Novel approach for streamflow forecasting using a hybrid ANFIS-FFA model

Zaher Mundher Yaseen, Isa Ebtehaj, Hossein Bonakdari, Ravinesh C. Deo, Ali Danandeh Mehr, Wan Hanna Melini Wan Mohtar, Lamine Diop, Ahmed El-shafie, Vijay P. Singh

Research output: Contribution to journalArticle

35 Citations (Scopus)

Abstract

The present study proposes a new hybrid evolutionary Adaptive Neuro-Fuzzy Inference Systems (ANFIS) approach for monthly streamflow forecasting. The proposed method is a novel combination of the ANFIS model with the firefly algorithm as an optimizer tool to construct a hybrid ANFIS-FFA model. The results of the ANFIS-FFA model is compared with the classical ANFIS model, which utilizes the fuzzy c-means (FCM) clustering method in the Fuzzy Inference Systems (FIS) generation. The historical monthly streamflow data for Pahang River, which is a major river system in Malaysia that characterized by highly stochastic hydrological patterns, is used in the study. Sixteen different input combinations with one to five time-lagged input variables are incorporated into the ANFIS-FFA and ANFIS models to consider the antecedent seasonal variations in historical streamflow data. The mean absolute error (MAE), root mean square error (RMSE) and correlation coefficient (r) are used to evaluate the forecasting performance of ANFIS-FFA model. In conjunction with these metrics, the refined Willmott's Index (Drefined), Nash-Sutcliffe coefficient (ENS) and Legates and McCabes Index (ELM) are also utilized as the normalized goodness-of-fit metrics. Comparison of the results reveals that the FFA is able to improve the forecasting accuracy of the hybrid ANFIS-FFA model (r = 1; RMSE = 0.984; MAE = 0.364; ENS = 1; ELM = 0.988; Drefined = 0.994) applied for the monthly streamflow forecasting in comparison with the traditional ANFIS model (r = 0.998; RMSE = 3.276; MAE = 1.553; ENS = 0.995; ELM = 0.950; Drefined = 0.975). The results also show that the ANFIS-FFA is not only superior to the ANFIS model but also exhibits a parsimonious modelling framework for streamflow forecasting by incorporating a smaller number of input variables required to yield the comparatively better performance. It is construed that the FFA optimizer can thus surpass the accuracy of the traditional ANFIS model in general, and is able to remove the false (inaccurately) forecasted data in the ANFIS model for extremely low flows. The present results have wider implications not only for streamflow forecasting purposes, but also for other hydro-meteorological forecasting variables requiring only the historical data input data, and attaining a greater level of predictive accuracy with the incorporation of the FFA algorithm as an optimization tool in an ANFIS model.

Original languageEnglish
Pages (from-to)263-276
Number of pages14
JournalJournal of Hydrology
Volume554
DOIs
Publication statusPublished - 1 Nov 2017

Fingerprint

streamflow
low flow
river system
seasonal variation

Keywords

  • ANFIS-FFA
  • Antecedent seasonal variations
  • Streamflow forecasting
  • Tropical environment

ASJC Scopus subject areas

  • Water Science and Technology

Cite this

Yaseen, Z. M., Ebtehaj, I., Bonakdari, H., Deo, R. C., Danandeh Mehr, A., Wan Mohtar, W. H. M., ... Singh, V. P. (2017). Novel approach for streamflow forecasting using a hybrid ANFIS-FFA model. Journal of Hydrology, 554, 263-276. https://doi.org/10.1016/j.jhydrol.2017.09.007

Novel approach for streamflow forecasting using a hybrid ANFIS-FFA model. / Yaseen, Zaher Mundher; Ebtehaj, Isa; Bonakdari, Hossein; Deo, Ravinesh C.; Danandeh Mehr, Ali; Wan Mohtar, Wan Hanna Melini; Diop, Lamine; El-shafie, Ahmed; Singh, Vijay P.

In: Journal of Hydrology, Vol. 554, 01.11.2017, p. 263-276.

