Streamflow forecasting by combining neural networks and fuzzy models using advanced methods of input variable selection

Dariane, Alireza B.; Azimi, Sh.

doi:10.2166/hydro.2017.076

Cited by 46 publications

(13 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The performance of the proposed hybrid model compared to that of least square SVM (LSSVM) model and the results were reported to be good. On this premise, an attempt towards the integration of ELM model with the genetic algorithm (GA) (an input selection algorithm) has been made by [30] for the forecasting of the monthly streamflow at the Ajichai Basin. The outcome of the study showed that the integration of ELM with GA enhanced the forecasting accuracy of the ELM model.…”

Section: Introductionmentioning

confidence: 99%

Non-Linear Input Variable Selection Approach Integrated With Non-Tuned Data Intelligence Model for Streamflow Pattern Simulation

et al. 2019

View full text Add to dashboard Cite

Streamflow modeling is considered as an essential component for water resources planning and management. There are numerous challenges related to streamflow prediction that are facing water resources engineers. These challenges due to the complex processes associated with several natural variables such as non-stationarity, non-linearity, and randomness. In this study, a new model is proposed to predict long-term streamflow. Several lags that cover several years are abstracted using the potential of Extreme Gradient Boosting (XGB) then after the selected inputs variables are imposed into the predictive model (i.e., Extreme Learning Machine (ELM)). The proposed model is compared with the stand-alone schema in which the optimum lags of the variables are supplied into the XGB and ELM models. Hydrological variables including rainfall, temperature and evapotranspiration are used to build the model and predict the streamflow at Goksu-Himmeti basin in Turkey. The results showed that XGB model performed an excellent result in which can be used for predicting the streamflow pattern. Also, it is clear from the attained results that the accuracy of the streamflow prediction using XGB technique could be improved when the high number of lags was used. However, the implementation of the XGB is tree-based technique in which several issues could be raised such as overfitting problem. The proposed schema XGBELM in which XGB approach is selected the correlated inputs and ranking them according to their importance; then after, the selected inputs are supplied into the ELM model for the prediction process. The XGBELM model outperformed the stand-alone schema of both XGB and ELM models and the high-lagged schema of the XGB. It is important to indicate that the XGBELM model found to improve the prediction ability with minimum variables number.INDEX TERMS Correlated variables, non-linear XGB approach, extreme learning machine, streamflow simulation.

show abstract

Section: Introductionmentioning

confidence: 99%

Non-Linear Input Variable Selection Approach Integrated With Non-Tuned Data Intelligence Model for Streamflow Pattern Simulation

et al. 2019

View full text Add to dashboard Cite

show abstract

“…ey are almost 1 for validation and more than 0.8 for all other cases. ese values are acceptable in the time-dependent highly nonlinear climate analyses [79][80][81][82]. However, out of them, LM algorithm has outperformed other training algorithms in the computational performance (highlighted in grey).…”

Section: Resultsmentioning

confidence: 99%

Inflow Forecast of Iranamadu Reservoir, Sri Lanka, under Projected Climate Scenarios Using Artificial Neural Networks

Karunanayake

Gunathilake

Rathnayake

2020

Applied Computational Intelligence and Soft Computing

View full text Add to dashboard Cite

Prediction of water resources for future years takes much attention from the water resources planners and relevant authorities. However, traditional computational models like hydrologic models need many data about the catchment itself. Sometimes these important data on catchments are not available due to many reasons. Therefore, artificial neural networks (ANNs) are useful soft computing tools in predicting real-world scenarios, such as forecasting future water availability from a catchment, in the absence of intensive data, which are required for modeling practices in the context of hydrology. These ANNs are capable of building relationships to nonlinear real-world problems using available data and then to use that built relationship to forecast future needs. Even though Sri Lanka has an extensive usage of water resources for many activities, including drinking water supply, irrigation, hydropower development, navigation, and many other recreational purposes, forecasting studies for water resources are not being carried out. Therefore, there is a significant gap in forecasting water availability and water needs in the context of Sri Lanka. Thus, this paper presents an artificial neural network model to forecast the inflows of one of the most important reservoirs in northern Sri Lanka using the upstream catchment’s rainfall. Future rainfall data are extracted using regional climate models for the years 2021–2050 and the inflows of the reservoir are forecasted using the validated neural network model. Several training algorithms including Levenberg–Marquardt (LM), BFGS quasi-Newton (BFG), scaled conjugate gradient (SCG) have been used to find the best fitting training algorithm to the prediction process of the inflows against the measured inflows. Results revealed that the LM training algorithm outperforms the other tests algorithm in developing the prediction model. In addition, the forecasted results using the projected climate scenarios clearly showcase the benefit of using the forecasting model in solving future water resource management to avoid or to minimize future water scarcity. Therefore, the validated model can effectively be used for proper planning of the proposed drinking water supply scheme to the nearby urban city, Jaffna in northern Sri Lanka.

show abstract

“…In the literature, there is a set of standard coefficients that provides agreement between measurements and predictions through a set of single parameter values. In most of the cases, authors report that their model is suitably based on the comparison of statistical parameters composed of measurement and model output data series by consideration one or few of the well-established agreement or association metrics among which are the most commonly used, accepted, and recommended ones are bias (BI), Percent Bias (PBI), coefficient of determination (R 2 ), mean square error (MSE) or root mean square error (RMSE), correlation coefficient (CC), Nash-Sutcliffe efficiency (NSE) and index of agreement (d) (Pearson 1895;Nash & Sutcliffe 1970;Willmott 1981;Santhi et al 2001;Gupta et al 2002;Moriasi et al 2007;Van Liew et al 2007;Özger & Kabataş2015;Tian et al 2015;Zhang et al 2016;Dariane & Azimi 2018).…”

Section: Introductionmentioning

confidence: 99%

A new innovative method for model efficiency performance

2021

View full text Add to dashboard Cite

In every aspect of scientific research, model predictions need calibration and validation as their representativity of the record measurement. In the literature, there are a myriad of formulations, empirical expressions, algorithms and software for model efficiency assessment. In general, model predictions are curve fitting procedures with a set of assumptions that are not cared for sensitively in many studies, but only a single value comparison between the measurements and predictions is taken into consideration, and then the researcher makes the decision as for the model efficiency. Among the classical statistical efficiency formulations, the most widely used ones are bias (BI), mean square error (MSE), correlation coefficient (CC) and Nash-Sutcliffe efficiency (NSE) procedures all of which are embedded within the visual inspection and numerical analysis (VINAM) square graph as measurements versus predictions scatter diagram. The VINAM provides a set of verbal interpretations and then numerical improvements embracing all the previous statistical efficiency formulations. The fundamental criterion in the VINAM is 1:1 (45°) main diagonal along which all visual, science philosophical, logical, rational and mathematical procedures boil down for model validation. The application of the VINAM approach is presented for artificial neural network (ANN) and adaptive network-based fuzzy inference system (ANFIS) model predictions.

show abstract

Streamflow forecasting by combining neural networks and fuzzy models using advanced methods of input variable selection

Cited by 46 publications

References 29 publications

Non-Linear Input Variable Selection Approach Integrated With Non-Tuned Data Intelligence Model for Streamflow Pattern Simulation

Non-Linear Input Variable Selection Approach Integrated With Non-Tuned Data Intelligence Model for Streamflow Pattern Simulation

Inflow Forecast of Iranamadu Reservoir, Sri Lanka, under Projected Climate Scenarios Using Artificial Neural Networks

A new innovative method for model efficiency performance

Contact Info

Product

Resources

About