A comparison of artificial intelligence-based classification techniques in predicting flow variables in sharp curved channels

  • Azadeh Gholami
  • Hossein BonakdariEmail author
  • Amir Hossein Zaji
  • Ali Akbar Akhtari
Original Article


Due to the complexity of variable distributions, it is essential to evaluate the flow patterns in sharp curved channels with different angles. It has recently become more common to use classification methods in combination with different artificial intelligence (AI) models to boost AI model performance. Gholami et al. (Appl Soft Comput 48:563–583, 2016a) obtained enhanced results by combining two common AI models with a classifier based on decision trees to estimate the flow velocity and flow depth in a 90° sharp curved channel. Their results represent the superior performance and accuracy of hybrid models over classic AI models alone. However, because the diversion angle in intake and bend channels has a considerable influence on the flow variables, the flow patterns differ in 60° and 90° bend channels. Hence, the present paper evaluates the goodness-of-fit results of the classifier multilayer perceptron (CMLP) and classifier radial basis function (CRBF) models designed based on decision trees. These models were designed to estimate the velocity and flow depth patterns in a 60° sharp curved channel based on 780 observational data for six flow rates: 5, 7.8, 13.6, 19.1, 25.3, and 30.6 l/s. According to the mean absolute relative error (MARE) and determination coefficient (R2) results of 0.0514 and 0.6 for velocity and 0.005 and 0.99 for flow depth prediction, the proposed CMLP model is more accurate than the classic MLP model. The CRBF model performed similar to CMLP, whereby CRBF predicted both velocity (MARE 0.086 and R2 = 0.9) and flow depth (MARE 0.0133 and R2 = 0.9) more accurately than the RBF model alone. Overall, the CMLP and CRBF models exhibited MARE reductions of 3% and 20% in velocity prediction and 36% and 22% in flow depth prediction compared to the individual MLP and RBF models, respectively. Moreover, CMLP and CRBF produce robust relationships based on different categories established according to various hydraulic conditions. The uncertainty analysis results show that, among the models examined, CMLP had the lowest uncertainty with the narrowest width of confidence bounds (WCB) of ± 0.1385 and ± 0.0031 in predicting velocity and flow depth, respectively. Moreover, CMLP and CRBF exhibited the greatest reliability with the lowest uncertainty indices in estimating the two flow variables in curved channels compared to the classic MLP and RBF models. Therefore, the combined AI-based classification models proposed in this study can serve as alternatives to the classic MLP and RBF models in the design and construction of curved channels with 60° and 90° bends.


Classification algorithms Uncertainty analysis 60° sharp bend Flow velocity Flow depth 



