An expert system with radial basis function neural network based on decision trees for predicting sediment transport in sewers.

In this study, an expert system with a radial basis function neural network (RBF-NN) based on decision trees (DT) is designed to predict sediment transport in sewer pipes at the limit of deposition. First, sensitivity analysis is carried out to investigate the effect of each parameter on predicting the densimetric Froude number (Fr). The results indicate that utilizing the ratio of the median particle diameter to pipe diameter (d/D), ratio of median particle diameter to hydraulic radius (d/R) and volumetric sediment concentration (C(V)) as the input combination leads to the best Fr prediction. Subsequently, the new hybrid DT-RBF method is presented. The results of DT-RBF are compared with RBF and RBF-particle swarm optimization (PSO), which uses PSO for RBF training. It appears that DT-RBF is more accurate (R(2) = 0.934, MARE = 0.103, RMSE = 0.527, SI = 0.13, BIAS = -0.071) than the two other RBF methods. Moreover, the proposed DT-RBF model offers explicit expressions for use by practicing engineers.

[1]  Hossein Bonakdari,et al.  Performance Evaluation of Adaptive Neural Fuzzy Inference System for Sediment Transport in Sewers , 2014, Water Resources Management.

[2]  Hossein Bonakdari,et al.  Design criteria for sediment transport in sewers based on self-cleansing concept , 2014 .

[3]  Jose J. Ota,et al.  Urban Storm Sewer Design: Approach in Consideration of Sediments , 2003 .

[4]  O. Kisi.  The potential of different ANN techniques in evapotranspiration modelling , 2008 .

[5]  Wei-Yin Loh,et al.  Classification and regression trees , 2011, WIREs Data Mining Knowl. Discov..

[6]  Robert Banasiak.  Hydraulic performance of sewer pipes with deposited sediments. , 2008, Water science and technology : a journal of the International Association on Water Pollution Research.

[7]  Martin D. Buhmann,et al.  Radial Basis Functions: Theory and Implementations: Preface , 2003 .

[8]  Hossein Bonakdari,et al.  Comparison of genetic algorithm and imperialist competitive algorithms in predicting bed load transport in clean pipe. , 2014, Water science and technology : a journal of the International Association on Water Pollution Research.

[9]  Hossein Bonakdari,et al.  Evaluation of Sediment Transport in Sewer using Artificial Neural Network , 2013 .

[10]  Hossein Bonakdari,et al.  Bed load sediment transport estimation in a clean pipe using multilayer perceptron with different training algorithms , 2016 .

[11]  C Nalluri,et al.  SEDIMENT TRANSPORT IN SMOOTH FIXED BED CHANNELS , 1975 .

[12]  Jaber Almedeij.  Rectangular Storm Sewer Design Under Equal Sediment Mobility , 2012 .

[13]  Jaber Almedeij,et al.  Remarks on Camp's Criterion for Self-Cleansing Storm Sewers , 2010 .

[14]  J J Ota,et al.  Particle velocity and sediment transport at the limit of deposition in sewers. , 2013, Water science and technology : a journal of the International Association on Water Pollution Research.

[15]  T Poggio,et al.  Regularization Algorithms for Learning That Are Equivalent to Multilayer Networks , 1990, Science.

[16]  Mukand S. Babel,et al.  Non-deposition design criteria for sewers with part-full flow , 2010 .

[17]  P. M. Brown,et al.  Self-cleansing conditions for sewers carrying sediment , 1989 .

[18]  Aminuddin Ab. Ghani.  Sediment transport in sewers , 1993 .

[19]  Hossein Bonakdari,et al.  Verification of equation for non-deposition sediment transport in flood water canals , 2014 .