An Automatic Correction Method of Harmonic Frequency of Servo Drive Signal in Numerical Control System

Abstract

The invention discloses an automatic correction method of a numerical control system servo driving signal harmonic frequency. The method comprises the following steps of collecting a numerical control system servo driving signals, carrying out fast Fourier transform, arranging magnitude spectra of the servo driving signals with an order from large to small according to amplitudes and acquiring an ordered sequence A of the amplitudes, wherein the A= (a1, a2, ..., an); using a Fibonacci sequence method or a golden section method to search and acquire a segment point in the ordered sequence A, acquiring average values A1 and A2 of amplitudes in a first segment and a second segment of the ordered sequence A respectively and further calculating and acquiring a threshold t; using the threshold t to extract and store each harmonic wave band in the magnitude spectra one by one; through using each extracted harmonic wave band and combining a correction formula, realizing automatic correction of the harmonic frequencies. In the invention, according to the amplitudes, the harmonic frequencies are extracted one by one and corrected; the threshold t which is greater than or equal to a first layer mean value and is less than or equal to a second layer mean value is used to effectively screen the harmonic wave band so that on-line and real-time rapid correction of the harmonic frequencies is realized.

Description

technical field [0001] The invention belongs to the field of signal analysis and processing, and more specifically relates to an automatic correction method for the harmonic frequency of a servo drive signal of a numerical control system. Background technique [0002] The CNC machine tool servo drive system is the basic component to realize the machining, cutting and forming movement of parts. The current data collected from it can track the position and speed of the CNC device sensitively and accurately, and can be used to diagnose the assembly quality of the CNC system and optimize the process parameters of the CNC system. The collected current data spectrum mainly includes the harmonics generated by the two harmonic sources of the transformer and the servo motor. These two harmonic frequencies need to be extracted for experimental verification and analysis during spectrum analysis. There are two main problems in harmonic extraction: fence effect and spectrum leakage. Fenc...

AI Technical Summary

Problems solved by technology

There are two main problems in harmonic extraction: fence effect and spectrum leakage. Fence effect refers to the phenomenon that harmonic spectral lines are "barred" by integer multiples of the frequency resolution of harmonic frequency parts. Spectrum leakage refers to the phenomenon of The truncation or fence effect causes harmonic energy to leak to the surroundings and affects the amplitude of adjacent spectral lines. The two will lead to deviations in the extraction of harmonic frequencies. Therefore, when extracting the harmonic frequency of the servo drive signal of the CNC system, it is necessary to check frequency correction

For example, CN201510260333.2 discloses a harmonic measurement channel calibration method based on the MIR-RSD high-precision cosine window interpolation FFT algorithm, which utilizes the spectral line k1 with the largest amplitude near the frequency fN in the frequency spectrum and its left spectral line k2 and right spectral line Line k3 calculates the weighted ratio coefficient β and the non-integer frequency component δ, corrects the amplitude, frequency and phase angle according to the δ value, but it fails to realize the automatic extraction of harmonic bands according to the spectrum characteristics; another example, Hunan University Wen Wen et al. discussed the spectral characteristics of the lowest and steepest falling windows of side lobes, and derived the spectrum correction formula combined with curve fitting; The correction formula for window function harmonic analysis; the typical algorithm for interpolation correction of FFT is bispectral interpolation correction algorithm, which needs to correct all spectral lines point by point with the algorithm mentioned in the above literature, and the calculation is complex and large , the computational efficiency is low

Images