Prediction of gas mixture reactivity based on detonation pipe vibrations

• Autorzy: Marcinkiewicz K. , Żbikowski M. , Lesiak P. , Teodorczyk A.
• Języki publikacji: EN

Abstrakty

EN

The aim of this paper is to make an approach of creation a machine learning system predicting gas mixture composition being burned in a process pipe, based on pipe vibrations measurements. Task is divided into two parts: performing an experiment to get a necessary experimental data, and developing prediction algorithm. First, the basic principles of machine learning and signal processing are presented. Machine learning is the subfield of computer science that focuses on creating algorithms that can learn from provided data and perform predictions, either classification or regression. Signal processing is a general statement for all activities performed on information in form of a signal. In this particular work the emphasis is put on Fourier transform. After introduction, a brief description of the pipe response to internal detonation and pressure load is provided. It is of most significance, since the sensors used in the experiment base on pipe vibrations. Finally, the experimental part is described. The experiment consisted of performing a series of hydrogen-air explosion in pipes, with various hydrogen concentration. Measurement is performed with three sensors: piezoelectric sensor, knock combustion sensor - both measuring vibrations of pipe - and a pressure sensor, measuring pressure. This data is fed to a machine learning algorithm, that works as follows: first, measurement from a sensor is interpolated using b-splines. Then it transposes data from time domain to frequency domain using Fourier transform. Afterwards it is merged into one array. The set is divided into training and scoring sets, using cross-validation techniques. Training sets are used to feed classificator: SVM, SGD, naive Bayes, logistic regression, linear SVC, Ada Boost, perceptron. From this algorithm the prediction score of each classificator is derived and arranged with each other. It appears, that the algorithms used in conjunction with piezoelectric sensor give the score averaging to 50 %. The analysis of frequency spectrum is needless, since there is not enough features. The best classifiers are Perceptron, Naive Bayes and Support Vector Machine. Data from pressure sensor give much better results, with accuracy even up to 90 %. Fourier transform boosts the accuracy of classifiers. The best one is logistic regression. Therefore prediction of gas mixture reactivity based on detonation pipe vibration is possible.

Słowa kluczowe

EN

detonation; machine learning; hydrogen; explosion; vibration; pipe; classification

• Wydawca: Komitet Termodynamiki i Spalania PAN
• Czasopismo: Archivum Combustionis
• Rocznik: 2017
• Tom: Vol. 37 nr 1
• Strony: 43--77
• Opis fizyczny: Bibliogr. 27 poz., rys., tab.

Twórcy

autor

Marcinkiewicz K.

• Warsaw Uniwersity of Technology Institute of Heat Engineering Nowowiejska 21/25, Warsaw 00-665, Poland

autor

Żbikowski M.

• Warsaw Uniwersity of Technology Institute of Heat Engineering Nowowiejska 21/25, Warsaw 00-665, Poland

autor

Lesiak P.

• Scientific and Research Centre for Fire Protection National Research Institute Aleja Nadwislanska 213, 05-420 Jozefow, Poland

autor

Teodorczyk A.

• Warsaw Uniwersity of Technology Institute of Heat Engineering Nowowiejska 21/25, Warsaw 00-665, Poland

