RU2730398C1 - Method of measuring linearity of speed and controlling its non-uniformity - Google Patents

Abstract

FIELD: measuring equipment.SUBSTANCE: invention relates to measurement equipment and can be used for accurate measurement of small linear velocities of object displacement, as well as for evaluation of speed deviation from preset value. Proposed method of linear velocity measurement and control of its non-uniformity involves three stages: 1) setting the required constant speed of the object along the straight line, measuring the coordinates of the object at equal time intervals specified by the microcontroller, calculating values of instantaneous speed Vfor each specified time interval, averaging obtained values of instantaneous speeds V, using method of sliding window, wherein the value of the window is selected experimentally, obtaining the averaged values of the instantaneous velocities V(t) and using these values V(t), calculating the velocity deviation on the required sections of the object movement path to estimate the non-uniformity of the measured linear velocity; 2) by means of Fourier transformation obtaining frequency spectrum of changes of instantaneous speed Vand analyzing it, then possible elimination of sources and causes of occurrence of non-uniformity of measured linear velocity; 3) repeating the actions performed at the first step, and additionally comparing the newly calculated speed deviation value with the previous one, if necessary repeating the second and third stages.EFFECT: disclosed method enables to measure the object linear velocity and determine the degree of its non-uniformity.1 cl, 4 dwg

Description

The invention relates to the field of measuring technology and can be used for accurate measurement of small linear velocities of movement of objects, as well as for assessing the deviation of the speed from a given value.

In various technological processes and installations, it is sometimes necessary to move objects at a given low constant linear speed. To accomplish this task, it is necessary to control the degree of deviation of the current speed from the given constant and, if possible, eliminate the causes of these deviations.

In particular, the method can be used to control the uniformity of movement of water-soluble elements during coating and to create thin films by uniformly immersing (extracting) these elements into a solution at a constant low linear velocity.

According to the inventor's certificate SU 36680, IPC G01P 3/38, a method for controlling the uniformity of movement is known. The sample is marked at exactly the same distance. In this case, a number of marks are applied at the controlled object during its movement at strictly equal intervals of time. Then the positions of the marks on the object and on the sample are compared and a conclusion is made about the uniformity of movement of the controlled object.

Also known is a method for automatic control and regulation of low speeds, using the effects of Mesbauer and Doppler (SU 159336, IPC G01P 5/10). This method measures the detuning from the resonance, which appears during the Doppler shift of the lines as a result of the movement of the radiation source relative to the absorber. The method allows non-contact and continuous measurement of very low speeds, down to fractions of a micron per second.

A known method for measuring the speed of an object (SU 1739297, G01P 3/64, publ. 07.06.1992), in which the linear speed of the object is measured when it moves along a given trajectory. Sensors are installed along the entire trajectory of the object. Then the speed value is determined at the sensor installation site by measuring the interaction time between the object and the sensor. In parallel, the time for the object to pass the entire trajectory of movement is recorded.

The disadvantage of all of the above methods of measuring the linear speed is the impossibility of determining the degree of unevenness of the speed of the object. In addition, it is impossible to determine the sources, causes of these irregularities, and, accordingly, eliminate them.

The problem to be solved by the present invention is to develop a method for measuring low linear speed, which also makes it possible to determine the degree of unevenness of the speed of an object by analyzing the speed of its movement along the entire path, and to identify the sources and causes of these irregularities, and then, based on the data obtained, to approximate the speed of the object to the required (constant) value.

The task is achieved by the fact that the developed method includes the following stages: the first stage, at which the required constant speed of movement of the object along a straight line is set and the coordinates of the object are measured at regular intervals, specified by the microcontroller. Next, the values of the instantaneous velocity V _i are calculated for each given period of time, and then the obtained values of the instantaneous velocities V _i are averaged using the sliding window method, and the value of the window w is chosen experimentally. Averaged values of instantaneous velocities V _i (t) are obtained, after which, using these values V _i (t), the velocity deviation is calculated on the required sections of the object's movement path for a primary assessment of the unevenness of the measured linear velocity. At the second stage, the initially obtained data (values of the instantaneous velocity V _i ) are processed using the Fourier transform, as a result, a spectrum of frequencies of changes in the instantaneous velocity V _{i is} obtained. After that, the obtained spectrum is analyzed and a conclusion is made about the causes of the arising velocity inhomogeneities and, if possible, the sources of occurrence of the detected irregularities are eliminated, trying to bring the object's speed as close as possible to the required (constant) value. At the third stage, the actions performed at the first stage are repeated, and in addition the newly calculated value of the velocity deviation is compared with the previous one, the effectiveness of the performed correction is estimated. Repeat the second and third steps if necessary.