Research output: Contribution to journalArticle

Yaseen, ZM, Ebtehaj, I, Bonakdari, H, Deo, RC, Danandeh Mehr, A, Wan Mohtar, WHM, Diop, L, El-shafie, A & Singh, VP 2017, 'Novel approach for streamflow forecasting using a hybrid ANFIS-FFA model', Journal of Hydrology, vol. 554, pp. 263-276. https://doi.org/10.1016/j.jhydrol.2017.09.007
Yaseen, Zaher Mundher ; Ebtehaj, Isa ; Bonakdari, Hossein ; Deo, Ravinesh C. ; Danandeh Mehr, Ali ; Wan Mohtar, Wan Hanna Melini ; Diop, Lamine ; El-shafie, Ahmed ; Singh, Vijay P. / Novel approach for streamflow forecasting using a hybrid ANFIS-FFA model. In: Journal of Hydrology. 2017 ; Vol. 554. pp. 263-276.
@article{e3f2aaa2f9de4194bcfb5f3de8ace96a,
title = "Novel approach for streamflow forecasting using a hybrid ANFIS-FFA model",
abstract = "The present study proposes a new hybrid evolutionary Adaptive Neuro-Fuzzy Inference Systems (ANFIS) approach for monthly streamflow forecasting. The proposed method is a novel combination of the ANFIS model with the firefly algorithm as an optimizer tool to construct a hybrid ANFIS-FFA model. The results of the ANFIS-FFA model is compared with the classical ANFIS model, which utilizes the fuzzy c-means (FCM) clustering method in the Fuzzy Inference Systems (FIS) generation. The historical monthly streamflow data for Pahang River, which is a major river system in Malaysia that characterized by highly stochastic hydrological patterns, is used in the study. Sixteen different input combinations with one to five time-lagged input variables are incorporated into the ANFIS-FFA and ANFIS models to consider the antecedent seasonal variations in historical streamflow data. The mean absolute error (MAE), root mean square error (RMSE) and correlation coefficient (r) are used to evaluate the forecasting performance of ANFIS-FFA model. In conjunction with these metrics, the refined Willmott's Index (Drefined), Nash-Sutcliffe coefficient (ENS) and Legates and McCabes Index (ELM) are also utilized as the normalized goodness-of-fit metrics. Comparison of the results reveals that the FFA is able to improve the forecasting accuracy of the hybrid ANFIS-FFA model (r = 1; RMSE = 0.984; MAE = 0.364; ENS = 1; ELM = 0.988; Drefined = 0.994) applied for the monthly streamflow forecasting in comparison with the traditional ANFIS model (r = 0.998; RMSE = 3.276; MAE = 1.553; ENS = 0.995; ELM = 0.950; Drefined = 0.975). The results also show that the ANFIS-FFA is not only superior to the ANFIS model but also exhibits a parsimonious modelling framework for streamflow forecasting by incorporating a smaller number of input variables required to yield the comparatively better performance. It is construed that the FFA optimizer can thus surpass the accuracy of the traditional ANFIS model in general, and is able to remove the false (inaccurately) forecasted data in the ANFIS model for extremely low flows. The present results have wider implications not only for streamflow forecasting purposes, but also for other hydro-meteorological forecasting variables requiring only the historical data input data, and attaining a greater level of predictive accuracy with the incorporation of the FFA algorithm as an optimization tool in an ANFIS model.",
keywords = "ANFIS-FFA, Antecedent seasonal variations, Streamflow forecasting, Tropical environment",
author = "Yaseen, {Zaher Mundher} and Isa Ebtehaj and Hossein Bonakdari and Deo, {Ravinesh C.} and {Danandeh Mehr}, Ali and {Wan Mohtar}, {Wan Hanna Melini} and Lamine Diop and Ahmed El-shafie and Singh, {Vijay P.}",
year = "2017",
month = "11",
day = "1",
doi = "10.1016/j.jhydrol.2017.09.007",
language = "English",
volume = "554",
pages = "263--276",
journal = "Journal of Hydrology",
issn = "0022-1694",
publisher = "Elsevier",

}

TY - JOUR

T1 - Novel approach for streamflow forecasting using a hybrid ANFIS-FFA model

AU - Yaseen, Zaher Mundher

AU - Ebtehaj, Isa

AU - Bonakdari, Hossein

AU - Deo, Ravinesh C.

AU - Danandeh Mehr, Ali

AU - Wan Mohtar, Wan Hanna Melini

AU - Diop, Lamine

AU - El-shafie, Ahmed

AU - Singh, Vijay P.

