CC BY
7TECHNICAL SCIENCES
ОЦ1НКА УМОВ РОБОТИ ВИХРОСТРУМОВОГО ПРИСТРОЮ ДИНАМ1ЧНОГО РЕЖИМУ
КОНТРОЛЮ
Абрамович А.О.
Нацюнальний техтчний унгверситет Украши «Кшвський полтехтчний тститут шет 1горя Ci-
корського», м. Кшв, Украша, к.т.н., iнженер
Шддубний В. О.
Нацюнальний техтчний утверситет Украти «Кшвський полiтехнiчний тститут iMew 1горя Ci-
корського», м. Кшв, Украша к.т.н., доцент
Баженов В.Г.
Нацюнальний техтчний унiверситет Украши «Кшвський полтехтчний тститут iменi 1горя Ci-
корського», м. Кшв, Украша к.т.н., доцент
EVALUATION OF THE OPERATING CONDITIONS OF THE EDDY CURRENT DEVICE IN THE
DYNAMIC CONTROL MODE
Abramovych A.
National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute" Kyiv, Ukraine,
engineer Piddubnyi V.
National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute" Kyiv, Ukraine,
assistant professor Bazhenov V.
National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute" Kyiv, Ukraine,
assistant professor DOI: 10.5281/zenodo.7298704
АНОТАЦ1Я
Розглядаються результата роботи динашчног системи вихрострумового контролю. Розроблено макет системи, що дозволяе проводити контроль металевих зразшв у динамiчному режима Розроблено матема-тичну модель, яка описуе змши шформацшних критерив сигналу вщгуку у залежносп вщ матерiалу дос-лщжуваних металевих об'екпв та швидкосп лшшного перемщення гх вщносно котушок системи контролю. У дослвдженш акцентовано увагу на пошуку оптимальних швидкостей роботи системи контролю. Запропоновано iнтегрально-диференцiйний пiдхiд для визначення максимуму шформацп м1ж сигналами-вiдгуками отриманими ввд рiзних метатв. Визначено оптимальний дiапазон швидкостей сканування, що дозволяе отримати максимальну чутливють iнформацiйних критерiíв для рiзних метатв, та може бути ви-користаним для побудови мобшьних систем вихрострумового експрес контролю.
ABSTRACT
The results of the dynamic eddy current control system are considered. A layout of the system was developed, which allows control of metal samples in dynamic mode. A mathematical model has been developed that describes changes in the information criteria of the response signal depending on the material of the investigated metal objects and the speed of their linear movement relative to the coils of the control system. The research focuses on finding optimal operating speeds of the control system. An integral-differential approach is proposed for determining the maximum information between feedback signals obtained from different metals. The optimal range of scanning speeds has been determined, which allows obtaining the maximum sensitivity of information criteria for various metals, and can be used to build mobile eddy current express control systems.
Ключовi слова: електромагнггш властивосп металiв, вихрострумовий перетворювач, дистанцшна вдентифжащя металiв, математичне моделювання, амплпудно-фазовий метод реестраци сигналiв.
Keywords: electromagnetic properties of metals, eddy current converter, remote identification of metals, mathematical modeling, amplitude-phase signal registration method.
• Постановка проблеми
The task of detection and identification of metal objects in various fields of science, technology, maintenance control and military affairs has always been relevant. This is related to the search for hidden explosive metal objects, communications, the determination of the metal from which an unknown object is made, the identification of found metal nuggets in mining. For
their search (detection), electronic devices (metal detectors) are used, built on the registration of differences in the electrical and magnetic characteristics of hidden objects and the environment in which they are located. Such devices, in addition to searching, carry out preliminary metal identification. They work in a dichoto-mous mode. That is, it is determined to which subgroup of the subgroup of metals the material of the investigated control object belongs. Known metal detectors
are built on the basis of the eddy current method for detecting metal objects and divide them into black and non-ferrous, but do not provide the ability to determine metal within subgroups, that is, they cannot distinguish gold from aluminum or copper, nickel from steel, etc.
The final identification of the composition of a metal object is carried out by laboratory analysis using various methods, in particular, X-ray, optical emission, eddy current [1]. It is quite effective to use the dynamic method proposed by the authors in [2], which is also based on the eddy current method of detecting metal objects [3]. Theoretical issues of the formation of signals of a dynamic eddy current system were considered in a number of works [3-5], but they do not fully explain the change in the shape of the response signal obtained as a result of probing the investigated metal object.
• Аналiз останшх дослщжень i публжацш
The existing mathematical models that explain the
operation of the eddy current method of detecting metal objects cannot fully explain the occurrence of the signal and do not reflect the informative features that characterize the differences of metals embedded in the form of the response signal. In particular, these are the models of K. Brushini [3], Hrynyova A.Yu. [4] and the model is taken from the theory of non-destructive testing [5].
