Hazi Mohammad Azamathulla

Neuro-Fuzzy GMDH to Predict the Scour Pile Groups due to Waves

AbstractIn this paper, the neuro-fuzzy based group method of data handling (NF-GMDH) as an adaptive learning network was used to predict the scour process at pile groups due to waves. The NF-GMDH network was developed using the particle swarm optimization (PSO) algorithm and gravitational search algorithm (GSA). Effective parameters on the scour depth include sediment size, geometric property, pile spacing, arrangement of pile group, and wave characteristics upstream of group piles. Seven dimensionless parameters were obtained to define a functional relationship between input and output variables. Published data were compiled from the literature for the scour depth modeling due to waves. The efficiency of training stages for both NF-GMDH-PSO and NF-GMDH-GSA models were investigated. The results indicated that NF-GMDH models could provide more accurate predictions than those obtained using model tree and traditional equations.

Gene expression programming for total bed material load estimation—a case study

This paper presents Gene-Expression Programming (GEP), which is an extension to the genetic programming (GP) approach to predict the total bed material load for three Malaysian rivers. The GEP is employed without any restriction to an extensive database compiled from measurements in the Muda, Langat, and Kurau rivers. The GEP approach demonstrated a superior performance compared to other traditional sediment load methods. The coefficient of determination, R(2) (=0.97) and the mean square error, MSE (=0.057) of the GEP method are higher than those of the traditional method. The performance of the GEP method demonstrates its predictive capability and the possibility of the generalization of the model to nonlinear problems for river engineering applications.

Group method of data handling to predict scour depth around bridge piers

In this study, group method of data handling network with quadratic polynomial was used to predict scour depth around bridge piers. Effective parameters on scour phenomena include sediment size, geometry of bridge pier, and upstream flow conditions. Different shapes of piers have been utilized to develop the GMDH network. Back propagation algorithm was performed to train the GHMD network which updated weighting coefficients of quadratic polynomial in each iteration of the training stage. The GMDH performed with the lowest errors of training and testing stages for cylindrical pier. Also, Richardson and Davis, Johnson’s equations produced relatively good performances for different types of piers. Finally, the results indicated that GMDH could be provided more accurate prediction than those obtained using traditional equations.

ANFIS-Based Approach for Predicting the Scour Depth at Culvert Outlets

The processes involved in the local scour at culverts are so complex and that makes it difficult to establish a general empirical model to provide accurate estimation for scour. This paper describes the use of adaptive neurofuzzy inference system (ANFIS) to estimate the scour depth at culvert outlets. The data sets of laboratory measurements were compiled from published literature and used to train the ANFIS network. The developed network was validated by using the observations that were not involved in training. The performance of ANFIS was found to be more effective ( R2 =0.94 ) when compared with the results of regression equations and artificial neural networks modeling in predicting the scour depth at culvert outlets ( R2 =0.78 ) . Further work is required to collect field data of scour at culvert outlets to train the genetic programming approach and validate its usefulness.

Prediction of pipeline scour depth in clear-water and live-bed conditions using group method of data handling

In the present study, the Group method of data handling (GMDH) network was utilized to predict the scour depth below pipelines. GMDH network was developed using back propagation. Input parameters that were considered as effective parameters on the scour depth included those of sediment size, geometry of pipeline, and approaching flow characteristics. Training and testing performances of the GMDH networks have been carried out using nondimensional data sets that were collected from the literature. These data sets are related to the two main situations of pipelines scour experiments namely clear-water and live-bed conditions. The testing results of performances were compared with the support vector machines (SVM) and existing empirical equations. The GMDH network indicated that using of back propagation produced lower error of scour depth prediction than those obtained using the SVM and empirical equations. Also, the effects of many input parameters on the scour depth have been investigated.

