Comparison between Multi-Layer Perceptron and Radial Basis Function Networks for Sediment Load Estimation in a Tropical Watershed

Memarian, Hadi; Balasundram, Siva K.

doi:10.4236/jwarp.2012.410102

Cited by 53 publications

(22 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Predicted BY= 0.67×BY+32e+03 (3) Predicted Y = 0.53×Y + 1.5e+03 (4) Overall, according to R 2 , the best model for prediction of BY and Y was MLP model using the equations of 3 and 4. This finding was in agreement with Memarian and Balasundram (2012) who reported that the MLP model showed a slightly better output than RBF network model in predicting suspended sediment discharge, especially in the training process.…”

Section: Prediction Of Biological Yield and Yield Of Barley Using Mlrsupporting

confidence: 92%

“…Among various methods of ANN and learning algorithms, multi-layer perceptron networks (MLP), and radial basis function (RBF) are the most popular neural network models. One of ANN applications is in agriculture science (Heinzow and Richard, 2002;Memarian and Balasundram, 2012). In the past years there has been an increasing interest in ANN modeling in different fields of agriculture, particularly for some areas where conventional statistical modeling failed.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Prediction of biological and grain yield of barley using multiple regression and artificial neural network models

Mokarram¹,

Bijanzadeh

2016

Aust J Crop Sci

View full text Add to dashboard Cite

Prediction of barley yield is an attempt to accurately forecast the outcome of a specific situation, using as input information extracted from a set of data features that potentially describe the situation. In this study, an attempt has been made to analyze and compare multiple linear regression (MLR), and artificial neural network (ANN) including m ulti-layer p erceptron (MLP) and r adial basis function (RBF) models to predicting biological yield (BY) and yield (Y) of barely. Data was collected from the literatures on the subject of barley production that was existed in http://sid.ir website. A total of 10563 data from 17 features were prepared in Excel software sheets. Then, the Matlab software was used to compare the models. Results of MLR model based on R 2 showed that Model 7, with 1000-kernel weight (gr), OC (%), grain/spike, soil pH, N applied (kg/ha), plant height (cm), and irrigation regime (according to FC) and Model 8 with 1000-kernel weight (gr), OC (%), soil pH, grain/spike, HI (%), plant height (cm), irrigation regime (according to FC), and plant density (plant/m 2 ), were the best models for prediction BY and Y of barley, respectively. The highest standardized coefficient (β) for prediction of BY was obtained in 1000-kernel weight (0.621), OC (0.396) and grain/spike (0.385). Also, for prediction of Y, 1000-kernel weight, OC, and grain/spike with 0.547, 0.403, and 0.347 had the highest β, respectively. Among the MLR, MLP and RBF models, MLP model had the highest R 2 values for prediction of BY (R 2 =0.894) and Y (R 2 =0.922). Overall, ANN models can be used to successfully estimate BY and Y from data.

show abstract

Section: Prediction Of Biological Yield and Yield Of Barley Using Mlrsupporting

confidence: 92%

Section: Introductionmentioning

confidence: 99%

Prediction of biological and grain yield of barley using multiple regression and artificial neural network models

Mokarram¹,

Bijanzadeh

2016

Aust J Crop Sci

View full text Add to dashboard Cite

show abstract

“…Clearly, all networks (S1 -S5) showed weak robustness in estimating sediment load with a large magnitude, especially for records higher than 4000 ton/day. Such a limitation in ANN application has also been reported by Hsu et al (1995), Morid et al (2002), Talebizadeh et al (2010) and Memarian and Balasundram (2012). Inefficiency of the ANN -GA model in estimating large magnitudes of sediment load can be attributed to different non-linear relationships governing the process of sediment detachment and final sediment load generated from a watershed.…”

Section: Scenariomentioning

confidence: 92%

An expert integrative approach for sediment load simulation in a tropical watershed