Bibliografia

• [1] Mallard E., Chatelier H.L.: Combustion des melanges gazeaux explosifs, Annales des Mines 8, 1883, pp. 971-982.
• [2] Mannan S.: Lees’ Process Safety Essentials., Waltham 2014. Elsvier Inc.
• [3] Zeldovich Ya. B.: On the theory of the propagation of detonation in gaseous systems. Sov. Phys. JETP, 10, 1940, pp. 542-568.
• [4] Neumann J.: Report on ”theory of detonation waves” (OD-2). Technical report, National Defence Research Committee of the Office if Scientific Research and Development Devision B, Section B-1, 1942.
• [5] Doring W.: Uber der detonation vergang in gasen. Ann. Phys., 46, 1946, pp. 421–436.
• [6] Markstein G.H.: A shock-tube study of flame front-pressure wave interaction, Proceedings of the Combustion Institute, 6, 1956, pp. 387–398.
• [7] Urtiew P.A, Oppenheim A.K.: Experimental observations of the transition to detonation on an explosive gas, Proceedings of the Royal Society of London Series A, 295, 1966, pp. 13–28.
• [8] Lee J.H., Knystautas R., Yoshikawa N.: Photochemical initiation of gaseous detonations, Acta Austronautica 5, 1977, pp. 971–982.
• [9] Sivashinsky G.I.: Some developments in premixed combustion modelling, Proceedings of the Combustion Institute, 29, 2002, pp.1737-1761.
• [10] Dorofeev S.B.: Flame acceleration and DDT in gas explosions, Journal de Physique IV France, 12, 2002, pp. 3–10.
• [11] Liang Z., Karnesky J., Shepherd J.E.: Structural response to reflected detonations and deflargation-todetonation transition in H2 − N2O Mixtures, Explosion Dynamics Laboratory Report FM2006.003: 2006.
• [12] Beltman W.M., Shepherd J.E.: Linear elastic response of tubes to internal detonation loading, Journal of Sound and Vibration, 252, 2002, pp. 6–17.
• [13] Karnesky J., Damazo J.E., Shepherd J.: Plastic Response of thin-walled tubes to detonation, Pressure Vessels and Piping Division Conference, 2010.
• [14] Karnesky J., Damazo J., Chow-Yee K., Rusinek A., Shepherd J.E.: Plastic Deformation due to Reflected Detonation, International Journal of Solids and Structures, 1, 2013, pp. 97-110.
• [15] U.S. Departament of Transportation. Pipeline Incident 20 Year Trends – significant incident 20 year trend. http://opsweb.phmsa.dot.gov/primis_pdm/significant_inc_trend.asp [Online; accessed 19December-2016].
• [16] Cooley J.W., Tukey J.W.: An algorithm for the machine calculation of complex Fourier series, Mathematics of Computation, 19, 1965, pp. 297–301.
• [17] Dongarra J., Sullivan F.: The Top 10 Algorithms, Computing ins Science & Engineering, 2, 2000, pp. 22-23.
• [18] Prautzsch H., Boehm W., Paluszny M.: Bezier and B-Spline Techniques, New York 2002. Springer.
• [19] James, G., Witten, D., Hastie, T., Tibshirani, R.: An Introduction to Statistical Learning. New York 2013. Springer.
• [20] Hastie T., Tibshirani, R., Friedman J.: The Elements of Statistical Learning, New York 2009. Springer.
• [21] Freund Y., Schapire R. E.: A Decision-Theoretic Generalization of On-Line Learning and an Application to Boosting, Journal of computer science and system sciences, 55, 1977, pp. 119–139.
• [22] Shepherd J. E.: Pressure loads and structural response of the BNL high-temperature detonation tube, US Nuclear Regulatory Commission 1992.
• [23] de Malherbe M.C., Wing R.D., Laderman A. J., Oppenheim A.K. : Response of a Cylindrical Shell to Internal Blast Loading, Journal of Mechanical Engineering Science, 8, 1966, pp. 93–98.
• [24] Shepherd J.E.: Structural Response of Piping to Internal Gas Detonation, Preceedings of PVP2006ICPVT-11, 2006.
• [25] Shepherd J.E., Pintgen F.: Elastic and Plastic Structural Response of tubes to Deflargation-to- Detonation Transition, Explosion Dynamics Laboratory Report FM2006-005, 2006.
• [26] Pingten F., Liang Z., Shepherd J.E.: Structural Response of Tubes to Deflagration-to-Detonation Transition, ICDERS, 21, 2007, pp. 23-27.
• [27] Piezo Film Sensors: Technical Manual, Norristown 2008. Measurement Specialties, Inc. 25.

Uwagi

PL

Opracowanie rekordu w ramach umowy 509/P-DUN/2018 ze środków MNiSW przeznaczonych na działalność upowszechniającą naukę (2018).