Performance of artificial neural networks in an inverse problem of laser beam diagnostics

• Autorzy: Pietrak Karol , Muszyński Radosław , Marek Adam , Łapka Piotr

Identyfikatory

DOI

10.24425/bpasts.2022.140100

• Języki publikacji: EN

Abstrakty

EN

The presented results are for the numerical verification of a method devised to identify an unknown spatio-temporal distribution of heat flux that occurs at the surface of a thin aluminum plate, as a result of pulsed laser beam excitation. The presented identification of boundary heat flux function is a part of the newly proposed laser beam profiling method and utilizes artificial neural networks trained on temperature distributions generated with the ANSYS Fluent solver. The paper focuses on the selection of the most effective neural network hyperparameters and compares the results of neural network identification with the Levenberg–Marquardt method used earlier and discussed in previous articles. For the levels of noise measured in physical experiments (0.25–0.5 K), the accuracy of the current parameter estimation method is between 5 and 10%. Design changes that may increase its accuracy are thoroughly discussed.

Słowa kluczowe

EN

inverse problems; pattern recognition; laser beams

PL

odwrotne problemy; rozpoznawanie wzorców; wiązka laserowa

• Wydawca: Polska Akademia Nauk, Wydział IV Nauk Technicznych
• Czasopismo: Bulletin of the Polish Academy of Sciences. Technical Sciences
• Rocznik: 2022
• Tom: Vol. 70, nr 1
• Strony: art. no. e140100
• Opis fizyczny: Bibliogr. 61 poz., rys., tab.

Twórcy

autor

Pietrak Karol

• Faculty of Power and Aeronautical Engineering, Warsaw University of Technology, ul. Nowowiejska 24, 00-665 Warsaw, Poland

• https://orcid.org/0000-0003-4723-1263

autor

Muszyński Radosław

• Faculty of Power and Aeronautical Engineering, Warsaw University of Technology, ul. Nowowiejska 24, 00-665 Warsaw, Poland

autor

Marek Adam

• Faculty of Power and Aeronautical Engineering, Warsaw University of Technology, ul. Nowowiejska 24, 00-665 Warsaw, Poland

autor

Łapka Piotr

• Faculty of Power and Aeronautical Engineering, Warsaw University of Technology, ul. Nowowiejska 24, 00-665 Warsaw, Poland