  1. 1.
    Akbari M, Vaghefi M (2017) Experimental investigation on streamlines in a 180° sharp bend. Acta Sci Technol 39(4):425–432Google Scholar
  2. 2.
    Akhtari AA, Abrishami J, Sharifi MB (2009) Experimental investigations water surface characteristics in strongly-curved open channels. J Appl Sci 9(20):3699–3706Google Scholar
  3. 3.
    Akhtari AA, Seyedashraf O (2018) Experimental and numerical investigation on Vanes’ effects on the flow characteristics in sharp 60° bends. KSCE J Civ Eng 22(4):1484–1495Google Scholar
  4. 4.
    Anwar HO (1986) Turbulent structure in a river bed. J Hydraul Eng 112:657–669Google Scholar
  5. 5.
    Araghinejad S, Fayaz N, Hosseini-Moghari SM (2018) Development of a hybrid data driven model for hydrological estimation. Water Resour Manag 1–14Google Scholar
  6. 6.
    Armfield Limited, Co. (1995) Instruction manual of miniature propeller velocity meter type H33. Armfield Limited, Co, LondonGoogle Scholar
  7. 7.
    Asl PF, Monjezi M, Hamidi JK, Armaghani DJ (2018) Optimization of flyrock and rock fragmentation in the Tajareh limestone mine using metaheuristics method of firefly algorithm. Eng Comput 34(2):241–251Google Scholar
  8. 8.
    Aydinalp M, Ugursal VI, Fung AS (2002) Modeling of the appliance, lighting, and space-cooling energy consumptions in the residential sector using neural networks. Appl Energy 71(2):87–110Google Scholar
  9. 9.
    Baghlani A (2012) Application of a high-resolution scheme in simulation of flow in curved channel using boundary-fitted curvilinear coordinates. Sci Iran 19(6):1463–1472Google Scholar
  10. 10.
    Barbhuiya AK, Talukdar S (2010) Scour and three dimensional turbulent flow fields measured by ADV at a 90° horizontal forced bend in a rectangular channel. Flow Meas Instrum 21:312–321Google Scholar
  11. 11.
    Bergs MA (1990) Flow processes in a curved alluvial channel. Ph. D. thesis, The University of IowaGoogle Scholar
  12. 12.
    Berry G, Armitage P (1995) Mid-P confidence intervals: a brief review. Statistician 44:417–423Google Scholar
  13. 13.
    Bilhan O, Emin Emiroglu M, Kisi O (2010) Application of two different neural network techniques to lateral outflow over rectangular side weirs located on a straight channel. Adv Eng Softw 41(6):831–837zbMATHGoogle Scholar
  14. 14.
    Bishop CM (1995) Neural networks for pattern recognition. Oxford University Press Inc, New YorkzbMATHGoogle Scholar
  15. 15.
    Blanckaert K, Graf WH (2001) Mean flow and turbulence in open channel bend. J Hydraul Eng ASCE 127(10):835–847Google Scholar
  16. 16.
    Bodnar T, Prihoda J (2006) Numerical simulation of turbulent free-surface flow in curved channel. Flow Turbul Combust 76:429–442zbMATHGoogle Scholar
  17. 17.
    Bonakdari H, Zaji AH (2016) Open channel junction velocity prediction by using a hybrid self-neuron adjustable artificial neural network. Flow Meas Instrum 49:46–51Google Scholar
  18. 18.
    Booij R (2003) Measurements and large eddy simulations of the flows in some curved flumes. J Turbul 4:8–16Google Scholar
  19. 19.
    Breiman L, Friedman JH, Olshen RA, Stone CJ (1984) Classification and regression trees. Wadsworth International Group, BelmontzbMATHGoogle Scholar
  20. 20.
    Broomhead DS, Lowe D (1988) Radial basis functions, multi-variable functional interpolation and adaptive networks. Royal Signals and Radar Establishment, MalvernzbMATHGoogle Scholar
  21. 21.
    Chen W, Fu ZJ, Chen CS (2014) Recent advances in radial basis function collocation methods. Springer, HeidelbergzbMATHGoogle Scholar
  22. 22.
    Christopher M (1995) Neural networks for pattern recognition. Oxford University Press, OxfordGoogle Scholar
  23. 23.
    Coppersmith D, Hong SJ, Hosking JRM (1999) Partitioning nominal attributes in decision trees. Data Min Knowl Discov 3:197–217Google Scholar
  24. 24.
