the damping ratio, and s is the Laplace variable.

4. The aging assessment and correction method for a feed system based on path position signals as described in Request 1, wherein, The spectrum analysis system performs a Fast Fourier Transform (FFT) on the error signal and extracts features from the main resonant frequency range.

5. The aging assessment and correction method for a feed system based on path position signals as described in Request 1, wherein, The adaptive differential evolutionary algorithm (SaDE) optimizes and identifies model coefficients through the following steps: (1) Randomly initialize multiple candidate solutions in a coefficient group for each model coefficient; (2) During the evolution process, randomly select several candidate solutions from the coefficient group and generate corresponding mutation vectors based on their coefficient differences; (3) Use the mutation vectors to generate a trial vector and calculate its fitness; (4) Compare the fitness values ​​of the trial vector with those of the original candidate solutions, and select the one with better fitness to retain in the next generation of coefficient group; (5) Repeat the above steps until the convergence condition is met, and obtain a set of optimal model coefficients as the dynamic model coefficient output of the feed system.

6. The aging assessment and correction method for a feed system based on path position signals as described in claim 1, wherein, It further includes a multi-axis integration step, which involves separately capturing and analyzing the position commands and feedback signal data of multiple feed axes, independently executing steps (B) to (E) for each feed axis to obtain the model of each axis, and transforming the model of each axis into a multi-axis synchronous system model through a mechanism chain matrix.

7. The aging assessment and correction method for a feed system based on path position signals as described in claim 6, wherein, The multi-axis integration step further includes: calculating the rotational lag using the mechanism chain matrix, converting it into a machining linear deviation, calculating appropriate feedforward parameter values ​​for the multi-axis to match it, simulating synchronization error trends, and providing adjustment suggestions for multi-axis coordinated control.

8. A feed system aging assessment and correction device based on path position signal, applied to a feed servo system, the feed servo system including a feed axis, a corresponding CNC controller, and a control parameter adjustment unit, characterized in that the device includes: a data acquisition module, for acquiring position command signals and position feedback signals of the feed axis during a reciprocating motion process from the CNC controller; a spectrum analysis module, for calculating the difference between the position command and position feedback signals from the data acquisition module, generating a position error signal, and performing frequency domain analysis on the position error signal to extract frequency domain features of a specific frequency band; an aging judgment module, for judging whether the feed axis has aging phenomena based on the position error signal from the aforementioned spectrum analysis module; and a model identification module, which includes a system model constructed using a standard second-order transfer function, and estimates the system model coefficients of the feed axis based on the frequency domain features of the spectrum analysis module through an adaptive differential evolution algorithm. A model coefficient update module writes back the identified model coefficients to the control parameter adjustment unit of the feed servo system for use in the feed axis compensation setting.

9. The aging assessment and correction device for a feed system based on path position signals as described in claim 8, wherein, The aging judgment module compares the position error signal with a preset aging threshold. When the position error signal exceeds the aging threshold, it outputs an aging warning signal.

10. The aging assessment and correction device for the feed system based on path position signal as described in claim 8 further includes a multi-axis integration module, which is used to synchronously process the commands and feedback data of multiple feed axes, perform independent frequency domain analysis and coefficient identification on them, and integrate their parameters as a synchronous dynamic model of the overall servo platform.

11. The aging assessment and correction device for a feed system based on path position signals as described in claim 10, wherein, The multi-axis integrated module further uses a mechanism chain matrix calculation to convert the rotational lag into machining linear deviation, in order to calculate appropriate feedforward parameter values ​​for each feed axis, and connects to the control parameter adjustment unit of multiple feed axes for synchronous control compensation settings.

Images