• https://orcid.org/0000-0003-3039-9588

Bibliografia

• [1] A.K. Dubey and V. Yadava, “Laser beam machining – A review”, Int. J. Mach. Tools Manuf., vol. 48, no. 6, pp. 609–628, 2008.
• [2] Q. Peng et al., “Lasers in medicine”, Reports Prog. Phys., vol. 71, no. 5, p. 056701, May 2008.
• [3] G.P. Perram, M.A. Marciniak, and M. Goda, “High energy laser weapons: technology overview”, Proc. SPIE, vol. 5414, pp. 1–25, 2004.
• [4] K. Pietrak and T.S. Wi ́sniewski, “Methods for experimental determination of solid-solid interfacial thermal resistance with application to composite materials”, J. Power Technol., vol. 94, no. 4, pp. 270–285, 2014.
• [5] L. Bruno, “Mechanical characterization of composite materials by optical techniques: A review”, Opt. Lasers Eng., vol. 104, pp. 192–203, May 2018.
• [6] C.B. Roundy, “Current technologies of laser beam profile measurements”, in Laser Beam Shaping, F.M. Dickey and S.C. Holswade, Eds. New York: CRC Press, 2000.
• [7] J.A. Hoffnagle and C.M. Jefferson, “Design and performance of a refractive optical system that converts a Gaussian to a flattop beam”, Appl. Opt., vol. 39, no. 30, pp. 5488–99, Oct. 2000.
• [8] J.L. Guttman, “Noninterceptive Beam Profiling of High-Power Industrial Lasers”, Laser Tech. J., vol. 12, no. 5, pp. 20–23, Nov. 2015.
• [9] M. Kujawi ́nska, P. Łapka, M. Malesa, K. Malowany, M. Prasek, and J. Marczak, “Investigations of high power laser beam interaction with material by means of hybrid FVM-FEM and digital image correlation methods,” in Proc. SPIE 10159, 2016, p. 1015916.
• [10] O. Aharon, “High power beam analysis”, Ind. Laser Solut. Manuf., vol. 28, no. 5, pp. 17–20, 2013.
• [11] M.A. Sutton, J.J. Orteu, and H. Schreier, Image Correlation for Shape, Motion and Deformation Measurements. Springer US, 2009.
• [12] K. Pietrak, P. Łapka, and M. Kujawińska, “Inverse analysis for the identification of temporal and spatial characteristics of a short super-Gaussian laser pulse interacting with a solid plate”, Int. J. Therm. Sci., vol. 134C, pp. 585–593, 2018.
• [13] M.N. Ozisik and H.R.B. Orlande, Inverse heat transfer. New York: Taylor&Francis, 2000.
• [14] P. Łapka, K. Pietrak, M. Kujawi ́nska, and M. Malesa, “Development and validation of an inverse method for identification of thermal characteristics of a short laser pulse”, Int. J. Therm. Sci., vol. 150, Apr. 2020.
• [15] J.V. Beck and K.A. Woodbury, “Inverse problems and parameter estmation: integration of measurements and analysis”, Meas. Sci. Technol., vol. 9, no. 6, pp. 839–847, Jun. 1998.
• [16] J. Taler and D. Taler, “Measurement of Heat Flux and Heat Transfer Coefficient”, in Heat Flux, processes, measurement techniques and applications, G. Cirimele and M. D’Elia, Eds. New York: Nova Science Publishers, Inc, 2012, pp. 1–104.
• [17] J. Taler and D. Taler, “Surface-heat transfer measurements using transient techniques”, in Encyclopedia of Thermal Stresses, R. B. Hetnarski, Ed. Dordrecht Heidelberg New York London: Springer, 2014, pp. 4774–4784.
• [18] M.S. Gadala and S. Vakili, “Assessment of Various Methods in Solving Inverse Heat Conduction Problems”, in Heat Conduction – Basic Research, V. Vikhrenko, Ed. InTech, 2011.
• [19] J. Krejsa, K.A. Woodbury, J.D. Ratliff, and M. Raudensky, “Assessment of strategies and potential for neural networks in the inverse heat conduction problem”, Inverse Probl. Eng., vol. 7, no. 3, pp. 197–213, Jun. 1999.
• [20] X. Wang and M. Yao, “Neural networks for solving the inverse heat transfer problem of continuous casting mould”, in 2011 Seventh International Conference on Natural Computation, 2011, vol. 2, pp. 791–794.
• [21] M. Grabarczyk and P. Furmański, “Predicting the effective thermal conductivity of dry granular media using artificial neural networks”, J. Power Technol., vol. 93, no. 2, pp. 59–66, May 2011.