    Dawson CW, Wilby RL (2005) Hydrological modeling using artificial neural networks. Prog Phys Geogr 25(1):80–108Google Scholar
  25. 25.
    DeMarchis M, Napoli E (2006) 3D numerical simulation of curved open channel flows. In: Proceedings of 6th international conference on water resources, hydraulics & hydrology, Chalkida, Evia Island, pp 86–91Google Scholar
  26. 26.
    DeVriend HJ, Geoldof HJ (1983) Main flow velocity in short river bends. J Hydraul Eng 109:991–1011Google Scholar
  27. 27.
    Ebtehaj I, Bonakdari H, Zaji AH (2016) A nonlinear simulation method based on a combination of multilayer perceptron and decision trees for predicting non-deposition sediment transport. Water Sci Technol Water Supply 16(5):1198–1206Google Scholar
  28. 28.
    Fenjan SA, Bonakdari H, Gholami A, Akhtari AA (2016) Flow variables prediction using experimental, computational fluid dynamic and artificial neural network models in a sharp bend. Int J Eng Trans A Basics 29(1):14Google Scholar
  29. 29.
    Feng G, Huang GB, Lin Q, Gay R (2009) Error minimized extreme learning machine with growth of hidden nodes and incremental learning. IEEE Trans Neural Netw 20(8):1352–1357Google Scholar
  30. 30.
    Ferguson RI, Parsons DR, Lane SN, Hardy RJ (2003) Flow in meander bends with recirculation at the inner bank. J Water Resour Res 39:1–13Google Scholar
  31. 31.
    Gerger R, Kisi O, Faruk Dursun O, Emin Emiroglu M (2016) Applicability of several soft computing approaches in modeling oxygen transfer efficiency at baffled chutes. J Irrig Drain Eng 143(5):04016085Google Scholar
  32. 32.
    Gharde KD, Kothari M, Mittal HK, Singh PK, Dahiphale PA (2015) Sediment yield modelling of Kal river in Maharashtra using artificial neural network model. Res J Recent Sci 4:120–130Google Scholar
  33. 33.
    Ghasemi E, Kalhori H, Bagherpour R (2016) A new hybrid ANFIS-PSO model for prediction of peak particle velocity due to bench blasting. Eng Comput 32(4):607–614Google Scholar
  34. 34.
    Gholami A, Akhtari AA, Minatour Y, Bonakdari H, Javadi AA (2014) Experimental and numerical study on velocity fields and water surface profile in a strongly-curved 90° open channel bend. Eng Appl Comput Fluid Mech 8(3):447–461Google Scholar
  35. 35.
    Gholami A, Bonakdari H, Zaji AH, Akhtari AA, Khodashenas SR (2015) Predicting the velocity field in a 90° open channel bend using a gene expression programming model. Flow Meas Instrum 46:189–192Google Scholar
  36. 36.
    Gholami A, Bonakdari H, Zaji AH, Akhtari AA (2015) Simulation of open channel bend characteristics using computational fluid dynamics and artificial neural networks. Eng Appl Comput Fluid Mech 9:355–369Google Scholar
  37. 37.
    Gholami A, Bonakdari H, Zaji AH, Michelson DG, Akhtari AA (2016) Improving the performance of multi-layer perceptron and radial basis function models with a decision tree model to predict flow variables in a sharp 90° bend. Appl Soft Comput 48:563–583Google Scholar
  38. 38.
    Gholami A, Bonakdari H, Akhtari AA (2016) Assessment of water depth change patterns in 120° sharp bend using numerical model. Water Sci Eng 9(4):336–344Google Scholar
  39. 39.
    Gholami A, Bonakdari H, Zaji AH, Ajeel Fenjan S, Akhtari AA (2016) Design of modified structure multi-layer perceptron networks based on decision trees for the prediction of flow parameters in 90 open-channel bends. Eng Appl Comput Fluid Mech 10(1):193–208Google Scholar
  40. 40.
    Gholami A, Bonakdari H, Zaji AH, Fenjan SA, Akhtari AA (2016) New radial basis function network method based on decision trees to predict flow variables in a curved channel. Neural Comput Appl 30(9):2771–2785Google Scholar
  41. 41.
    Gholami A, Bonakdari H, Akhtari AA (2016) Developing finite volume method (FVM) in numerical simulation of flow pattern in 60° open channel bend. J Appl Res Water Wastewater 3(1):193–200Google Scholar