The invention is illustrated by the following figures.

FIG. 1 shows a general view of a device for linear movement of an object with a low linear speed.

FIG. 2 shows the graphs of the dependence of the obtained values of the instantaneous speed on time: a) - without averaging, b) - with averaging over 500 points, c) - the initial section of movement (enlarged), d) - the final section of motion (enlarged).

FIG. 3 shows the speed spectrum: a) - a segment for frequencies from 0 to 300 Hz, b) - a segment for frequencies from 0 to 4.5 Hz.

FIG. 4 shows the speed spectra for three different linear motion systems.

To implement the claimed method (see Fig. 1), the required constant speed of movement of the platform 1 (object) is initially set; in an example of a specific implementation, this speed was 1.65 mm / s. The structure of the device for linear movement of the platform 1 includes rails 2 with carriages 3, a movement drive 4, a stepping motor (SM) 5 with a planetary gearbox (not shown in the figure). To measure the coordinates of platform 1, which was carried out at regular intervals, an optical ruler 6 of the DC-11 series of the corresponding length was used. For this series of rulers 6, the coordinate sampling (resolution) is 1 micron. To analyze the data from optical line 6, a Speedometer microcontroller with the appropriate software was used. In the "Speedometer" microcontroller, its own crystal frequency generator generates its own time, independent of computer interruptions. Once every 500 microseconds (2000 times per second), the microcontroller polls and writes to the buffer the coordinate value from the optical ruler 6. Then, data is transferred via the USB interface to the computer, in which the "Speedometer" program unit records the time and coordinates into a csv file for further analysis. Based on the obtained data, the values of the instantaneous speed V _i are calculated for each predetermined period of time, that is, 2000 values of the instantaneous speed per second are obtained. The graph of the dependence of the obtained values of the instantaneous speed V _i on time is presented in Fig. 2a.

то есть ΔV=2 мм/с. Таким образом, мгновенные значения V_i за каждый заданный промежуток времени могут меняться с интервалом, кратным 2 мм/с. Чтобы получить реальную скорость перемещения объекта, необходимо провести усреднение. Поэтому далее проводят обработку полученных значений мгновенных скоростей V_i, используя метод скользящего окна. Величину окна w выбирают экспериментально. В конкретной реализации способа оптимальным оказалось значение w=500. Получают усредненные значения мгновенных скоростей V_i(t), на основе которых строят новый график зависимости этих значений от времени (см. фиг. 2б). Видно, что скорость имеет периодический характер, в котором выделяются как низкие частоты, так и высокие. Для оценки неравномерности измеряемой линейной скорости вычисляют ее девиацию, причем делают это на требуемых участках перемещения платформы 1.Since the sampling of the used optical ruler 6 is ΔS = 1 μm, and measurements are carried out with a time interval Δt = 0.0005 s, the measured values of the velocity V _i will be multiples of the value

that is, ΔV = 2 mm / s. Thus, the instantaneous values of V _i for each given period of time can change with an interval divisible by 2 mm / s. To get the real speed of movement of the object, it is necessary to carry out averaging. Therefore, further processing of the obtained values of the instantaneous velocities V _{i is} carried out using the sliding window method. The window size w is chosen experimentally. In a specific implementation of the method, the value w = 500 turned out to be optimal. Averaged values of instantaneous velocities V _i (t) are obtained, on the basis of which a new graph of the dependence of these values on time is constructed (see Fig. 2b). It can be seen that the speed is of a periodic nature, in which both low and high frequencies are distinguished. To assess the unevenness of the measured linear speed, its deviation is calculated, and this is done at the required sites of movement of the platform 1.

DeV = (V _max - V _min ) / U _average ,

where Dev is the speed deviation at the estimated movement section,

V _max - maximum speed on the estimated area of movement,

V _min is the minimum speed in the estimated area of movement,

V _average - the calculated average speed in the estimated travel section.

