The Impact of the Number of Neurons in the Hidden Layer on the Performance of MLP Neural Network: Application to the Fast Identification of Toxic Gases

Authors: Slimane Ouhmad, Abdellah Halimi

Abstract:

In this work, neural networks methods MLP type were applied to a database from an array of six sensors for the detection of three toxic gases. The choice of the number of hidden layers and the weight values are influential on the convergence of the learning algorithm. We proposed, in this article, a mathematical formula to determine the optimal number of hidden layers and good weight values based on the method of back propagation of errors. The results of this modeling have improved discrimination of these gases and optimized the computation time. The model presented here has proven to be an effective application for the fast identification of toxic gases.

Keywords: Back-propagation, Computing time, Fast identification, MLP neural network, Number of neurons in the hidden layer.

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1338002

References:

[1] M. W. Gardner, S.R. Dorling, “Neural network modeling and prediction of hourly NOX and NO2 concentrations in urban air in London,” Atmospheric Environment, 33, 5, pp. 709–719, 1999.
[2] Kukkonen,et al, “Extensive evaluation of neural network models for the prediction of NO2 and PM10 concentrations compared with a deterministic modeling system and measurements in central Helsinski,” Atmospheric Environment,37,pp. 4539–4550, 2003.
[3] D. Jiang, Y. Zhang, X. Hu, Y. Zeng, J. Tan, D. Shao, “in developing an ANN model for air pollution index forecast,” Atmospheric Environment, 38,pp. 7055–7064, 2004.
[4] U. Brunelli, V. Piazza, L. Pignato, F. Sorbello, S. Vitabile, “Two-day ahead prediction of daily maximum concentrations of SO2, O3, PM10, NO2, CO in the urban area of Palermo,” Atmospheric Environment, 41,pp. 2967–2995, 2007.
[5] K.D. Karatzas, S. Kaltsatos, “Air pollution modelling with the aid of computational intelligence methods,” in Thessaloniki. Greece, Simulation Modelling Practice and Theory, pp. 1310–1319, doi. 10.1016/ j.simpat.2007.09.005.
[6] J. Galloway, et al, “Nitrogen cycles: past, present and future,” Biogeochemistry, 70, pp.153–226, 2004.
[7] H.J.S. Fernando, et al, “Forecasting PM10 in metropolitan areas: Efficacy of neural networks,” Environmental Pollution, 163, pp.62-67, 2012.
[8] J. Fenger, “Air pollution in the last 50 years—from local to global,” Atmos Environ, pp. 13–22, 2009.
[9] D. Gacquer, et al, “Comparative study of supervised classification algorithms for the detection of atmospheric pollution,” Engineering Applications of Artificial Intelligence, 24, pp. 1070–1083, 2011.
[10] A.B. Chelani, M.Z. Hasan, “Forecasting nitrogen dioxide concentration in ambient air using artificial neural-networks,” International Journal of Environmental Studies, 58, 4,pp. 487–499, 2001.
[11] K.M. Venables. et al, “Thunderstorm-related asthma - the epidemic of 24/25 June 1994,” Clinical and Experimental Allergy, 27, pp.725-736, 1997.
[12] N. Lokeshwari, G. Srinikethan, V. S. Hegde, “Management of Air Pollutants from Point Sources,” International Journal of Environmental Earth Science and Engineering, Vol.7, pp.12, 2013.
[13] J. Burlakovs, M. Klavins, R. Ernsteins, A. Ruskulis, “Contamination in Industrial Areas and Environmental Management in Latvia,” World Academy of Science. Engineering and Technology, Vol.76, pp.04-23, 2013.
[14] R. Lalauze, C. Pijolat, “A new approach to selective detection of gas by an SnO2 solid-state sensor,” Sensors and Actuators, 5, pp. 55–63, 1984.
[15] P. Romppainen, V. Lantto, S. Leppavuori, “Effect of water vapour on the CO response behavior of tin dioxide sensors in constant temperature and temperature-pulsed modes of operation,” Sensors and Actuators, B. Chemicall, pp. 73–78, 1990.
[16] P. van Geloven, M. Honore, J. Roggen, “The influence of relative humidity on the response of tin oxide gas sensors to carbon monoxide,” Sensors and Actuators, B. Chemical, 4,pp.185–188, 1991.
[17] J. Watson, K. Ihokura, G.S.V. Coles, “ The tin dioxide gas sensor,” Measurement Science and Technology, 4,pp. 711–719, 1993.
[18] H. Yamaura, T. Jinkawa, J. Tamaki, K. Moriya, N. Miura, N. Yamazoe,“Indium oxide-based gas sensor for selective detection of CO,” Sensors and Actuators, B: Chemical, 35–56,pp.325–332, 1996.
[19] H. Debeda, P. Massok, C. Lucati, F. Menil, J.L. Aucouturier, “Methane sensing:from sensitive thick films to a reliable selective device,” Measurement Science and Technology,vol.8, pp.99–110, 1997.