Memarian

Balasundram

Tajbakhsh

2013

Journal of Integrative Environmental Sciences

Self Cite

View full text Add to dashboard Cite

Prediction of highly non-linear behaviour of suspended sediment flow in rivers is of prime importance in environmental studies and watershed management. In this study, the predictive performance of artificial neural network (ANN) integrated with genetic algorithm (GA) was assessed. GA was used to optimize the parameters and architecture of the ANN. Five simulation scenarios (S1 -S5) were developed using daily time series of suspended sediment discharge, water discharge, precipitation and reservoir level. The scenario S1 was composed of only water discharge input. The scenarios S2-S4 were composed of water discharge input and precipitation records at different stations. The inputs water discharge, precipitation and reservoir level formed the last scenario S5. Assessment metrics such as normalized mean square error, correlation coefficient, Nash-Sutcliffe efficiency and trend accuracy were used to evaluate the performance of ANN-GA on the simulation scenarios. Based on error analysis, differences between various scenarios in terms of error metrics were trivial, especially during the testing process. Meanwhile, S1 and S3 showed better accuracy in predicting the trend of sediment load time series, as compared to other scenarios. Based on error and sensitivity analyses, S1 with the Nash -Sutcliff efficiency and correlation coefficient of 0.56 and 0.81, respectively, was chosen as the most appropriate scenario. All networks showed a weak robustness in estimating large magnitudes of sediment load, mostly attributable to scarcity of large observed values in the training data-set. This weakness was also originated from different non-linear relationships governing the process of sediment detachment and final sediment load by a high storm event, as compared to those by low or medium storm events. Furthermore, there was an obvious sediment load overestimation in the 2008 exemplars due to a high level of daily water discharge and the outlined generalization rules. Nevertheless, ANN -GA showed reliable performance for sediment load simulation in the studied watershed.

show abstract

“…Some case studies (e.g. Mustafa et al [10], Memarian and Balasundram [20]) employ only discharge or rainfall as the input. There are no other reasons besides data unavailability.…”

Section: Datamentioning

confidence: 99%

Coupling Singular Spectrum Analysis with Artificial Neural Network to Improve Accuracy of Sediment Load Prediction

Heng¹,

Suetsugi²

2013

JWARP

View full text Add to dashboard Cite

Sediment load estimation is generally required for study and development of water resources system. In this regard, artificial neural network (ANN) is the most widely used modeling tool especially in data-constraint regions. This research attempts to combine SSA (singular spectrum analysis) with ANN, hereafter called SSA-ANN model, with expectation to improve the accuracy of sediment load predicted by the existing ANN approach. Two different catchments located in the Lower Mekong Basin (LMB) were selected for the study and the model performance was measured by several statistical indices. In comparing with ANN, the proposed SSA-ANN model shows its better performance repeatedly in both catchments. In validation stage, SSA-ANN is superior for larger Nash-Sutcliffe Efficiency about 24% in Ban Nong Kiang catchment and 7% in Nam Mae Pun Luang catchment. Other statistical measures of SSA-ANN are better than those of ANN as well. This improvement reveals the importance of SSA which filters noise containing in the raw time series and transforms the original input data to be near normal distribution which is favorable to model simulation. This coupled model is also recommended for the prediction of other water resources variables because extra input data are not required. Only additional computation, time series decomposition, is needed. The proposed technique could be potentially used to minimize the costly operation of sediment measurement in the LMB which is relatively rich in hydrometeorological records.

show abstract

Comparison between Multi-Layer Perceptron and Radial Basis Function Networks for Sediment Load Estimation in a Tropical Watershed

Cited by 53 publications

References 24 publications

Prediction of biological and grain yield of barley using multiple regression and artificial neural network models

Prediction of biological and grain yield of barley using multiple regression and artificial neural network models

An expert integrative approach for sediment load simulation in a tropical watershed

Coupling Singular Spectrum Analysis with Artificial Neural Network to Improve Accuracy of Sediment Load Prediction

Contact Info

Product

Resources

About