In the specific case of the implementation of the method, taking into account the design of the movement device, which is based on the shaft-nut system, the control of unevenness (calculation of deviation) is necessary at least in the initial I and final II sections of the path of movement of the platform 1 (see Fig.2c and Fig.2d ). For normal operation of the mechanical system, which ensures the movement of the platform 1 (Fig. 1), free play is required, but at the same time, because of this, vibration sources are inevitably present in the system. The period of the speed deviation coincides with the rotational speed of the travel shaft. Since the vibration intensity increases from the initial point of the platform 1 movement to the final one (see Figs. 2b-d), then the velocity deviation, as an indicator of unevenness, will differ in the initial I and final II sections. The value of the velocity deviation was 6% at the beginning, and it increases by the end of the movement to 7.4%. This is due to the fact that there are a number of "thin places" in the movement system, such as the connections of the shafts of the motor 5, the gearbox and the travel shaft, which have angular and / or radial misalignments between them. For example, the connection of the stepper motor 5 and the planetary reduction gear is carried out rigidly by means of a longitudinally cut sleeve. This connection is clamped with one screw, respectively, the entire joint is pulled to the side by the amount of elasticity of the elements (shafts, housing elements, bearings). Accordingly, in this connection, the elastic force of the system in certain angles gives resistance to rotation, in other angles it accelerates rotation.

FIG. 2c-d shows the averaged values of the instantaneous speed V _i (t), consisting of a slowly changing average speed of the platform 1 and a fast vibration speed superimposed on the average. For most consumers, the magnitude of small speed deviations is unimportant in their tasks of moving the object. However, in some cases (in precision positioning systems in diamond micromilling machines, in installations for applying coatings on water-soluble elements, etc.), increased requirements are imposed on the smoothness and uniformity of movement of movement devices in technological processes characterized by relatively low speeds of about 1.5-2 mm per second.

If the value of the calculated deviation turns out to be unsatisfactory (too large) for the tasks of precision movement of the object, then the sources and causes of the "increased" unevenness of the measured linear velocity are identified. Using the Fourier transform, a spectrum of frequencies of changes in the instantaneous velocity V _{i is} obtained and the resulting spectrum is analyzed (see Fig. 3a and Fig. 3b). If we consider the process from the time-frequency point of view, the values of the instantaneous velocity V _i taken after a time of 500 μs, according to the Kotelnikov theorem, allow us to obtain and analyze its spectrum with an upper frequency limit of 1 kHz. For this task, this value of the upper limit is quite sufficient, since all mechanical resonances are in the frequency range less than 200 Hz. With the lower frequency limit, there are no such restrictions.

Vibration can be caused both by sources external to the displacement device, and internal, each of which is represented by the corresponding components in the obtained spectrum. Vibration is generally the harmonic vibration of a mechanical system. Internal sources of vibration are mechanically active elements of the installation - motors, bearings, carriages of movement. The direction of maximum vibration displacement depends on many factors. In the presented case, the data is analyzed from one coordinate of movement, that is, the projections of vibration displacement or vibration velocity to one selected coordinate are considered.

The drive system 4 of the shaft-nut type has a non-uniformity of speed caused by misalignment of the connected parts (shafts of the motor 5, gearbox and travel shaft), and such small speed deviations are invariably repeated throughout the entire period of motion. Therefore, the speed spectrum will contain a component with a frequency equal to the rotation frequency of these compounds. Accordingly, the speed spectrum of the displacement device will contain components corresponding to both the properties (defects) of the system itself and the influence of external sources of vibration (its projection onto the displacement axis).

FIG. 3a shows the spectrum of the speed at frequencies up to 300 Hz, which contains components with average frequencies of 0.413 Hz, 4.13 Hz, 82.5 Hz, 165 Hz, 247.5 Hz.

Analysis of the higher frequency portion of the speed spectrum can identify sources of vibration in the system. In the presented case, the support nodes of the carriage "hum". For this type of carriages, the spectrum contains many components that are grouped around frequencies f _n = n⋅A⋅V,

where n is an integer: 1,2,3 ...,

А = 50, - coefficient typical for this type of HIWIN supports, [1 / mm],

V - carriage speed, [mm / s].

Thus, f ₁ = 82.5 Hz, f ₂ = 165 Hz, f ₃ = 247.5 Hz for the case V = 1.65 mm / s.

FIG. 3b separately shows a section of the lowest frequency region of the spectrum from 0 to 4.5 Hz, which contains components with average frequencies of 0.413 Hz (drive shaft rotation frequency) and 4.13 Hz (ШД5 shaft rotation frequency in front of the planetary reduction gear (1:10) ).

To measure the linear velocity according to the proposed method, three linear displacement devices with different types of mechanical system providing movement were sequentially considered and compared.

