RU2535334C2 - Method of vibration diagnostics of machine resilient system with application of power impact generator included into "machine-accessory-tool-part" system - Google Patents

Abstract

FIELD: measurement equipment.

SUBSTANCE: invention is designed for diagnostics of a resilient system in metal-cutting machines. The method of vibration diagnostics of the resilient system of the machine using a power impact generator, included into the "machine-accessory-tool-part" system, consisting in the fact that harmonic, pulse or random excitation is arranged at the inlet in the resilient system of the machine, and system response at the outlet is measured, at the same time to produce dynamic characteristics, they excite the investigated structure with the help of the measured dynamic force, differing by the fact that harmonic and random excitation is provided with the help of a piezoceramic contact vibrator, and to create pulse power impact, a generator is used, afterwards signals are supplied to a two-channel spectrum analyser, in which they produce, with the help of spectral analysis of complex signals, based on quick Fourier transform, frequency characteristics, and analogue signals arriving to the analyser inlets are filtered, selected and converted with the help of an analogue-to-digital converter into a digital form for production of series of digital data - implementation, and by speed of selection and duration of implementation they determine the frequency range and the resolving capacity during analysis of investigated characteristics, and the force supplied to the investigated object during turning of the mandrel with the cutter is measured with the help of a piezoelectric dynamometer.

EFFECT: increased accuracy of measurements, as well as expansion of technological capabilities during diagnostics of a resilient system of a machine.

2 cl, 5 dwg

Description

The invention is intended for diagnostics of an elastic system of machine tools.

Currently, the industry produces stands and devices for controlling the parameters of vibroacoustic signals, which can be used to judge the dynamics of the elastic system of the machine by the state of the bearing assemblies [Balitsky F.Ya., Ivanova MA, Sokolova AG, Khomyakov E.I. . Vibroacoustic diagnostics of incipient defects. - M .: Nauka, 1984. - p. 78-83]. The assembly of high-speed spindle units is carried out in thermostatically controlled rooms, according to a strictly defined method, with strict control of the deviations of individual parts from a given geometry, and after assembly, the spindle is run for many hours on a special stand with registration of temperature at several points of the unit and the moment of resistance to rotation.

The disadvantages of the known methods include the fact that by controlling only the temperature, it is impossible to penetrate into the essence of the processes occurring in the spindle units during idle rotation of the spindle, when operating under load and with increasing temperature. Today, the need has ripened for the application of new methods and methods of vibroacoustic diagnostics, which can penetrate much more deeply than temperature into the essence of the processes occurring in the spindle units when the spindle is idle, when working under load and when the temperature rises.

The closest technical solution to the technical nature and the achieved result is a method for diagnosing a spindle assembly according to the patent of the Russian Federation No. 21244966, class. B23B 25/06, G01M 13/02 (prototype). According to the prototype, the diagnosis is implemented as follows. After selecting the test mode, the machine turns on, and the middle part of the mandrel is machined with a cutter. The signals from the displacement sensors located in two cross sections of the mandrel are fed first to the amplification-converting equipment, and then to the computer, where the trajectory of the axis of the mandrel in two sections is constructed. As a result of the movement, the tip of the cutter describes a curve on the surface of the mandrel, which forms a "geometric image" of the processed section. The software allows building a “geometric image” on a display screen in three-dimensional space, which determines a whole set of accuracy parameters of the processed mandrel, i.e. carry out harmonic, pulsed excitation at the input in the elastic system of the machine and measure the response of the system at the output, while, to obtain dynamic characteristics, the investigated structure is excited with the help of the measured dynamic force.

A disadvantage of the known technical solution is the relatively low accuracy of reproducing the geometric image of the processed cross section of the reference workpiece and the lack of the possibility of vibro-acoustic diagnostics, which allows a much deeper assessment of the nature of the processes occurring in the spindle units when the spindle is idle, when operating under load and when the temperature rises.

A technically achievable result is an increase in the accuracy of measurements, as well as the expansion of technological capabilities during diagnostics of the elastic system of the machine.