• [22] K. Goudarzi, A. Moosaei, and M. Gharaati, “Applying artificial neural networks (ANN) to the estimation of thermal contact conductance in the exhaust valve of internal combustion engine”, Appl. Therm. Eng., vol. 87, pp. 688–697, Aug. 2015.
• [23] B. Czél, K.A. Woodbury, and G. Gróf, “Simultaneous estimation of temperature-dependent volumetric heat capacity and thermal conductivity functions via neural networks”, Int. J. Heat Mass Transf., vol. 68, pp. 1–13, Jan. 2014.
• [24] S. Chudzik, S. Grys, and W. Minkina, “The application of the artificial neural network and hot probe method in thermal parameters determination of heat insulation materials Part 1-thermal model consideration”, in Proceedings of the IEEE International Conference on Industrial Technology, 2009.
• [25] S. Chudzik, “Measuring system with a dual needle probe for testing the parameters of heat-insulating materials”, Meas. Sci. Technol., vol. 22, no. 7, p. 075703, Jun. 2011.
• [26] S. Chudzik, “Measurement of thermal diffusivity of insulating material using an artificial neural network”, Meas. Sci. Technol., vol. 23, no. 6, 2012.
• [27] S. Chudzik, W. Minkina, and S. Grys, “The application of the artificial neural network and hot probe method in thermal parameters determination of heat insulation materials. Part 2 – application of the neural network”, 2009, pp. 1–5.
• [28] A. Mirsepahi, L. Chen, and B. O’Neill, “An Artificial Intelligence Solution for Heat Flux Estimation Using Temperature History; A Two-Input/Two-Output Problem”, Chem. Eng. Commun., vol. 204, no. 3, pp. 289–294, Mar. 2017.
• [29] A. Mirsephai, M. Mohammadzaheri, L. Chen, and B. O’Neill, “An artificial intelligence approach to inverse heat transfer modeling of an irradiative dryer”, Int. Commun. Heat Mass Transf., vol. 39, no. 1, pp. 40–45, Jan. 2012.
• [30] A. Mirsepahi, L. Chen, and B. O’Neill, “A comparative artificial intelligence approach to inverse heat transfer modeling of an irradiative dryer”, Int. Commun. Heat Mass Transf., vol. 41, pp. 19–27, Feb. 2013.
• [31] A. Mirsepahi, L. Chen, and B. O’Neill, “A comparative approach of inverse modelling applied to an irradiative batch dryer employing several artificial neural networks”, Int. Commun. Heat Mass Transf., vol. 53, pp. 164–173, Apr. 2014.
• [32] R.-M. Ricardo, H.-L. Juan Manuel, D.-G. Héctor Martín, and P.-V. Arturo, “Use of Artificial Neural Networks for Prediction of Convective Heat Transfer in Evaporative Units”, Ing. Investig. y Tecnol., vol. 15, no. 1, pp. 93–101, Jan. 2014.
• [33] R. Romero-Méndez, P. Lara-Vázquez, F. Oviedo-Tolentino, H.M. Durán-García, F.G. Pérez-Gutiérrez, and A. Pacheco-Vega, “Use of Artificial Neural Networks for Prediction of the Convective Heat Transfer Coefficient in Evaporative Mini-Tubes”, Ing. Investig. y Tecnol., vol. 17, no. 1, pp. 23–34, Jan. 2016.
• [34] M. Mohanraj, S. Jayaraj, and C. Muraleedharan, “Applications of artificial neural networks for thermal analysis of heat exchangers – A review”, Int. J. Therm. Sci., vol. 90, pp. 150–172, 2015.
• [35] O. Cortés, G. Urquiza, and J.A. Hernández, “Optimization of operating conditions for compressor performance by means of neural network inverse”, Appl. Energy, vol. 86, no. 11, pp. 2487–2493, Nov. 2009.
• [36] J.A. Hernández, D. Colorado, O. Cortés-Aburto, Y. El Hamzaoui, V. Velazquez, and B. Alonso, “Inverse neural network for optimal performance in polygeneration systems”, Appl. Therm. Eng., vol. 50, no. 2, pp. 1399–1406, Feb. 2013.
• [37] Y. El Hamzaoui, B. Ali, J. A. Hernandez, O. Cortés-Aburto, and O. Oubram, “Search for Optimum Operating Conditions for a Water Purification Process Integrated to a Heat Transformer with Energy Recycling using Artificial Neural Network Inverse Solved by Genetic and Particle Swarm Algorithms”, Chem. Prod. Process Model., vol. 7, no. 1, Jan. 2012.
• [38] S. Samarasinghe, Neural Networks for Applied Sciences and Engineering: From Fundamentals to Complex Pattern Recognition. Boca Raton: Auerbach, 2007.
