Do we need a more sophisticated multilayer artificial neural network to compute roughness coefficient?
DOI:
https://doi.org/10.32347/2411-4049.2023.4.170-182Keywords:
activation functions, artificial neural networks, Chézy’s roughness coefficient, comparison, dropout algorithm, hidden layers, modifications, neuronsAbstract
Artificial neural networks (ANNs) are one of the most rapidly growing fields of soft computing. Along with deep learning, they are currently the most widely used machine learning techniques. Artificial neural networks are especially suitable for problem-solving where a researcher deals with incomplete data sets and no algorithms or specific sets of rules to be followed.
This article deals with a case of comparison of several modifications of neural networks that may be applied to compute Chézy’s roughness coefficient. Neural network modelling is often started with one hidden layer. Having even one hidden layer, a neural network presents a powerful computing system to give good results. If it is necessary, the number of hidden layers may increase. Usually, two or three hidden layers of neurons are used. Diverse activation functions may also apply. The article aims to explore the necessity of developing sophisticated multilayer artificial neural networks to compute Chézy’s roughness coefficient.
Under the study, the following modifications of the neural network computing Chézy’s roughness coefficient were considered and analysed: (1) Application of two hidden layers of neurons; (2) Application of three hidden layers of neurons; (3) Use of a dropout algorithm for training neural networks by randomly dropping units during training to prevent their co-adaptation; (4) Apart from the sigmoid (logistic) activation function, the use of other artificial neuron transfer functions – hyperbolic tangent (tanh) and rectifying activation function (ReLU).
The training and testing of the considered neural network options were carried out using the actual hydro-morphological and hydrological data related to the channel section on the Dnieper River (downstream of Kyiv), the Desna River section near Chernihiv, and the Pripyat River section near the town of Turiv. The Python object-oriented programming environment was applied to build and train the neural networks. The test results confirm the acceptability and sufficiency of computing the Chézy roughness coefficient using the ANN of direct propagation with one hidden layer and a sigmoid logistic activation function. The formation of a qualitative set of training data, as well as data arrangement and choosing a relevant computing model based on empirical knowledge, are, as concluded, among more actual issues than creating more sophisticated neural networks.
References
Zadeh, L.A. (1994). Fuzzy logic, neural networks and soft computing. Fuzzy Systems. Communications of the ACM, Vol. 37, No. 3, 77–84.
Ibrahim, D. (2016). An overview of soft computing. Procedia Computer Science, 102, 34–38. https://doi.org/10.1016/j.procs.2016.09.366
Bele, S.B. (2020). The concept of soft computing. International Journal of Recent Advances in Multidisciplinary Research, Vol. 07, Issue 03, 5623–5625.
Carlos, M. (2022). Characteristics of Soft Computing and its Applications. Int. J. of Swarm Intelligent and Evolutionary Comp. 11:275. https://doi.org/10.35248/2090-4908.22.11.275
Soft computing approach for mathematical modeling of engineering problems. (2022). Ed. by A. Ahmadian, and S. Salahshour. CRC Press. Taylor & Francis Group, London, N.Y., 203 p.
Singh, Ju. (2016). A Parameterized Comparison of Fuzzy Logic, Neural Network and Neuro-Fuzzy System: A Literature. International Journal of Computer Science and Mobile Computing, Vol. 5, Issue 5, 478-482.
Thakur, A., Dhiman, K., and Phansikar, M. (2021). Neuro-Fuzzy: Artificial Neural Networks & Fuzzy Logic. International Journal for Research in Applied Science & Engineering Technology, Vol. 9, Issue IX, 128-135.
Haghbin, M., Sharafati, A., Motta, D. et al. (2021). Applications of soft computing models for predicting sea surface temperature: a comprehensive review and assessment. Progress in Earth and Planetary Science, 8:4. https://doi.org/10.1186/s40645-020-00400-9
Adnan, R.M., Meshram, S.G., Mostafa, R.R., et al. (2023). Application of Advanced Optimized Soft Computing Models for Atmospheric Variable Forecasting. Mathematics, 11, 1213. https://doi.org/10.3390/math11051213
Pham, B.T., Nguyen, M.D., Al-Ansari, N. et al. (2021). A Comparative Study of Soft Computing Models for Prediction of Permeability Coefficient of Soil. Mathematical Problems in Engineering, Vol. 2021, 7631493. https://doi.org/10.1155/2021/7631493
Redjimi, H., and Tar, J.K. (2021). A Simple Soft Computing Structure for Modeling and Control. Machines, 9, 168. https://doi.org/10.3390/machines9080168
Afaq, H., and Saini, S. (2011). Swarm Intelligence based Soft Computing Techniques for the Solutions to Multiobjective Optimization Problems. International Journal of Computer Science Issues, Vol. 8, Issue 3, No. 2, 498-510.
