USE OF A ROLLED DEEP NEURAL NETWORK TO DETERMINE THE CHANGE OF THE STRESSED AND DEFORMED STATE OF VERTICAL STEEL CYLINDERAL RIVER

Authors

  • Khristina Pankiv
  • Yurii Pankiv

DOI:

https://doi.org/10.31471/1993-9981-2020-2(45)-5-12

Keywords:

deep neural network; convolution; puling; cross entropy.

Abstract

The main place in determining the mechanical characteristics of the RVS material is occupied by methods and means of VAT control, which include the method of coercive force; magnetic anisotropy method; Barkhausen method, magnetic metal memory (MPM) method, strain gauge method. The advantages of magnetic methods are the ability to control without decommissioning the RVS, safety. The disadvantages inherent in almost all magnetic control methods include the need to prepare the controlled surface, the difficulty of determining the position of the sensors in relation to the maximum loads, the dependence of control results on methods and conditions of measurement, the influence of the air layer between the sensor and the controlled surface.It's shown that to determine the further safe operation possibility of vertical steel cylindrical tank it is necessary to know their stress-strain state. The shortcomings of the existing experimental and mathematical methods of its estimation were highlighted. It's proposed to use a convolutional deep neural network to determine the stress state of a vertical steel cylindrical tank. As input data, it's proposed to use data on the movement of its wall obtained, for example, as a result of geometric calibration at two points in time. The input data was presented in the form of an array of dimensions 8x12, then used convolution and max-pulling. The last layer is fully connected. It's proposed to use cross-entropy as a cost function. To increase the amount of training data, it is proposed to use the values of displacements on stresses obtained by modeling different effects on a cylindrical tank with different shape defects using the SolidWorks package. Possible ways to improve the proposed method are proposed.For further research and improvement of the proposed method, you can try to use other hyper parameters in the neural network, in particular to change the number of feature maps, the size of the local receptive field, the size of the shift step of the receptive field and others.You can also try using the source layer with a different number of neurons and softmax as a function of cost.

Downloads

Download data is not yet available.

References

Kandakov G. P. Problemy otechestvennogo rezervuarostroeniya i vozmozhnye puti ikh resheniya [Tekst] / G. P. Kandakov // Promyshlennoe i grazhdanskoe stroitelstvo. – 1998. – № 5. – S.24-26.

Ibragimov I.G. Monitoring sostoyaniya obolochkovykh konstrukcij metodom magnitnogo skanirovaniya [Elektronnij resurs] / I. G. Ibragimov, R. G. Vildanov // Neftegazovoe delo. – 2004. – Rezhim dostupu do zhurn. : http://www.ogbus.ru.

Razrusheniya v processe ehkspluatacii vertikalnykh cilindricheskikh rezervuarov so stacionarnoj kryshej [Elektronnij resurs] / S. M. Kupreishvili // S. : Khimstalkon. – Rezhim dostupu : http://www.himstalcon.ru/node/2582.

Nastanova z provedennia tekhnichnoho diahnostuvannia vertykalnykh stalevykh rezervuariv [Tekst] : DSTU–N B A.3.1–10:2008. – [Chynnyi vid 2009–07–01]. – K.: Minrehionbud Ukrainy, 2009. – 63 s. – (Natsionalnyi standart Ukrainy).

Zagorodnev V. I. Ostorozhno, svarka [Tekst] / V. I. Zagorodnev // Industriya. – 2005. – № 3 (41). – S. 2–3.

Systema zabezpechennia nadiinosti ta bezpeky budivelnykh obiektiv. Zahalni pryntsypy zabezpechennia nadiinosti ta konstruktyvnoi bezpeky budivel, sporud, budivelnykh konstruktsii ta osnov. [Tekst] : DBN V.1.2-14-2009. [Chynnyi vid 2009-01-12]. – K. : Minrehionbud Ukrainy, 2009. – (Derzhavni budivelni normy Ukrainy)

Kontrol napryazhenno–deformirovannogo sostoyaniya oborudovaniya i konstrukcij pri ocenke ostatochnogo resursa na obektakh promyshlennosti i transporta [Elektronnij resurs] / OOO «EhnergodiagnostikA». – Reutov. – Rezhim dostupu : http://www.energodiagnostika.ru/ru/application_mmm/app_mmm_sss_inspection.aspx.

Nikitina N. E. Preimushchestva metoda akustouprogosti dlya nerazrushayushchego kontrolya mekhanicheskikh napryazhenij v detalyakh mashin [Tekst] / N. E. Nikitina, S. V. Kazachek // Vestnik nauchnO–tekhnicheskogo razvitiya. – 2010. – № 4 (32). – S. 18–28.

