scholarly journals Reliability-based design and implementation of crow search algorithm for longitudinal dispersion coefficient estimation in rivers

2021 ◽  
Author(s):  
Alireza Ghaemi ◽  
Tahmineh Zhian ◽  
Bahareh Pirzadeh ◽  
Seyedarman Hashemi Monfared ◽  
Amir Mosavi
Keyword(s):  
Failure Probability ◽  
Polynomial Regression ◽  
Dispersion Coefficient ◽  
Search Algorithm ◽  
Quality Parameters ◽  
Longitudinal Dispersion ◽  
Longitudinal Dispersion Coefficient ◽  
Reliability Based Design ◽  
Evolutionary Polynomial Regression ◽  
Probability Prediction

AbstractThe longitudinal dispersion coefficient (LDC) of river pollutants is considered as one of the prominent water quality parameters. In this regard, numerous research studies have been conducted in recent years, and various equations have been extracted based on hydrodynamic and geometric elements. LDC’s estimated values obtained using different equations reveal a significant uncertainty due to this phenomenon’s complexity. In the present study, the crow search algorithm (CSA) is applied to increase the equation’s precision by employing evolutionary polynomial regression (EPR) to model an extensive amount of geometrical and hydraulic data. The results indicate that the CSA improves the performance of EPR in terms of R2 (0.8), Willmott’s index of agreement (0.93), Nash–Sutcliffe efficiency (0.77), and overall index (0.84). In addition, the reliability analysis of the proposed equation (i.e., CSA) reduced the failure probability (Pf) when the value of the failure state containing 50 to 600 m2/s is increasing for the Pf determination using the Monte Carlo simulation. The best-fitted function for correct failure probability prediction was the power with R2 = 0.98 compared with linear and exponential functions.

scholarly journals Investigation of a parameter estimation method for contaminant transport in aquifers

2001 ◽  
Vol 3 (4) ◽  
pp. 203-213 ◽  
Author(s):  
Channa Rajanayaka ◽  
Don Kulasiri
Keyword(s):  
Experimental Data ◽  
Hydraulic Conductivity ◽  
Contaminant Transport ◽  
Dispersion Coefficient ◽  
Estimation Method ◽  
Longitudinal Dispersion ◽  
Longitudinal Dispersion Coefficient ◽  
System Noise ◽  
Estimated Parameters ◽  
Comparison Of The Results

Real world groundwater aquifers are heterogeneous and system variables are not uniformly distributed across the aquifer. Therefore, in the modelling of the contaminant transport, we need to consider the uncertainty associated with the system. Unny presented a method to describe the system by stochastic differential equations and then to estimate the parameters by using the maximum likelihood approach. In this paper, this method was explored by using artificial and experimental data. First a set of data was used to explore the effect of system noise on estimated parameters. The experimental data was used to compare the estimated parameters with the calibrated results. Estimates obtained from artificial data show reasonable accuracy when the system noise is present. The accuracy of the estimates has an inverse relationship to the noise. Hydraulic conductivity estimates in a one-parameter situation give more accurate results than in a two-parameter situation. The effect of the noise on estimates of the longitudinal dispersion coefficient is less compared to the effect on hydraulic conductivity estimates. Comparison of the results of the experimental dataset shows that estimates of the longitudinal dispersion coefficient are similar to the aquifer calibrated results. However, hydraulic conductivity does not provide a similar level of accuracy. The main advantage of the estimation method presented here is its direct dependence on field observations in the presence of reasonably large noise levels.

A Genetic Algorithm-Artificial Neural Network Method for the Prediction of Longitudinal Dispersion Coefficient in Rivers

2011 ◽  
pp. 358-374
Author(s):  
Jianhua Yang ◽  
Evor L. Hines ◽  
Ian Guymer ◽  
Daciana D. Iliescu ◽  
Mark S. Leeson ◽  
...  
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Dispersion Coefficient ◽  
Longitudinal Dispersion ◽  
Longitudinal Dispersion Coefficient ◽  
Artificial Neural ◽  
Input Variables ◽  
Training Examples ◽  
Artificial Neural Network Ann ◽  
Trained Neural Network

In this chapter a novel method, the Genetic Neural Mathematical Method (GNMM), for the prediction of longitudinal dispersion coefficient is presented. This hybrid method utilizes Genetic Algorithms (GAs) to identify variables that are being input into a Multi-Layer Perceptron (MLP) Artificial Neural Network (ANN), which simplifies the neural network structure and makes the training process more efficient. Once input variables are determined, GNMM processes the data using an MLP with the back-propagation algorithm. The MLP is presented with a series of training examples and the internal weights are adjusted in an attempt to model the input/output relationship. GNMM is able to extract regression rules from the trained neural network. The effectiveness of GNMM is demonstrated by means of case study data, which has previously been explored by other authors using various methods. By comparing the results generated by GNMM to those presented in the literature, the effectiveness of this methodology is demonstrated.

Sign in / Sign up

Export Citation Format

Share Document