• Видшення невиршених частин загально! проблеми
Eddy current devices, which distinguish materials from each other by magnetic permeability and electrical conductivity, are used to detect metallic objects. They can have different schematic implementations, structural and technical characteristics. However, such a parameter as scanning speed is not regulated for them, which largely depends on the shape of the response signal. And this is important for solving the problem of identification of metal objects.
• Мета стал
This publication proposes a mathematical model of the formation of the response signal of a dynamic eddy current system [2], which allows identifying the metal from which it is made based on the electromagnetic parameters of the object under study, and provides the results of optimizing the scanning speed of the system in question.
• Виклад основного матерiалу
We will show that the differences in metals depend on their electromagnetic characteristics. For this, we will use the model described in [5]. In this model, the voltage at the output of the receiving coil of the Пвд eddy current converter is determined by the electromagnetic properties of the investigated control object (ОС) and the parameters of the coils:
Uefl{x,t) = ja /u0%NдЫ3Rд Im с о s (со t)-^
да
•|ф! (X, p) J1 (xRЗ*) J1 (x)e~xh*dx(3 . 2)
0
where N д, N 3 - the number of turns in the receiving and transmitting coils of the frame type; J -Bessel function of the first kind of the first order; X -
integral transformation parameter; R- the radius of
the receiving coil; R3 - the radius of the transmitting coil;
h* —
h, + h
Д .
Rz
R 3* —
R
Д
Rz
where - the height of the transmitting coil, h^ - the height of the receiving coil.
That is, the magnitude of the signal presented in the receiving coil depends on the parameters of the coils. In our case, the coil system had the following geometric dimensions: R^ = 80 mm, R = 40 mm, =
5 mm, = 5 mm , N3 = 85, N^ = 280, x = X R3
In this model, such a characteristic as the influence function is introduced pj (x, P) [6,7]. It takes into account the electrical and magnetic properties of metals, and can be used to estimate the difference in feedback signals obtained for different metals:
фi (xp) —
Vr - Vx
ix
jP 2
Vr +Jx 2 + jP 2
(2)
- relative magnetic permeability
r
1 -
integration parameter [1/m], a - electrical conductivity of
where x = X R 3,
P = R W ® V a a
metal, ^ - absolute magnetic permeability.
For example, we will show the difference in function values for steel, nickel, lead and electrotechnical copper. As is known, the relative magnetic permeability of steel =100, copper =1, nickel
=100...600 (use 100), lead fir =1, conductivity of steel a =7,69'106 sm/m, copper a =58,1 * 106 sm/m, nickel a=11,5 >106 sm/m, lead a =4,81 >106 sm/m. The authors calculated the values of the influence function (x, P) for various metals:
11 (x,P)| = 0,802 ,|plnlckel (x,P) = 0,833 , |*icoPPer (*, P)| = 0,9 9 2, |pUead (x, P) = 0,9 7 2 .
Data analysis shows that the difference between the influence functions for copper and steel is more than 19.0%, and for copper and nickel - more than 16.0%. This makes it possible to identify the metal from which the OK is made by the value of the Ubm voltage and its change over time.
Information about the material of the metal OC is also carried by the phase shift between the probing signal and the signal transmitted in the receiving antenna. In works [8, 9] it is proposed to use the spectral characteristics of the response signal, and in [2,10] the method of graphic-digital images, which allows analyzing the signal in the time domain.
Consider the structural diagram of the modified dynamic system (Fig. 1). It consists of electronic units used in metal detectors (probing signal generator, output amplifier, low-noise input amplifier of the signal presented in the receiving coil, phase detector) [3] and
a digital signal processing unit based on a microcontroller. The system is supplemented with a rotary setup, which is used for dynamic measurement of the response signal with the possibility of changing the frequency of rotation of the OK on the coils and for studying the effect on the shape of the response signal of the speed of movement of the investigated object relative to the block of coils (dynamic measurement of the response signal).
The coil unit (transmitting and receiving coils) and analog electronic unit and advanced digital processing
unit based on a modern microcontroller, which provides identification of the material of the metal object using the graphic-digital image and spectral methods developed by the author [3], are unchanged. It is the presence of a rotary device (Fig. 1) that allows the dynamic measurement method to be implemented. The general appearance of the installation is shown in Fig. 2. The studied samples are moved with the help of a mechanical system made on the basis of an electric motor of the KD6-4 type (Fig. 3). The minimum sample is 10x10x1 mm and the maximum is 80x80x5 mm, for this size range the results of the study are correct.