In the first case, the system is arranged as follows: rails 2 with carriages 3 (manufactured by HIWIN) are placed on the supporting surface, the displacement drive 4 is a steel shaft with a caprolon nut, connected to a stepper motor 5 of the FL57 type (NEMA 23 or ISM-7402E) through a split sleeve. The measured deviation of the speed in this system was 6 ÷ 7.4%, while the speed of movement was 1.65 mm / s.

In the second case of the implementation of a mechanical system that provides movement, the displacement drive 4 was made on the basis of a shaft with a ball screw (ball screw), which was connected to the engine through a reduction gear (1:10). This solution made it possible to reduce the deviation of the displacement speed to 3 ÷ 4%.

In the third case, the system uses a linear stepper motor 5 (LSM), which made it possible to obtain a speed deviation of generally less than 0.5%. In the second and third cases, the travel speed was 1.6 mm / s.

FIG. 4 shows parts of the spectrum for the three above-described mechanical systems with different displacement drives, but using the same HIWIN guides and moving the platform 1 at the above speeds: a) for the shaft system with a caprolon nut, b) for the shaft-ball screw system, c) for the system with linear motor.

The magnitude of the deviation is uniquely related to the number of components in the spectrum, concentrated around the frequencies f _n . Since the deviation value for the first type of mechanical system was 6–7.4%, then in Fig. 4a shows many components concentrated around frequencies of 82.5 Hz and 165 Hz.

For the system with ball screws, the velocity deviation was 3–4%, which is almost two times less than for the first system. Therefore, the number of spectral components around frequencies of 80 Hz and 160 Hz is much less (Fig. 4.b). Also, a component appears on the spectrum at f _{ball screw} = 64 Hz. This component appears due to vibration introduced by the ball screw (ball screw), so it is not on the spectra of other systems. f _ball screw = 0.8⋅A⋅V,

where A = 50 [1 / mm], V = 1.6 [mm / s].

The fact that it is narrow indicates that the input movement (rotation) comes to it with minimal deviations in the SM system - a flexible coupling, and the ball screw system (a threaded shaft and a nut with balls on the shaft) makes the main contribution to the speed deviation.

FIG. 4c shows the spectrum for the Direct drive system, and the same spectrum components are visible at 80 Hz and 160 Hz, but they are very “narrow” compared to the components in the previous spectra, since the deviation for the third system was less than 0.5%.

Thus, the developed method makes it possible to obtain data, on the basis of which it can be concluded that, of the three systems considered, the best characteristics are possessed by a system using a linear stepper motor drive (LSD or direct drive).

In addition, it is possible to use the developed method for measuring the linear speed and monitoring its irregularity to identify the source / cause of the irregularity of the travel speed (wear, parts defect, inaccurate parts joining) within one device. For this, the actions described in the developed method are performed, that is, the deviation is calculated and the spectrum is analyzed, after which a decision is made on the need to disassemble the system and carry out defect identification of parts or compensation for inaccuracies in the connection of parts.

For example, in a linear displacement device (according to Fig. 1), it was found that the main contribution to the uneven motion is made by the misalignment of the connection when the rotation of the engine is transmitted to the rotation of the drive shaft. To compensate for existing angular and / or radial misalignments, an elastic coupling was used with a specific, selectable angle of inclination. For this, an adjustable element is added to the central section of a standard split coupling.

Thus, the developed method makes it possible to measure the linear velocity of an object and determine the degree of its unevenness, as well as to identify the sources and causes of velocity unevenness, and, based on the data obtained, bring the object's velocity as close as possible to the required (constant) value.

Claims (1)

A method for measuring the linear velocity and controlling its unevenness, including the following steps: at the first stage, the required constant speed of the object moving along a straight line is set, the coordinates of the object are measured at regular intervals specified by the microcontroller, the values of the instantaneous speed V _i are calculated for each given period of time, then averaged the obtained values of the instantaneous velocities V _i , using the sliding window method, and the window size is chosen experimentally, the averaged values of the instantaneous velocities V _i (t) are obtained, after which, using these values V _i (t), the velocity deviation is calculated on the required sections of the object's path to assess the unevenness of the measured linear velocity, at the second stage, using the Fourier transform, a spectrum of frequencies of changes in the instantaneous velocity V _{i is} obtained and analyzed, then, if possible, the sources and causes of the unevenness of the measured linear velocity are eliminated, at the third stage the actions are repeated i performed at the first stage, and additionally compare the newly calculated value of the velocity deviation with the previous one, if necessary repeat the second and third stages.

Images