  42. 42.
    Gholami A, Bonakdari H, Ebtehaj I, Shaghaghi S, Khoshbin F (2017) Developing an expert group method of data handling system for predicting the geometry of a stable channel with a gravel bed. Earth Surf Process Landf 42(10):1460–1471Google Scholar
  43. 43.
    Gholami A, Bonakdari H, Ebtehaj I, Akhtari AA (2017) Design of an adaptive neuro-fuzzy computing technique for predicting flow variables in a 90° sharp bend. J Hydroinform jh2017200Google Scholar
  44. 44.
    Gholami A, Bonakdari H, Ebtehaj I, Gharabaghi B, Khodashenas SR, Talesh SHA, Jamali A (2018) A methodological approach of predicting threshold channel bank profile by multi-objective evolutionary optimization of ANFIS. Eng Geol 239:298–309Google Scholar
  45. 45.
    Gholami A, Bonakdari H, Zeynoddin M, Ebtehaj I, Gharabaghi B, Khodashenas SR (2018) Reliable method of determining stable threshold channel shape using experimental and gene expression programming techniques. Neural Comput Appl 1–19Google Scholar
  46. 46.
    Hagan MT, Demuth HB, Mark H, Beale MH (1996) Neural network design. PWS Publishing Company, BostonGoogle Scholar
  47. 47.
    Haykin S (1990) Neural networks. A comprehensive foundation, 2nd edn. Prentice-Hall Inc, Englewood CliffszbMATHGoogle Scholar
  48. 48.
    Isa IS, Saad Z, Omar S, Osman MK, Ahmad KA, Sakim HM (2010) Suitable MLP network activation functions for breast cancer and thyroid disease detection. In: 2010 second international conference on computational intelligence, modelling and simulation (CIMSiM). IEEE, New York, pp 39–44Google Scholar
  49. 49.
    Karsoliya S (2012) Approximating number of hidden layer neurons in multiple hidden layer BPNN architecture. Int J Eng Trends Technol 3(6):714–717Google Scholar
  50. 50.
    Latifoğlu L, Kişi O, Latifoğlu F (2015) Importance of hybrid models for forecasting of hydrological variable. Neural Comput Appl 26(7):1669–1680Google Scholar
  51. 51.
    Leschziner MA, Rodi W (1979) Calculation of strongly curved open channel flow. J Hydraul Div 105:1297–1314Google Scholar
  52. 52.
    Levenberg K (1944) A method for the solution of certain nonlinear problems in least-squares. Q Appl Math 2:164–168MathSciNetzbMATHGoogle Scholar
  53. 53.
    Li X, Chan CW (2010) Application of an enhanced decision tree learning approach for prediction of petroleum production. Eng Appl Artif Intell 23(1):102–109Google Scholar
  54. 54.
    Maier HR, Dandy GC (1999) Empirical comparison of various methods for training feed Forward neural networks for salinity forecasting. Water Resour Res 35(8):2591–2596Google Scholar
  55. 55.
    Mount NJ, Dawson CW, Abrahart RJ (2013) Legitimizing data-driven models: exemplification of a new data-driven mechanistic modeling framework. Hydrol Earth Syst Sci 17:2827–2843Google Scholar
  56. 56.
    Naji MA, Ghodsian M, Vaghefi M, Panahpur N (2010) Experimental and numerical simulation of flow in a 90° bend. Flow Meas Instrum 21:292–298Google Scholar
  57. 57.
    Newcombe RG (1998) Two-sided confidence intervals for the single proportion: comparison of seven methods. Stat Med 17(8):857–872Google Scholar
  58. 58.
    Nayak SC, Misra BB, Behera HS (2016) Efficient forecasting of financial time-series data with virtual adaptive neuro-fuzzy inference system. Int J Bus Forecast Market Intell 2(4):379–402Google Scholar
  59. 59.
    Nkiaka E, Nawaz NR, Lovett JC (2016) Using self-organizing maps to infill missing data in hydro-meteorological time series from the Logone catchment, Lake Chad basin. Environ Monit Assess 188(7):400Google Scholar
  60. 60.
    Oliver N (2001) Nonlinear system identification: from classical approaches to neural networks and fuzzy models. Springer, New YorkzbMATHGoogle Scholar
  61. 61.
    Parvin H, MirnabiBaboli M, Alinejad-Rokny H (2015) Proposing a classifier ensemble framework based on classifier selection and decision tree. Eng Appl Artif Intell 37:34–42Google Scholar