• [39] Engineering ToolBox, “Metals – Melting Temperatures,” 2005. [Online]. Available: https://www.engineeringtoolbox.com/melting-temperature-metals-d_860.html. [Accessed: 28-May-2018].
• [40] Y.A. Cengel and A.J. Ghajar, Heat and Mass Transfer, Fundamentals & Applications, 5th-th ed. New York: McGraw-Hill Education, 2015.
• [41] “FLIR SC7500 instrument specifications”, Flir Systems Inc., 2014. [Online]. Available: https://issuu.com/netservis/docs/flir-sc7500.
• [42] A. Gulli and S. Pal, Deep Learning with Keras. Packt Publishing Ltd, 2017.
• [43] S. Ruder, “An overview of gradient descent optimization algorithms,” Sep. 2016.
• [44] B. Ding, H. Qian, and J. Zhou, “Activation functions and their characteristics in deep neural networks”, in Proceedings of the 30th Chinese Control and Decision Conference, CCDC 2018, 2018, pp. 1836–1841.
• [45] C. Nwankpa, W. Ijomah, A. Gachagan, and S. Marshall, “Activation Functions: Comparison of trends in Practice and Research for Deep Learning,” Nov. 2018.
• [46] D.-A. Clevert, T. Unterthiner, and S. Hochreiter, “Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)”, 4th Int. Conf. Learn. Represent. ICLR 2016 – Conf. Track Proc., Nov. 2015.
• [47] G. Klambauer, T. Unterthiner, A. Mayr, and S. Hochreiter, “Self-Normalizing Neural Networks”, Adv. Neural Inf. Process. Syst., vol. 2017-December, pp. 972–981, Jun. 2017.
• [48] Keras Documentation, “Earlystopping”, 2020. [Online]. Available: https://keras.io/api/callbacks/early_stopping/. [Accessed: 29-Jul-2020].
• [49] R.J. Hyndman and A.B. Koehler, “Another look at measures of forecast accuracy”, Int. J. Forecast., vol. 22, no. 4, pp. 679–688, Oct. 2006.
• [50] J. Brownlee, “How to ConFigure the Number of Layers and Nodes in a Neural Network,” 2018. [Online]. Available: https://machinelearningmastery.com/how-to-conFigure-the-number-of-layers-and-nodes-in-a-neural-network/. [Accessed: 30-Jul-2020].
• [51] N. Srivastava, G. Hinton, A. Krizhevsky, and R. Salakhutdinov, “Dropout: A Simple Way to Prevent Neural Networks from Overfitting,” 2014.
• [52] D. Stathakis, “How many hidden layers and nodes?”, Int. J. Remote Sens., vol. 30, no. 8, pp. 2133–2147, 2009.
• [53] V. Kůrková, “Kolmogorov’s theorem and multilayer neural networks”, Neural Networks, vol. 5, no. 3, pp. 501–506, Jan. 1992.
• [54] G. Cybenko, “Approximation by superpositions of a sigmoidal function”, Math. Control. Signals, Syst., vol. 2, no. 4, pp. 303–314, Dec. 1989.
• [55] W.S. Sarle, “Should I normalize/standardize/rescale the data?”, 2002. [Online]. Available: http://www.faqs.org/faqs/ai-faq/neural-nets/part2/section-16.html. [Accessed: 24-Aug-2020].
• [56] S. Ioffe and C. Szegedy, “Batch normalization: Accelerating deep network training by reducing internal covariate shift”, in 32nd International Conference on Machine Learning, ICML 2015, 2015, vol. 1, pp. 448–456.
• [57] J. Brownlee, “How to use Data Scaling Improve Deep Learning Model Stability and Performance”, 2019. [On-line]. Available: https://machinelearningmastery.com/how-to-improve-neu ral-network-stability-and-modeling-performance-with-data-sca ling/. [Accessed: 29-Jul-2020].
• [58] S. Nayak, B.B. Misra, and H.S. Behera, “Impact of Data Normalization on Stock Index Forecasting”, Int. J. Comput. Inf. Syst. Ind. Manag. Appl., vol. 2014, pp. 257–269, 2014.
• [59] Tibo Spotfire Documentation, “Normalization by subtracting the median,” 2014. [Online]. Available: https://docs.tibco.com/pub/spotfire/6.5.0/doc/html/norm/norm_subtract_the_median.htm. [Accessed: 25-Aug-2020].
• [60] J. Jones, “Stats: Measures of Central Tendency,” 2020. [Online]. Available: https://people.richland.edu/james/lecture/m170/ch03-ave.html. [Accessed: 25-Aug-2020].
• [61] K. Pietrak, P. Łapka, and M. Kujawińska, “Numerical tests of an inverse method for the characterization of pulsed and continuous laser beams”, in Contemporary Issues of Heat and Mass Transfer, 2019, vol. 1, pp. 481–502.