Estimation of dimension and time variation of local scour at short abutment

Accurate prediction of the local scour at abutments is an important criterion to design a safe depth for the bridge foundation. In this paper, the dimension and variation of local scour with time at a vertical-wall abutment were investigated experimentally under clear-water conditions. The multiple linear regression (MLR), gene expression programming (GEP) and artificial neural networks (ANNs), feed forward back propagation and radial basis function were used to predict the time variation of scour depth at a short abutment. Results indicated that the dimension of the scour hole in the x-direction ranged from 3L to 5L upstream and downstream of the abutment, respectively, and also 4L in the y-direction. Statistical analysis showed that, although the ANNs technique produced better results (R 2 = 0.997, RMSE = 0.0113 and MAE = 0.0071) in comparison with the GEP (R 2 = 0.959, RMSE = 0.068 and MAE = 0.044) and MLR techniques (R 2 = 0.958, RMSE = 0.059 and MAE = 0.041), both GEP and MLR are more practical methods. Finally, sensitivity analysis indicated that the local scour was greatly affected by the three studied parameters in the following order, time ratio (t/t e) > abutment length ratio (L/y) > velocity ratio (U/U c).

Soft computing for prediction of river pipeline scour depth

This study presents gene-expression programming (GEP) as an alternative soft computing tool for the prediction of scour below underwater pipeline across river. Actual laboratory measurements were used for the model development. The scour depth was formulated in terms of several influencing parameters. The results indicate that GEP is a very promising approach to predict the river pipeline scour depth.

ANFIS-based approach for the estimation of transverse mixing coefficient.

To develop a proper indicator which could predict water quality and trace pollution sources is critically important for the management of sustainable aquatic ecosystem. In our study, seven water samples collected from Wuliangsuhai Lake in Inner Mongolia were used. UV-visible spectra and synchronous fluorescence spectra were applied to investigate the humification degree and aromatic structure of dissolved organic matter (DOM) extracted from water samples. The results showed that both samples from W1 site and W3 site display lower humification degree and less aromatic structure, where industrial wastewater and domestic sewage, and reclaimed water of farmland irrigation, were accepted respectively. After computing the values of SUVA(254), A(280), A(250/365), A(253/203) and A(226-400), we reached the conclusion that they have a consistent trend (W4> W6> W5> W2> W7> W1> W3). Fluorescence index (f(450/500)) was always utilised to interpret the origin of organic matter in a complex aquatic environment system. Values of f(450/500) are closer to 1.60, indicating that humic substances derived from terrestrial sources and biological sources. Our study demonstrated that reclaimed water of farmland irrigation, industrial wastewater and domestic sewage will definitely influence the humification degree and amount of the aromatic structure of DOM.Understanding of the fate of pollutants, disposed of in streams, is a matter of concern in recent years for the effective control of pollution. Transverse mixing of the pollutants in open channels is arguably more important than the longitudinal mixing and near-field mixing. Several attempts have been made to establish the relationship between the transverse mixing coefficient and bulk channel and flow parameters such as width, depth, shear velocity, friction factor, curvature and sinuosity. This paper presents adaptive neuro fuzzy inference system (ANFIS) approach to predict the transverse mixing coefficient in open channel flows. Available laboratory and field data for the transverse mixing coefficients covering wide range of channel and flow conditions are used for the development and testing of the proposed method. The proposed ANFIS approach produces satisfactory results (R(2)=0.945) compared to the artificial neural network (ANN) model and existing predictors for mixing coefficient.

Determination of optimum relaxation coefficient using finite difference method for groundwater flow

Solution of Laplace’s equation can be done by iteration methods likes Jacobi, Gauss–Seidel, and successive over-relaxation (SOR). There is no new knowledge about the relaxation coefficient (ω) in SOR method. In this paper, we used SOR for solving Laplace’s differential equation with emphasis to obtaining the optimum (minimum) number of iterations with variations of the relaxation coefficient (ω). For this purpose, a code in FORTRAN language has been written to show the solution of a set of equations and its number of iterations. The results demonstrate that the optimum value of ω with minimum iterations is achieved between 1.7 and 1.9. Also, with increasing β = ∆x/∆y from 0.25 to 10, the number of iterations reduced and the optimum value is obtained for β = 2.