[20] J. Judith Vijaya, L. John Kennedy, G. Sekaran, K.S. Nagaraja, “Utilization of Sr(II)-added calcium aluminate for the detection of volatile organic compounds,” Industrial and Engineering Chemistry Research, vol. 46.19,pp. 6251–6258, 2007.
[21] B.W. Licznerski, K. Nitsch, H. Teterycz, T. Soba´nski, K. Wi´sniewski, “Characterization of electrical parameters for multilayer SnO2gas sensors,” Sensors and Actuators B: Chemical. vol. 103, 1–2, pp. 69–75, 2004.
[22] A. Gulbag, F. Temurtas, “A study on quantitative classification of binary gas mixture using neural networks and adaptive neuro -fuzzy inference systems,” Sensors and Actuators.B. Chemical, vol. 115, pp. 252–262, 2006.
[23] U. Brunelli, V. Piazza, L. Pignato, F. Sorbello, S. Vitabile, “days ahead prediction of daily maximum concentrations of SO2, O3, PM10, NO2, CO in th eurban area of Palermo.Italy,”. Atmospheric Environment, vol.41,pp.2967–2995, 2007.
[24] M. Kolehmainen, H. Martikainen, J. Ruuskanen, “Neural networks and periodic components used in air quality forecasting,” Atmospheric Environment, vol. 35, pp. 815–825, 2001.
[25] J. Yi, V.R. Prybutok, “A neural network model forecasting for prediction of daily maximum ozone concentration in an industrialized urban area,” Environmental Pollution, vol. 92, pp. 349–357, 1996.
[26] E. I. Vlahogianni, “Optimized and meta-optimized neural networks for short-term traffic flow prediction: A genetic approach,” Transportation Research Part C, vol. 1, pp. 211–234, 2005.
[27] A. Eswaradass, X.-H. Sun, M. Wu, “A neural network based predictive mechanism for available bandwidth”, proceeding of 19th, IEEE International Conference on Parallel and Distributed Proceeding Symposium,Workshop.10,vol.11,p.228,2005.. doi>10.1109/IPDPS.p.234.2005.
[28] H. Yousefia, “Utilizing neural networks to reduce packet loss in selfsimilar tele traffic,” proceeding of IEEE International. Conference on Communications, vol.3.pp. 1942–1946, 2003.
[29] M. Badura, et al, “Statistical assessment of quantification methods used in gas sensor system”, Sensors and Actuators: B. Chemical, vol.188, pp.815– 823, November 2011.
[30] P. Chandra and Y. Singh, "An activation function adapting training algorithm for sigmoidal feedforward networks," Neurocomputing, vol.61, pp. 429–437, 2004.
[31] W. Duch and N. Jankowski, “Survey of neural transfer functions”, Neural Comput Appl, vol. 2, pp.163–212, 1999.
[32] W. Duch and N. Jankowski, “Transfer functions: Hidden possibilities for better neural networks,” In 9th European symposium on artificial neural networks, pp. 81–94, 2001.
[33] Y. Singh and P. Chandra, “A class +1 sigmoidal activation functions for FFANNs,” J Econ Dynamic Control, vol. 28, pp.183–187, 2003.
[34] B.W. Licznerski, C K. Nitsch, H. Teterycz, T. Sobanski, K. Wisniewski, “haracterisation of electrical parameters for multilayer SnO2gas sensors”, Sensors and Actuators. B. Chemical,vol.103, No. 1–2, pp. 69– 75, 2004.
[35] A. Lfakir, “Identification and Quantification of a complex gaseous atmosphere with an intelligent multi-sensor system. Application to Detection of mixtures compounds H2S, NO2, SO2 wet atmosphere variable”, Thesis of Metz University, 2006. “Identification et Quantification d’une atmosphère gazeuse complexe à l’aide d’un système multi-capteurs intelligent. Application La détection de mélanges composés de H2S, NO2, SO2 en atmosphère humide variable”, thèse d’Université Metz, 2006.
[36] O. Helli, “Gas multi -sensor for the design of an electronic nose for air pollution monitoring”, Thesis of Metz University, 2003. “Multi-capteurs de gaz pour la conception d'un nez électronique de surveillance de la pollution atmosphérique”, thèse d’Université Metz, 2003.
[37] T. Seiyama, N. Yamazoe, S. Yamauchi,Chemical Sensor Technology, Vol. 1-4, Elsevier Science Ltd, 1988.
[38] Rumelhart, D.E. Hinton, G.E. Williams, “Learning internal representations by error propagation,” In Rumelhart, D.E. Mc Cleland, J.L. (Eds.), Parallel Distributed Processing. MIT Press, Cambridge, pp. 318–362, 1986.
[39] W.S. McCulloch, W.H. Pitts, “A logical calculus of the ideas immanent in nervous activity”, Bulletin of Mathematical Biophysics, vol 5, pp. 115–133, 1943.
[40] Cybenko, “Approximation by superposition of a sigmoidal function, ” Math. Cont. Sig. Syst 2, pp. 303–314.
[41] S. Ouhmad, H. Roukhe, A, Roukhe “Real time identification of toxic gases based on artificial neural networks, ” International Journal of Computational Engineering Research,vol. 04, issue. 4, pp. 67-72, 2014.