This is achieved by the fact that in the method of vibration diagnostics of the elastic system of the machine using a power generator included in the machine-tool-tool-component system, which consists in the fact that harmonic, pulse or random excitation is performed at the input in the elastic system of the machine and measure the response of the system at the output, in order to obtain dynamic characteristics, the investigated structure is excited with the help of a measured dynamic force, while harmonic and random excitation provide a piezoelectric ceramic contact vibrator, while the alternating force is created by piezoceramic rings, which are supplied with electric voltage, and the thickness of the piezoelectric element is changed, and to create a pulsed force, a generator is used, which is a cylindrical mandrel in which a longitudinal groove of a given depth is realized that implements the amplitude of the input pulsed impact, and which is rigidly fixed in the spindle unit of the machine, and the frequency of the input pulse action is set by the rotation speed spindle, in this case, the width of the groove changes the pulse duration, and hence the frequency range of the excitation spectrum, after which the signals are fed to a two-channel spectrum analyzer, which is obtained by spectral analysis of complex signals, the basis of which is the fast Fourier transform, frequency characteristics, and analyzer inputs the analog signals are filtered, selected and converted using an analog-to-digital converter into digital form to obtain a series of digital data - implementation, presentation They represent the time history of signals over the corresponding time intervals, and the frequency range and resolution in the analysis of the studied characteristics are determined by the sampling speed and duration of implementation.

Figure 1 shows a block diagram of dynamic excitation during vibration diagnostics of the elastic systems of machine tools, figure 2 shows a diagram of a piezoelectric vibrator for contact loading of an elastic system, figure 3 shows a diagram of a pulse generator included in the system "machine-tool-tool- detail ", figure 4 shows the installation diagram of a piezoelectric dynamometer, figure 5 shows a block diagram of a two-channel spectrum analyzer.

A device for implementing the method of vibration diagnostics of an elastic system of a machine using a force generator included in the machine-tool-tool-component system (Fig. 1) includes a frame 1 on which a spindle unit 2 and a generator 3 are mounted, respectively, to obtain Signals: 4 - the simplest sinusoidal oscillation, 5 - a set of oscillations, 6 - addition of sinusoidal oscillations, 7 - excitation in the form of a pulse signal. To obtain a harmonic sinusoidal signal, a vibrator is used, and for a pulsed cylindrical mandrel with a longitudinal groove. To obtain dynamic characteristics, it is necessary to excite the investigated structure with the help of a measured dynamic force. Figure 1 shows a block diagram of dynamic excitation in the diagnosis of elastic systems of machines. Block 8 implements a sinusoidal signal, and block 9 serves to generate pulsed signals. Block 10 is designed to process the received signals of dynamic effects on the elastic system of the machine and their processing, as well as graphing. With random and pulsed excitation, the frequency characteristics can be obtained using spectral analysis of complex signals, the basis of which is the fast Fourier transform.

Harmonious or random excitation is provided using an electromagnetic non-contact vibrator (not shown in the drawing), which is placed on the machine so that the force developed by it coincides with the cutting force. When contact excitation using a piezoelectric vibrator (figure 2), the variable force is created by piezoceramic rings 13, which are supplied with electrical voltage through the connector 17. Because of this voltage, the thickness of the piezoelectric element changes. Changing the linear size of the column of piezoelectric elements through the mandrel 14, the measuring piezoelectric elements 16, the tip 15 is transmitted to the part of the machine, which you want to apply force. The magnitude of the static force is controlled using strain gauges 18 glued to the deformable part of the base 11. The non-conductive housing 12 protects the researcher from high voltage applied to the piezoelectric elements.

To create a pulsed force action, a pulsed action generator is used, which is included in the “machine-tool-tool-component” system, the circuit of which is shown in FIG. 3. The generator is a cylindrical mandrel 19, in which a longitudinal groove 20 is made of a given depth, which implements the amplitude of the input pulse action, and which is rigidly fixed in the spindle unit of the machine. The frequency of the input pulse action is set by the spindle speed. In a section perpendicular to the axis of the spindle assembly of the machine (Fig. 3), the groove 20 is made with inclined side surfaces lying in planes intersecting along a line coinciding with the axis of the mandrel 19 and in a plane perpendicular to the axis of the spindle coinciding with the center of the circle. Moreover, the surface connecting the side planes is a part of the cylindrical surface, the equidistant external cylindrical surface of the mandrel 19.