Figure 2 - Appearance of the laboratory identification system
The response signal resulting from the interaction of the signal emitted by the transmission coil with a metal object carries information that depends on its electrical and magnetic properties, and on the basis of which a response signal specific to a particular metal is formed. This signal is not an absolute characteristic of a particular metal, but depends on a number of technical parameters of the system, such as the speed and stabil-
ity of the relative movement of the OK and the coil system. Therefore, there is a need to take into account these parameters and stabilize the trajectory of the coil relative to the object. Rotating laboratory installation (Fig. 3), which allows changing the relative speed of movement of metal samples and the antenna in a plane parallel to the antenna in the range from 2 to 8 m/s.
%
^^ у
>J
Figure 3 - Rotating installation: the process of controlling metal samples
The feedback signal at the input of the phase detector, which is the main element of the control system that forms the investigated signal, depends on the position of the OK relative to the coil block. Without its fixation, the signal-echo is a relative characteristic of the metal, which becomes absolute only after comparing the signals obtained from different metals, taken at the same relative speed of movement, and creating a database of signals, which is necessary for their mutual comparison. Write the feedback signal: V
U 3 = U130 cos (— t)cos(ffl t + ф(?)) L
.(6)
where
фО) = ф,
Д ф
ф мах
t , фп
- the in-
itial value of the phase shift, m
A t m ax
value of the initial phase shift.
- the maximum Дф - initial phase
change step, Ui3 - voltage in the receiving coil, taking
U,
13 0
V - linear L
into account the movement of the metal OK
the basic value of the voltage on the coil
speed of OK movement along the turns of the coil
- the distance between the receiving and transmitting coils.
The signal U3 is fed to the input of the synchronous detector [11], at the output of which there is a signal corresponding to the phase shift between the reference
(probe) signal and the signal presented in the receiving antenna.
The signal at the output of the synchronous phase detector Ua corresponds to the nature of the change in phase and amplitude:
0 fз ^ fо пор
U a =
1 П -с/s 2
со sep /,=/<
о п ор
U (V, а, ц) ^[Ф]
where 0 - a nonlinear filter, after passing through which information coefficients are extracted.
where Ua - voltage at the output of the synchronous phase detector, U3 - voltage at the input of the phase detector, f - frequency of the signal at the input of the phase detector, fonop - frequency of the reference signal (the frequency emitted by the coils system), ф -phase shift between signals, Uonop - reference signal voltage.
Next, the Ua signal is converted into a digital code and processed in the electronic unit by digital methods proposed in [2], which allows identification of the metal from which the control object is made.
The purpose of processing is to extract information coefficients characterizing a specific metal or alloy. These are the number of extremes, their mutual relationship, the area of the spectrum under the contour, the curvature of the spectrum contour, the lower and upper limits of the spectra. They can be imagined as a row matrix f (K%, Ecum , КОР, Sxs, fn, fv) .
Schematically, the identification process can be imagined as:
f (K %, E CUm , КОР, S xS, fn, fv ) ,
At different speeds of movement of the OK relative to the coil, the shape of the signal will be different, accordingly, the informative coefficients will also be
different. Correlation of coefficients among themselves for different metals will lead to better or worse identification. Therefore, it is appropriate to determine the optimal speeds (rotation frequencies) of the rotary installation, which will ensure the maximum possible accuracy of identification of the OK material.
Previous results of row matrices did not contain information about the speed of movement of the OK relative to the coil system, because the filter extracts only information coefficients. Therefore, it was appropriate to investigate the effect on the response signal of the speed of movement of the OK relative to the antenna system.
The classic analysis of finding extremums [13,14], which involves calculating the derivative of one of the variables and equating the result to zero, is not acceptable, because the criterion for maximum information between metals is a row matrix, not an initial function.
In order to find the optimal speed of movement of metal samples above the coil, we suggest to alternately compare the derivative functions of the response signal by graphical and integral method.
The maximum graphic-integral difference of the derivatives for two different metals is a confirmation of the maximum correlation between the parameters of the metals being compared. Schematically, this method can be written as:
J
UX(V, a, |a) U 2(V, a, |a)
ô V V
ô V
max
U j(V, a, U13 ocos(-1 )cos(© t + 9(t))
ô V
t V
--U130 sin (— t) c o s (© t + 9(t))
L L
ô V
The values of the integral characteristics for 4 different metals are calculated in the work. They are presented in Fig. 6 (on the y-axis - area, graphically - integral difference). They show the cross-correlation characteristics for the range of rotation frequencies of the
moving part of the rotary installation from 10 to 18 Hz, which corresponds to the speed of linear movement of the OK relative to the antenna system of 2...8 m/s.
