Assessment of Technical Condition of Polyurethane Foam Thermal Insulation Pipelines of Heating Networks Using Neural Network Technologies

  • Authors

    • Emil R. Saifullin
    • Shamil G. Ziganshin
    • Yury V. Vankov
    • Airat R. Zagretdinov
    2018-09-27
    https://doi.org/10.14419/ijet.v7i4.7.20556
  • diagnostics, corrosion, defect, pipelines, acoustic signal, heat networks, finite element model, neural network

  • Abstract

    Pipelines of heat networks are an important element of heat supply to cities and industrial facilities. To increase the reliability of the operation of pipelines of heating networks, reducing the number of their accidents and increasing the economic parameters of transportation of heat energy, it is required to constantly increase the volumes and quality of complex diagnostics. The instruments currently used for the diagnosis of pipelines have many shortcomings. Among them, low reliability of detection of defects and subjectivity of decision-making, as well as lack of funds for diagnostics of pre-insulated pipelines (in polyurethane foam insulation). To simplify, accelerate and improve the reliability of monitoring the technical condition of pipelines, the authors set the goal of diagnosing the object of research using acoustic methods, using neural network technologies to process acoustic signals. The article describes experimental studies of pipelines of heating networks in polyurethane foam insulation with various sizes of defects and an analysis of the acoustic signals obtained at the same time is made. The frequency of natural oscillations of the pipeline is chosen as the determining parameter of the acoustic signal. To process and analyze the frequencies obtained as a result of the experiments, a neural network of back propagation of the error was constructed.

    The results of the classification of the neural network of back propagation of the error trained by the neural network showed its good ability to analyze unknown samples and a high degree of reliability of their recognition.

     

     

     

  • References

    1. [1] Yu.V. Vankov, Influence of defects on natural oscillation frequencies of gas turbine engine blades. Izvestiya Vysshikh Uchebnykh Zavedenij. Aviatsionnaya Tekhnika, (3), 2003, pp. 69-71.

      [2] E.V. Belov, Yu.V. Vankov, I.V. Ivshin, A.V. Kochergin, D.N. Pervukhin, A.P. Tunakov. Diagnostics of Gas-Turbine Engines by Acoustic Sensing of Their Components. Acoustical Physics, 42 (1), 1996, pp. 14-17.

      [3] Yu.V. Vankov, R.B. Kazakov, , D.N. Pervukhin. Diagnostic system for flaw detection in the railroad car axles by acoustic parameters. Zavodskaya Laboratoriya. Diagnostika Materialov, 70 (3), 2004, pp. 34-38.

      [4] Yu.V. Vankov, A.E. Kondratev. Acoustic spectral defectoscope for the detection of defects in composite materials (Part 1). Pribory i Sistemy Upravleniya, (1), 2004, pp. 45-53.

      [5] Yu.V. Vankov. Acoustic spectral defectoscope for defects detection in composite materials (part 2). Pribory i Sistemy Upravleniya, (2), 2004, pp. 53-58.

      [6] Yu.V. Vankov, Sh.G. Ziganshin, E.V. Izmailova, A.R. Zagretdinov. The effect of the size and position of transverse cracks in the natural frequencies of the rods. IOP Conference Series: Materials Science and Engineering, 86 (1), art. no. 012036, 2015.

      [7] Yu.V. Vankov, Sh. G. Ziganshin, E. V. Izmailova, T.O. Politova. Methods check the piping using a neural network. IOP Conference Series: Materials Science and Engineering, 86 (1), no. 012037, 2015.

      [8] Sh.G.Ziganshin, Yu.V. Vankov, T.G. Gorbunova. Reliability of Thermal Networks for City Development Original Research Article. Procedia Engineering, vol 150, 2016, pp. 2327-2333

      [9] Yu.V. Vankov, Sh.G. Ziganshin, E.V. Izmailova, V.V. Serov. Determination of the oscillation frequencies of corrosion defects finite element methods in order to develop methods of acoustic monitoring of pipelines. IOP Conference Series: Materials Science and Engineering , 86 (1), no. 012035, 2015.

      [10] T.O. Politova, Sh.G. Ziganshin, R.R. Salyakhova, A.O. Malakhov. Correlation analysis of the factors influencing the failures of pipelines of heat networks. Engineering journal of Don. 2015, â„– 3.

      [11] S.G. Ziganshin, V.V. Serov, Yu.V. Vankov, E.V. Izmaylova. Free Oscillation Technique for Pipeline System Inspection. NDT World 2013. No. 3(61), pp. 56-60.

      [12] Yu. Ya. Petrushenko, Yu. V. Vankov, Sh. G. Ziganshin, V.N. Tyryshkin. Vibrating-acoustic method and diagnostic complex for definition of defects of pipelines with use of the neural network. Proceedings of the higher educational institutions. Energy sector problems. 2009, no. 9–10, pp. 149–154.

      [13] Yu. Ya. Petrushenko, Yu. V. Vankov, Sh. G. Ziganshin, V.V. Serov. Definition of informative parameters of defects of pipelines by the finite element method. Proceedings of the higher educational institutions. Energy sector problems. 2009, no. 7–8, pp. 149–154.

      [14] Yu. V. Vankov, V.V. Serov, Sh. G. Ziganshin, E.V. Izmaylova. Studying of influence of corrosion defects on parameters of fluctuations of pipelines at the early stage of origin. Proceedings of the higher educational institutions. Energy sector problems. 2011, â„– 11-12, pp. 146-154.

      [15] Yu.V. Vankov, Sh.G.Ziganshin, T.G. Gorbunova, T.O. Politova, R.M. Khabibullin. Analysis of damage to heating systems Kazan and develop recommendations to improve reliability. Proceedings of the higher educational institutions. Energy sector problems. 2012, â„– 7-8, pp. 9-18.

      [16] Sh.G. Ziganshin, Yu.V. Vankov, E.V. Izmailova. Control of the technical state of pipelines by acoustic. Kazan, Publishing house of Kazan University, 2015.

      [17] Yakhyayeva G. E. Fuzzy sets and neural networks //Moscow: Internet University of information technologies. - 2006.

      [18] Haikin S. Neural networks: full course, 2nd edition. - Williams publishing house, 2008.

      [19] Callan R. Basic concepts of neural networks = the Essence of Neural Networks First Edition. - Moscow: Williams, 2001. - 288 p.

      [20] Cherkasov Yuri. The learning algorithm of multilayer neural network method with back propagation error// habrahabr.ru/post/198268/-2013

  • Downloads

  • How to Cite

    R. Saifullin, E., G. Ziganshin, S., V. Vankov, Y., & R. Zagretdinov, A. (2018). Assessment of Technical Condition of Polyurethane Foam Thermal Insulation Pipelines of Heating Networks Using Neural Network Technologies. International Journal of Engineering & Technology, 7(4.7), 241-244. https://doi.org/10.14419/ijet.v7i4.7.20556

    Received date: 2018-09-29

    Accepted date: 2018-09-29

    Published date: 2018-09-27