The force applied to the test object during turning by the cutter 21 of the mandrel 19 is measured using a piezoelectric dynamometer 22 (Fig. 4) mounted between the supporting surfaces of the support and the cutter.

The method of vibration diagnostics of the elastic system of the machine using a pulse generator included in the system "tool-tool-part" is as follows.

The method of vibration diagnostics of the elastic system of the machine is that at the input, harmonic, impulse or random excitation is carried out in the elastic system of the machine and measure the response of the system at the output. To obtain dynamic characteristics, it is necessary to excite the investigated structure with the help of a measured dynamic force. To create a pulsed force action, a generator is used, which is a cylindrical mandrel 19, in which a longitudinal groove 20 is made of a given depth, which implements the amplitude of the input pulsed action, and which is rigidly fixed in the spindle unit of the machine. The frequency of the input pulse action is set by the spindle rotation speed (Fig. 3), the width of the groove 20 changes the pulse duration, and hence the frequency range of the excitation spectrum.

Harmonious or random excitation is provided using an electromagnetic non-contact vibrator (figure 2), while the alternating force is created by piezoceramic rings 13, which are supplied with electrical voltage, due to which the thickness of the piezoelectric element changes. Changing the linear size of the column of piezoelectric elements through the mandrel 14, the measuring piezoelectric elements 16, the tip 15 is transmitted to the part of the machine, which you want to apply force. The magnitude of the static force is controlled using strain gauges 18 glued to the deformable part of the base 11.

Figure 4 presents a block diagram of a two-channel spectrum analyzer. With random and pulsed excitation, the frequency characteristics are obtained by spectral analysis of complex signals, the basis of which is the fast Fourier transform. The principles of spectral analysis are considered (Fig. 4) by the example of a two-channel analyzer performing a fast Fourier transform. The analyzer can be used as a "black box" that measures the excitation and reaction signals and determines the frequency characteristics based on these measurements. The analog signals arriving at the analyzer inputs are filtered, selected, and converted using an analog-to-digital converter into digital form to obtain series of digital data called implementations. These implementations represent the time history of signals over the corresponding time intervals. The sampling rate and the duration of the implementation determine the frequency range and resolution in the analysis.

Figure 5 presents the stages of the conversion of the signal and spectra in the spectrum analyzer. Registered implementations can be multiplied by the weight function. Thus, data is narrowed at the beginning and end of the implementation, which makes them more convenient for block analysis. Weighted realizations are converted to the frequency domain in the form of complex spectra using the discrete Fourier transform.

The evaluation function W ₁ , equal to the ratio of the mutual spectrum of the reaction and the force to the intrinsic spectrum of the force, is used to minimize noise at the output of the system; random noise at the output is removed in the process of averaging the mutual spectrum. With an increase in the number of averagings, W ₁ tends to the true frequency response W (ω).

Estimated function W ₂ equal to $$\frac{G_{yy}\left(\omega \right)}{G_{yF}\left(\omega \right)}$$

where G _yy (ω) is the transfer function of the displacement, G _yF (ω) is the transfer function of the displacement and the transmitted force during blade cutting, it is used to minimize the influence of input noise, since it is removed from the mutual spectrum during averaging.

With an increase in the number of averaging cycles, W ₂ tends to the true frequency response W (ω). In case of random excitation and investigation of resonances, the best estimate of the frequency response is W ₂ , since it compensates for input noise and is less sensitive to scattering. In the study of antiresonance zones, W ₁ is considered the best estimate of the frequency response, since the main thing in this case is its low sensitivity to output noise. When there is noise at the output and at the input, the functions W ₁ and W ₂ can be considered the limits of the confidence interval for the true frequency response W (ω). However, this does not apply to non-linear systems and to cases with coherent noise at the input and output.