2500
2000
500
Ag-Cu Ag-Fe
—10 Hz —»
Ag-Ni Cu-Fe Cu-Ni Ni-Fe
12 Hz ••£••14 Hz 16 Hz 18 Hz
Figure 6. - Diagram of cross-correlation integral characteristics of 4 different metals
As the results of the experiments showed, at a significantly low rotation frequency, the informational effect is poorly manifested - the form of feedback signals is similar and does not carry information about the type of metals. And at too high a frequency, the responses turn into one extreme, which can be explained by insufficient charging of the OK by the secondary field.
The analysis of the diagram shows that there is no clearly defined optimal frequency, but there is a frequency range of 12...14 Hz, which provides the maximum correlation integral characteristic.
CONCLUSIONS
The proposed model for evaluating the results of the comparison of controlled metals is based on an integral-differential approach to determining the correlation difference of information criteria of spectral and
graphic-digital methods of identification of metal objects, allowed to experimentally determine the range of the optimal speed of movement of a metal object relative to the coil system, allowed to optimize the operation of the eddy current dynamic control method, which in turn significantly increased the accuracy of metal identification.
The obtained experimental results showed that the maximum of the correlation difference of the information parameters of the proposed eddy current dynamic system, for a rotary installation with a shoulder of 500 mm, lies in the range of 12...14 Hz, which corresponds to the range of linear velocities of 4...6 m/s.
References
1. Abramovych A. O. Eddy-Current AmplitudePhase Based Method for Identifying Conductive (Metal) Objects / A. O. Abramovych, V. O. Poddubny //International scientific and technical journal" Metal-lophysics and the latest technologies". Kyiv. - 2020. -T. 42. - №. 8. - C. 1069-1085. DOI: https://doi.org/10.15407/mfint.42.08.1169
2. Abramovych A. O. The remote eddy-current analysis of a composition of metal objects / A. O. Abramovych, I. S. Kashirsky, V. O. Piddubniy // Metallophysics and advanced technologies . - 2017.-Vol.39, №8. - P. 1035-1049. DOI: https://doi.org/10.15407/mfint.39.08.1035
3. Bruschini C. A Multidisciplinary analysis of frequency domain metal detectors for humanitarian demining: thesis dissertation of Doctor in Applied Sciences / Claudio Bruschini.- Brussels: Vrije Universiteit Brussel, 2002.- 242 p.
4. Voprosy podpoverkhnostnoy radyolokatsyy /A. Yu. Hrynev Moskva: Radyotekhnyka: 2005 (in Russian).
5. Nerazrushayushchiy kontrol'. V 5-ti knigakh. Kniga 3. Elektromagnitnyy kontrol' / V. V. Sukhorukov Moskva: Vysshaya shkola: 1992 (in Russian).
6. Kontrol' nerazrushayushchiy vikhretokovyy. Terminy i opredeleniya, (GOST 24289-80) (in Russian).
7. Nakladnye i ekrannye datchiki (dlya kontrolya metodom vikhrevykh tokov) / V. S. Sobolev, Yu. M. Shkarlet Novosibirsk: Nauka: 1967. - 144p (in Russian).
8. Abramovych A. O. Rationing of signals of eddy
- current converters for correct comparison of them / A. O. Abramovych, V. O. Poddubny // Bulletin of the Ternopil National Technical University. Ser: Technical sciences. -Ternopil. - 2017. - Volume 86 (№2). - P.76-83.
9. Abramovych A. O. Solenoid antenna of radio engineering system of metal identification / A. O. Abramovych, V. O. Piddubnyi // The Journal of Zhytomyr State Technological University" / Engineering. - 2019. - Vol. №1(83). - P. 188-196. DOI: https://doi.org/10.26642/tn-2019-1(83)-188-196
10. Abramovych A. O. Radio engineering system identification of metals on the basis of eddy-current converters / A. O. Abramovych, Y. S. Agalidi, V. O. Piddubnyi // Scientific Bulletin of Zaporizhzhya National Technical University, Radioelektronika, Informatics, Management. - 2020. - №1. - P. 7-17. DOI: https://doi.org/10.15588/1607-3274-2020-1-1
11. Poluprovodnikovaya skhemotekhnika: Spravochnoe rukovodstvo. / U. Tittse, K. Shenk Moskva: Mir: 1982. - 450p (in Russian).
12. Salih M. Fourier Transform - Signal Processing / M. Salih. - Rijeka, Croatia.: InTech, 2012.
- 366p. ISBN 978-953-51-0453-7.
13. Ifeachor E. Digital Signal Processing: A Practical Approach 2nd Edition / E. Ifeachor, B. Jervis.
- Hoboken, USA.: Prentice Hall, 2001. - 933p. ISBN 978-0201596199
14. Stroustrup B. Programming: principles and practice using C++ (2nd Edition) / Bjarne Stroustrup. -Boston, USA.: Addison-Wesley, 2014. - 2339p.-ISBN 978-0321-992789.
