A comparative assessment between artificial neural network, neuro-fuzzy, and support vector machine models in splash erosion modelling under simulation circumstances

Abstract Splash erosion, as the first step of soil erosion, causes the movement of the soil particles and lumps and is considered an important process in soil erosion. Given the complexity of this process in nature, one way of identifying and modeling the process is to use a rainfall simulator and to study it under laboratory circumstances. For this purpose, transported material was measured with various rainfall intensities and different amounts of poly-acryl-amide. In the next step, artificial neural network (ANN), adaptive neuro-fuzzy inference system (ANFIS), and support vector machine (SVM) were used to model the transported materials. The results showed that among the three methods, the best values of evaluation criteria were related to SVM, and ANFIS respectively. Among the three studied durations, the experiment with a duration of 30 minutes received the best results. The results based on available data showed by increasing the number of membership functions, over-fitting happens in the ANFIS method. To reduce the complexity of the model and the likelihood of over-fitting, some rules were eliminated. The results showed that the performance of the model improved by eliminating some rules.

Download Full-text

A comparative study of artificial neural network, adaptive neuro fuzzy inference system and support vector machine for forecasting river flow in the semiarid mountain region

Journal of Hydrology ◽

10.1016/j.jhydrol.2013.11.054 ◽

2014 ◽

Vol 509 ◽

pp. 379-386 ◽

Cited By ~ 156

Author(s):

Zhibin He ◽

Xiaohu Wen ◽

Hu Liu ◽

Jun Du

Keyword(s):

Neural Network ◽

Artificial Neural Network ◽

Support Vector Machine ◽

Comparative Study ◽

Fuzzy Inference System ◽

River Flow ◽

Fuzzy Inference ◽

Support Vector ◽

Inference System ◽

Neuro Fuzzy

Download Full-text

Flood hydrograph modeling using artificial neural network and adaptive neuro-fuzzy inference system based on rainfall components

Arabian Journal of Geosciences ◽

10.1007/s12517-021-06683-6 ◽

2021 ◽

Vol 14 (5) ◽

Author(s):

Saeid Janizadeh ◽

Mehdi Vafakhah

Keyword(s):

Neural Network ◽

Artificial Neural Network ◽

Fuzzy Inference System ◽

Fuzzy Inference ◽

Inference System ◽

Neuro Fuzzy ◽

Flood Hydrograph ◽

Artificial Neural

Download Full-text

Comparison of different modeling methods toward predictive capability evaluation of the bonding strength of wood laminated products

Proceedings of the Institution of Mechanical Engineers Part E Journal of Process Mechanical Engineering ◽

10.1177/09544089211053074 ◽

2021 ◽

pp. 095440892110530

Author(s):

Morteza Nazerian ◽

Seyed Ali Razavi ◽

Ali Partovinia ◽

Elham Vatankhah ◽

Zahra Razmpour

Keyword(s):

Neural Network ◽

Artificial Neural Network ◽

Response Surface Methodology ◽

Response Surface ◽

Bonding Strength ◽

Fuzzy Inference ◽

Molar Ratio ◽

Inference System ◽

Neuro Fuzzy ◽

Artificial Neural

The main aim of this study is usability evaluation of different approaches, including response surface methodoloy, adaptive neuro-fuzzy inference system, and artificial neural network models to predict and evaluate the bonding strength of glued laminated timber (glulam) manufactured using walnut wood layers and a natural adhesive (oxidized starch adhesive), with respect to this fact that using the modified starch can decrease the formaldehyde emission. In this survey, four variables taken as the input data include the molar ratio of formaldehyde to urea (1.12–1.52), nanocellulose content (0%–4%, based on the dry weight of the adhesive), the mass ratio of the oxidized starch adhesive to the urea formaldehyde resin (30:70–70:30), and the press time (4–8 min). In order to find the best predictive performance of each selected evaluation approach, different membership functions were used. The optimal results were obtained when the molar ratio, nanocellulose content, mass ratio of the oxidised starch, and press time were set at 1.22, 3%, 70:30, and 7 min, respectively. Based on the performance criteria including root mean square error (RMSE) and mean absolute percentage error (MAPE) obtained from the modeling of response surface methodology, adaptive neuro-fuzzy inference network, and artificial neural network, it became evident that response surface methodology could offer a better prediction of the response with the lowest level of errors. Beside, artificial neural network and adaptive neuro-fuzzy inference system support the response surface methodology approach to evaluate bonding strength response with high precision as well as to determine the optimal point in fabrication of laminated products.

Download Full-text

Predicting Rainfall-Induced Soil Erosion Based on a Hybridization of Adaptive Differential Evolution and Support Vector Machine Classification

Mathematical Problems in Engineering ◽

10.1155/2021/6647829 ◽

2021 ◽

Vol 2021 ◽

pp. 1-20

Author(s):

Tuan Vu Dinh ◽

Hieu Nguyen ◽

Xuan-Linh Tran ◽

Nhat-Duc Hoang

Keyword(s):

Neural Network ◽

Machine Learning ◽

Artificial Neural Network ◽

Support Vector Machine ◽

Soil Erosion ◽

Differential Evolution ◽

Influencing Factors ◽

Support Vector ◽

Adaptive Differential Evolution ◽

Artificial Neural

Soil erosion induced by rainfall is a critical problem in many regions in the world, particularly in tropical areas where the annual rainfall amount often exceeds 2000 mm. Predicting soil erosion is a challenging task, subjecting to variation of soil characteristics, slope, vegetation cover, land management, and weather condition. Conventional models based on the mechanism of soil erosion processes generally provide good results but are time-consuming due to calibration and validation. The goal of this study is to develop a machine learning model based on support vector machine (SVM) for soil erosion prediction. The SVM serves as the main prediction machinery establishing a nonlinear function that maps considered influencing factors to accurate predictions. In addition, in order to improve the accuracy of the model, the history-based adaptive differential evolution with linear population size reduction and population-wide inertia term (L-SHADE-PWI) is employed to find an optimal set of parameters for SVM. Thus, the proposed method, named L-SHADE-PWI-SVM, is an integration of machine learning and metaheuristic optimization. For the purpose of training and testing the method, a dataset consisting of 236 samples of soil erosion in Northwest Vietnam is collected with 10 influencing factors. The training set includes 90% of the original dataset; the rest of the dataset is reserved for assessing the generalization capability of the model. The experimental results indicate that the newly developed L-SHADE-PWI-SVM method is a competitive soil erosion predictor with superior performance statistics. Most importantly, L-SHADE-PWI-SVM can achieve a high classification accuracy rate of 92%, which is much better than that of backpropagation artificial neural network (87%) and radial basis function artificial neural network (78%).

Download Full-text