Flood flow forecasting using ANN, ANFIS and regression models

Rezaeianzadeh, Mehdi; Tabari, Hossein; Yazdi, Azam Arabi; Işık, Sabahattin; Kalin, Latif

doi:10.1007/s00521-013-1443-6

Cited by 189 publications

(94 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Prediction models with artificial neural networks (ANNs) have been employed for flood susceptibility evaluation by various scholars (Kia et al, 2012;Seckin et al, 2013a;Rezaeianzadeh et al, 2014;Radmehr and Araghinejad, 2014); previous works have shown that an ANN is a capable nonlinear modeling tool. Nevertheless, ANN learning is prone to overfitting, and its performance has been shown to be inferior to that of support vector machines (SVMs; Hoang and Pham, 2016).…”

Section: A Review Of Related Work On Flood Susceptibility Predictionmentioning

confidence: 99%

A Bayesian framework based on a Gaussian mixture model and radial-basis-function Fisher discriminant analysis (BayGmmKda V1.1) for spatial prediction of floods

Bui

Hoang

2017

Geosci. Model Dev.

View full text Add to dashboard Cite

Abstract. In this study, a probabilistic model, named as BayGmmKda, is proposed for flood susceptibility assessment in a study area in central Vietnam. The new model is a Bayesian framework constructed by a combination of a Gaussian mixture model (GMM), radial-basis-function Fisher discriminant analysis (RBFDA), and a geographic information system (GIS) database. In the Bayesian framework, GMM is used for modeling the data distribution of flood-influencing factors in the GIS database, whereas RBFDA is utilized to construct a latent variable that aims at enhancing the model performance. As a result, the posterior probabilistic output of the BayGmmKda model is used as flood susceptibility index. Experiment results showed that the proposed hybrid framework is superior to other benchmark models, including the adaptive neuro-fuzzy inference system and the support vector machine. To facilitate the model implementation, a software program of BayGmmKda has been developed in MATLAB. The BayGmmKda program can accurately establish a flood susceptibility map for the study region. Accordingly, local authorities can overlay this susceptibility map onto various land-use maps for the purpose of land-use planning or management.

show abstract

Section: A Review Of Related Work On Flood Susceptibility Predictionmentioning

confidence: 99%

A Bayesian framework based on a Gaussian mixture model and radial-basis-function Fisher discriminant analysis (BayGmmKda V1.1) for spatial prediction of floods

Bui

Hoang

2017

Geosci. Model Dev.

View full text Add to dashboard Cite

show abstract

“…Sanikhani et al (2015) modelled two different ANFIS models -the ANFIS with grid partition (ANFIS-GP) and the ANFIS with subtractive clustering (ANFIS-SC) with gene expression programming (GEP) -in order to forecast one, two, and three months in advance the lake level fluctuations at Manyas and Tuz lakes in Turkey [10]. Rezaeianzadeh et al (2014) applied the ANN, ANFIS and regression models for forecasting the maximum daily flow at the outlet of the Khosrow Shirin watershed, located at the Fars Province in Iran [11]. Tien Bui et al (2016) proposed a neural fuzzy inference system and metaheuristic optimization for the flood susceptibility modelling.…”

Section: Introductionmentioning

confidence: 99%

Application of artificial neural networks for hydrological modelling in Karst

Lakušić¹

2018

JCE

View full text Add to dashboard Cite

Application of artificial neural networks for hydrological modelling in karstThe possibility of short-term water flow forecasting in a karst region is presented in this paper. Four state-of-the-art machine learning algorithms are used for the one day ahead forecasting: multi-layer perceptron neural network, radial basis function neural network, support vector machines for regression (SVR), and adaptive neuro fuzzy inference system (ANFIS). The results show that the ANFIS model outperforms other algorithms when the root mean square error and mean absolute error are used as quality indicators. Primjena neuronskih mreža za hidrološko modeliranje u krškom području U ovom radu je razmatrana mogućnost kratkoročnog predviđanja protoka vode u krškom području. Četiri suvremena algoritma strojnog učenja su korištena za predviđanje protoka jedan dan unaprijed i to: neuronska mreža s višeslojnim perceptronom, neuronska mreža s radijalnom aktivacijskom funkcijom, metoda potpornih vektora za regresiju (SVR), adaptivni neuroneizraziti (fuzzy) sustav zaključivanja (ANFIS). Rezultati pokazuju da ANFIS model nadmašuje ostale algoritme kada se korijen srednje kvadratne popogreške i srednja apsolutna pogreška upotrijebe kao mjere kvalitete.

show abstract

“…About 19% of the total land area is eroded due to the public road system [19]. Some researchers reported that most of the erosion occurred during the rainy season after a disturbance and half of the erosions were caused by roads from the logging operation.…”

Section: Introductionmentioning

confidence: 99%

Rainfall Induced Erosion Assessment on Hill Slope of GCE

Jaafar¹,

Islam²,

Hin³

et al. 2018

Preprint

View full text Add to dashboard Cite

This study investigates the effect of northeast monsoon season rainfall to erosion on the hillslope of the Guthrie corridor expressway (GCE), Malaysia. The main focus of this study is to develop a relationship among the vegetation cover, slope steepness, and soil loss. A plot scale basis study area is established on hillslope along the GCE. To account for various vegetation densities, a number of five plots are proposed with various percentage of vegetation density. Experimental equipment is set up at those plots to allow collection of rainfall and soil loss for duration of five months, i.e. from November 2014 till March 2015, with twice collection per month. Based on the collected data, the effect of vegetation associated to soil loss is examined followed by propose of a new model. There are few points to summarize based on this study; 1) the observed rainfall pattern shows that mostly the recorded rainfall depth is about 5 mm with 20 minutes duration of rainfall; 2) plot without vegetation cover (i.e. NBNM) yields greater soil loss and with maximum runoff whereas the least is exhibited by the plot having natural dense microbe (NDM). Meanwhile, the planted vegetation is found affected the rate of soil loss; 3) a positive relationship is found between runoff and slope steepness regardless of the vegetation cover type; and 4) A newly proposed model reveals that higher Fvs value implies a low density of vegetation; hence contributes to more soil loss.

show abstract

Flood flow forecasting using ANN, ANFIS and regression models

Cited by 189 publications

References 39 publications

A Bayesian framework based on a Gaussian mixture model and radial-basis-function Fisher discriminant analysis (BayGmmKda V1.1) for spatial prediction of floods

A Bayesian framework based on a Gaussian mixture model and radial-basis-function Fisher discriminant analysis (BayGmmKda V1.1) for spatial prediction of floods

Application of artificial neural networks for hydrological modelling in Karst

Rainfall Induced Erosion Assessment on Hill Slope of GCE

Contact Info

Product

Resources

About