The coherence function provides a means for assessing the degree of linearity of the coupling of input and output signals:

, где 0≤γ²(ω)≤1, $${\gamma }^2\left(\omega \right)=\frac{{|\; |\; |G_{yF}\left(\omega \right)|\; |\; |}^2}{G_{yy}\left(\omega \right)\cdot G_{FF}\left(\omega \right)}$$

where 0≤γ ² (ω) ≤1,

where G _FF (ω) is the transfer function of the transmitted force during blade cutting. The boundary values of the coherence function are 1 in the absence of noise and 0 in the presence of pure noise. As an interpretation of the coherence function, we can say that for each frequency it indicates the degree of linear dependence between the signals at the input and output of the system. The coherence function is similar to the square of the correlation coefficient used in statistics. In dynamic studies, this important property of the coherence function is used to identify a number of possible errors.

Using the frequency characteristics obtained in one way or another, it is possible to evaluate the vibration resistance of the machine’s dynamic system. For example, with blade processing, the maximum width of the cut layer: $$b_{PR}=\frac{\mathrm{one}}{K|\; |\; |{\mathrm{Re}}_{\mathrm{At}\mathrm{FROM}}^0|\; |\; |}$$

- отрезок, отсекаемый годографом упругой системы станка на отрицательной части вещественной оси. Чем больше отрезок $${\mathrm{Re}}_{УС}^0$$

, тем меньше предельная ширина срезаемого слоя и ниже виброустойчивость динамической системы станка.where K is the cutting coefficient (specific cutting force): $${\mathrm{Re}}_{\mathrm{At}\mathrm{FROM}}^0$$

- the segment cut off by the hodograph of the elastic system of the machine on the negative part of the material axis. Larger segment $${\mathrm{Re}}_{\mathrm{At}\mathrm{FROM}}^0$$

, the smaller the limiting width of the cut layer and lower the vibration resistance of the dynamic system of the machine.

Claims (2)

1. The method of vibration diagnostics of the elastic system of the machine using a power generator included in the system "machine-fixture-tool-part", which consists in the fact that the input is harmonic, pulsed or random excitation in the elastic system of the machine and measure the response of the system at the output in this case, to obtain dynamic characteristics, the investigated structure is excited with the help of a measured dynamic force, characterized in that the harmonic and random excitation is provided using a piezo contact vibrator, while the alternating force is created by piezoceramic rings, which are supplied with electric voltage, and the thickness of the piezoelectric element is changed, and to create a pulsed force action, a generator is used, which is a cylindrical mandrel in which a longitudinal groove is made of a given depth, which implements the amplitude of the input pulse action , and which is rigidly fixed in the spindle unit of the machine, and the frequency of the input pulse action is set by the spindle speed, In this case, the pulse width changes the width of the pulse, and hence the frequency range of the excitation spectrum, after which the signals are fed to a two-channel spectrum analyzer, which is obtained by spectral analysis of complex signals, the basis of which is the fast Fourier transform, frequency characteristics, and analog inputs to the analyzer the signals are filtered, selected and converted using an analog-to-digital converter into digital form to obtain a series of digital data - implementations representing the temporal history of the signals over the corresponding time intervals, and the frequency range and resolution are determined by the sampling speed and duration of the implementation when analyzing the characteristics under study, and the force applied to the object under study when turning the mandrel with a cutter is measured using a piezoelectric dynamometer installed between the supporting surfaces of the caliper and cutter. 2. The method of vibration diagnostics of the elastic system of the machine using a force generator included in the system "machine-tool-tool-part" according to claim 1, characterized in that the registered implementation is multiplied by the weight function and thereby narrowing the data at the beginning and at the end of the implementation, which makes them more convenient for block analysis, then the weighted implementations are converted to the frequency domain in the form of complex spectra using the discrete Fourier transform, and as a result of the inverse transformation to the semi initial time sequences are used, the averaging method is used, as a result of which the noise is eliminated and the degree of statistical reliability is improved, then the estimated function is determined equal to the ratio of the mutual reaction spectrum and force to its own force spectrum, and it is used to minimize noise at the system output, and random noise at the output is removed in the process of averaging the mutual spectrum, and with an increase in the number of averagings, a true frequency response is obtained, and according to the obtained frequency characteristics, vibration resistance of the dynamic system of the machine.

Images