  62. 62.
    Poggio T, Girosi F (1990) Regularization algorithms for learning that are equivalent to multilayer networks. Science 247:978–982MathSciNetzbMATHGoogle Scholar
  63. 63.
    Pourgholi R, Tabasi SH, Zeidabadi H (2018) Numerical techniques for solving system of nonlinear inverse problem. Eng Comput 34(3):487–502Google Scholar
  64. 64.
    Pradhan A (2014) Analysis of flow along the meander path of a highly sinuous rigid channel. Dissertation National Institute of Technology, RourkelaGoogle Scholar
  65. 65.
    Ramamurthy A, Han S, Biron P (2013) Three-dimensional simulation parameters for 90° open channel bend flows. J Comput Civ Eng 27:282–291Google Scholar
  66. 66.
    Rowiński PM, Piotrowski A, Napiórkowski JJ (2005) Are artificial neural network techniques relevant for the estimation of longitudinal dispersion coefficient in rivers? J Hydrol Sci 50:175–187Google Scholar
  67. 67.
    Rozovskii IL (1961) Flow of water in bends of open channels. Academy of Sciences of the Ukrainian SSR. Israel Program for Science Translation, Kiev, pp 1–233Google Scholar
  68. 68.
    Saxena A, Saad A (2007) Evolving an artificial neural network classifier for condition monitoring of rotating mechanical systems. Appl Soft Comput 7(1):441–454Google Scholar
  69. 69.
    Seyedashraf O, Akhtari AA (2017) Three-dimensional CFD study of free-surface flow in a sharply curved 30 open-channel bend. J Eng Sci Technol Rev 10(3):85–89Google Scholar
  70. 70.
    Shaghaghi S, Bonakdari H, Gholami A, Kisi O, Shiri J, Binns AD, Gharabaghi B (2018) Stable alluvial channel design using evolutionary neural networks. J Hydrol 566:770–782Google Scholar
  71. 71.
    Tiwari DK, Tiwari HL, Mishra S, Nateriya R (2018) A literature review on ANN based hydrological modeling on Narmada river basin. Int J Innov Adv Comput Sci 74Google Scholar
  72. 72.
    Ye J, McCorquodale JA (1998) Simulation of curved open channel flows by 3D hydrodynamic model. J Hydraul Eng 124:687–698Google Scholar
  73. 73.
    Yuhong Z, Wenxin H (2009) Application of artificial neural network to predict the friction factor of open channel flow. Commun Nonlinear Sci Numer Simul 14(5):2373–2378Google Scholar
  74. 74.
    Wang L, von Laszewski G, Huang F, Dayal J, Frulani T, Fox G (2011) Task scheduling with ANN-based temperature prediction in a data center: a simulation-based study. Eng Comput 27(4):381–391Google Scholar
  75. 75.
    Zabihi F, Saffarian M (2018) A not-a-knot meshless method with radial basis functions for numerical solutions of Gilson–Pickering equation. Eng Comput 34(1):37–44Google Scholar
  76. 76.
    Zaji AH, Bonakdari H (2015) Efficient methods for prediction of velocity fields in open channel junctions based on the artificial neural network. Eng Appl Comput Fluid Mech 9(1):220–232Google Scholar
  77. 77.
    Zaji AH, Bonakdari H, Shamshirband S (2016) Support vector regression for modified oblique side weirs discharge coefficient prediction. Flow Meas Instrum 51:1–7Google Scholar
  78. 78.
    Zaji AH, Bonakdari H (2018) Velocity field simulation of open-channel junction using artificial intelligence approaches. Iran J Sci Technol Trans Civ Eng 1–12Google Scholar
  79. 79.
    Zamanlooy B, Mirhassani M (2014) Efficient VLSI implementation of neural networks with hyperbolic tangent activation function. IEEE Trans Very Large Scale Integr VLSI Syst 22(1):39–48Google Scholar
  80. 80.
    Zeng J, Constantinescu G, Blanckaert K, Weber L (2008) Flow and bathymetry in sharp open-channel bends: experiments and predictions. Water Resourc Res 44(w09401):1–22Google Scholar

Copyright information

© Springer-Verlag London Ltd., part of Springer Nature 2019

Authors and Affiliations

  1. 1.Department of Civil EngineeringRazi UniversityKermanshahIran

Personalised recommendations