Soft Computing: Recent Advances and Applications in Engineering and Mathematical Sciences. (2023). Edited by P. Debnath, O. Castillo, and P. Kumam. CRC Press. Taylor & Francis Group, London, N.Y., 233 p.
Haikin, S. (2008). Neural Networks and Learning Machines (3rd Edition), Prentice Hall, 906 p.
Choi, R.Y., Coyner, A.S., Kalpathy-Cramer, J., Chiang, M.F., and Campbell, J.P. (2020). Introduction to machine learning, neural networks, and deep learning. Trans Vis Sci Tech., Special Issue, Vol. 9, No. 2, Article 14:2. https://doi.org/10.1167/tvst.9.2.14
Taye, M.M. (2023). Understanding of Machine Learning with DeepLearning: Architectures, Workflow, Applications and Future Directions. Computers, 12, 91. https://doi.org/10.3390/computers12050091
Russell, S.J., and Norvig, P. (2010). Artificial Intelligence: A Modern Approach. 3rd ed. Pearson Education, Inc.: Upper Saddle River, New Jersey, 1132 p.
Dhar, V. (2013). Data science and prediction. Communications of the ACM 56 (12): 64. https://doi.org/10.1145/2500499
Zhang, G., Patuwo, B.E., Hu, M.Y. (1998). Forecasting with artificial neural networks: The state of the art. International Journal of Forecasting, 14, 35–62. https://doi.org/10.1016/S0169-2070(97)00044-7
Montaño Moreno, J.J., Pol, A.P., and Gracia, P.M. (2011). Artificial neural networks applied to forecasting time series. Psicothema, Vol. 23, No 2, 322–329.
Stefanyshyn, D.V., Khodnevich, Y.V., Korbutiak, V.M. (2021). Estimating the Chezy roughness coefficient as a characteristic of hydraulic resistance to flow in river channels: a general overview, existing challenges, and ways of their overcoming. Environmental safety and natural resources, 39(3), 16–43. https://doi.org/10.32347/2411-4049.2021.3.16-43
Khodnevych, Y.V., Stefanyshyn, D.V. (2022). Data arrangements to train an artificial neural network within solving the tasks for calculating the Chézy roughness coefficient under uncertainty of parameters determining the hydraulic resistance to flow in river channels. Environmental safety and natural resources, 42(2), 59–85. https://doi.org/10.32347/2411-4049.2022.2.59-85
Chollet, F. (2018). Deep Learning with Python. Manning Publications Co., 384 p.
Trask, A.W. (2019). Grokking Deep Learning. Manning Publications Co., 336 p.
Cybenko, G.V. (1989). Approximation by Superpositions of a Sigmoidal function. Mathematics of Control, Signals and Systems, Vol. 2, No. 4, 303–314.
Keim, R. (2020). How Many Hidden Layers and Hidden Nodes Does a Neural Network Need? Available from https://www.allaboutcircuits.com/technical-articles/how-many-hidden-layers-and-hidden-nodes-does-a-neural-network-need/
Baldi, P., and Sadowski, P. (2014). The dropout learning algorithm. Artificial Intelligence, 210, 78–122. http://dx.doi.org/10.1016/j.artint.2014.02.004
Nielsen, M. Neural Networks and Deep Learning, 224 p. Available from https://static.latexstudio.net/article/2018/0912/neuralnetworksanddeeplearning.pdf
Brownlee, J. (2019). A Gentle Introduction to the Rectified Linear Unit (ReLU). In Deep Learning Performance. Available from https://machinelearningmastery.com/rectified-linear-activation-function-for-deep-learning-neural-networks/
Muller, A., and Guido, S. (2016). Introduction to Machine Learning with Python, Published by O’Reilly Media, 378 p.
Khodnevych, Ya. (2022). The software implementation of a neural network computational algorithm for predicting the Chézy roughness coefficient. Available from https://github.com/yakhodnevych/ANN_approximation_C
Nash, J.E., and Sutcliffe, J.V. (1970). River flow forecasting through conceptual models part I – A discussion of principles, Journal of Hydrology, 10 (3), 282–290. https://doi.org/10.1016/0022-1694(70)90255-6
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2023 Khodnevych Y.V., Stefanyshyn D.V.
This work is licensed under a Creative Commons Attribution 4.0 International License.
The journal «Environmental safety and natural resources» works under Creative Commons Attribution 4.0 International (CC BY 4.0).
The licensing policy is compatible with the overwhelming majority of open access and archiving policies.