Konstruktsii budivel i sporud. Stalevi konstruktsii. Normy proektuvannia, vyhotovlennia i montazhu. [Tekst] : DBN V.2.6-163 2010. – [Chynnyi vid 2011-01-12]. – K. : Minrehionbud Ukrainy, 2011. – (Derzhavni budivelni normy Ukrainy)

Vlasov V. Z. Obshchaya teoriya obolochek i ee prilozhenie v tekhnike [Tekst] : ucheb. posobie / V. Z. Vlasov. – M.: Gostekhizdat, 1979. – 784 s.

Streng G. Teoriya metoda konechnykh ehlementov [Tekst] : ucheb. posobie / G. Streng, Dzh. Fiks. ; perev. s angl. V. I. Agoshkova, V. A. Vasilenko, V. V. Shajdurova. – M.: Mir, 1977. – 349 s.

Rezervuary vertykalni tsylindrychni stalevi dlia nafty ta naftoproduktiv. Zahalni tekhnichni umovy (HOST 31385–2008, NEQ) [Tekst] : DSTU B V.2.6–183:2011. – [Chynnyi vid 2012–10–01]. – K. : Minrehion Ukrainy, 2012. – 77 s. – (Natsionalnyi standart Ukrainy).

Fomin A. V. Raschetno-ehksperimentalnye metody mekhaniki deformiruemogo tela v usloviyakh ogranichennoj iskhodnoj informacii [Tekst] : dis. na soiskanie uchenoj stepeni dokt. tekhn. nauk / A. V. Fomin. – M. : 1989.

Prejs A. K. Opredelenie napryazhenij v ob"eme detali po dannym izmerenij na poverkhnosti [Tekst] / A. K. Prejs. – M.: Nauka, 1979. – 128s.

Zamikhovskyi L. M. Matematychne modeliuvannia napruzheno-deformovanoho stanu rezervuariv z defektamy formy stinky [Tekst] / L. M. Zamikhovskyi, Kh. V. Pankiv // Visnyk Khmelnytskoho natsionalnoho universytetu. – Khmelnytskyi, 2007. – T.1 (93), № 3. – S. 212–214.

Zamikhovskyi L. M. Otsinka napruzheno-deformovanoho stanu vertykalnykh stalnykh tsylindrychnykh rezervuariv za peremishchenniamy tochok yikh poverkhni [Tekst] / L. M. Zamikhovskyi, Kh. V. Pankiv // Visnyk KDPU. – 2007. – Chast. 1, Vyp. 4. – S. 141–143.

Michael A. Nielsen, "Neural Networks and Deep Learning", Determination Press, 2015 [Elektronnyi resurs] – Rezhym dostupu : http://neuralnetworksanddeeplearning.com.

Metrolohiia. Rezervuary stalevi vertykalni tsylindrychni. Metodyka povirky (HOST 8.570–2000. MOD) [Tekst] : DSTU 4147–2003. – [Chynnyi vid 2003–01–01]. – K.: Derzhspozhyvstandart, 2003. – 74 s. – (Natsionalnyi standart Ukrainy).

Tiku SH. Ehffektivnaya rabota: SolidWorks 2004 : ucheb. posobie [Tekst] / Tiku SH. – SPb.: Piter, 2005. – 768 s. – ISBN 5–94723–841.

Martyniuk Kh. V. Metod otsinky napruzheno-deformovanoho stanu vertykalnykh stalnykh tsylindrychnykh rezervuariv [Tekst] / Kh. V. Martyniuk, L. M. Zamikhovskyi // Эffektyvnost realyzatsyy nauchnoho, resursnoho, promыshlennoho potentsyala v sovremennыkh uslovyiakh : 7-ma shchorichna mizhnar. prom. konf. i blits-vystavka 12-16 liutoho 2007 r.: materialy konferentsii. – s. Slavske, 2007. – S. 321-322.

Published

2021-04-12

How to Cite

Паньків, Х., & Паньків, Ю. (2021). USE OF A ROLLED DEEP NEURAL NETWORK TO DETERMINE THE CHANGE OF THE STRESSED AND DEFORMED STATE OF VERTICAL STEEL CYLINDERAL RIVER. METHODS AND DEVICES OF QUALITY CONTROL, (2(45). https://doi.org/10.31471/1993-9981-2020-2(45)-5-12

Issue

Section

METHODS AND EQUIPMENT OF NON-DESTRUCTIVE CONTROL