PY - 2017/11/1

Y1 - 2017/11/1

N2 - The present study proposes a new hybrid evolutionary Adaptive Neuro-Fuzzy Inference Systems (ANFIS) approach for monthly streamflow forecasting. The proposed method is a novel combination of the ANFIS model with the firefly algorithm as an optimizer tool to construct a hybrid ANFIS-FFA model. The results of the ANFIS-FFA model is compared with the classical ANFIS model, which utilizes the fuzzy c-means (FCM) clustering method in the Fuzzy Inference Systems (FIS) generation. The historical monthly streamflow data for Pahang River, which is a major river system in Malaysia that characterized by highly stochastic hydrological patterns, is used in the study. Sixteen different input combinations with one to five time-lagged input variables are incorporated into the ANFIS-FFA and ANFIS models to consider the antecedent seasonal variations in historical streamflow data. The mean absolute error (MAE), root mean square error (RMSE) and correlation coefficient (r) are used to evaluate the forecasting performance of ANFIS-FFA model. In conjunction with these metrics, the refined Willmott's Index (Drefined), Nash-Sutcliffe coefficient (ENS) and Legates and McCabes Index (ELM) are also utilized as the normalized goodness-of-fit metrics. Comparison of the results reveals that the FFA is able to improve the forecasting accuracy of the hybrid ANFIS-FFA model (r = 1; RMSE = 0.984; MAE = 0.364; ENS = 1; ELM = 0.988; Drefined = 0.994) applied for the monthly streamflow forecasting in comparison with the traditional ANFIS model (r = 0.998; RMSE = 3.276; MAE = 1.553; ENS = 0.995; ELM = 0.950; Drefined = 0.975). The results also show that the ANFIS-FFA is not only superior to the ANFIS model but also exhibits a parsimonious modelling framework for streamflow forecasting by incorporating a smaller number of input variables required to yield the comparatively better performance. It is construed that the FFA optimizer can thus surpass the accuracy of the traditional ANFIS model in general, and is able to remove the false (inaccurately) forecasted data in the ANFIS model for extremely low flows. The present results have wider implications not only for streamflow forecasting purposes, but also for other hydro-meteorological forecasting variables requiring only the historical data input data, and attaining a greater level of predictive accuracy with the incorporation of the FFA algorithm as an optimization tool in an ANFIS model.

AB - The present study proposes a new hybrid evolutionary Adaptive Neuro-Fuzzy Inference Systems (ANFIS) approach for monthly streamflow forecasting. The proposed method is a novel combination of the ANFIS model with the firefly algorithm as an optimizer tool to construct a hybrid ANFIS-FFA model. The results of the ANFIS-FFA model is compared with the classical ANFIS model, which utilizes the fuzzy c-means (FCM) clustering method in the Fuzzy Inference Systems (FIS) generation. The historical monthly streamflow data for Pahang River, which is a major river system in Malaysia that characterized by highly stochastic hydrological patterns, is used in the study. Sixteen different input combinations with one to five time-lagged input variables are incorporated into the ANFIS-FFA and ANFIS models to consider the antecedent seasonal variations in historical streamflow data. The mean absolute error (MAE), root mean square error (RMSE) and correlation coefficient (r) are used to evaluate the forecasting performance of ANFIS-FFA model. In conjunction with these metrics, the refined Willmott's Index (Drefined), Nash-Sutcliffe coefficient (ENS) and Legates and McCabes Index (ELM) are also utilized as the normalized goodness-of-fit metrics. Comparison of the results reveals that the FFA is able to improve the forecasting accuracy of the hybrid ANFIS-FFA model (r = 1; RMSE = 0.984; MAE = 0.364; ENS = 1; ELM = 0.988; Drefined = 0.994) applied for the monthly streamflow forecasting in comparison with the traditional ANFIS model (r = 0.998; RMSE = 3.276; MAE = 1.553; ENS = 0.995; ELM = 0.950; Drefined = 0.975). The results also show that the ANFIS-FFA is not only superior to the ANFIS model but also exhibits a parsimonious modelling framework for streamflow forecasting by incorporating a smaller number of input variables required to yield the comparatively better performance. It is construed that the FFA optimizer can thus surpass the accuracy of the traditional ANFIS model in general, and is able to remove the false (inaccurately) forecasted data in the ANFIS model for extremely low flows. The present results have wider implications not only for streamflow forecasting purposes, but also for other hydro-meteorological forecasting variables requiring only the historical data input data, and attaining a greater level of predictive accuracy with the incorporation of the FFA algorithm as an optimization tool in an ANFIS model.

KW - ANFIS-FFA

KW - Antecedent seasonal variations

KW - Streamflow forecasting

KW - Tropical environment

UR - http://www.scopus.com/inward/record.url?scp=85029511044&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85029511044&partnerID=8YFLogxK

U2 - 10.1016/j.jhydrol.2017.09.007

DO - 10.1016/j.jhydrol.2017.09.007

M3 - Article

AN - SCOPUS:85029511044

VL - 554

SP - 263

EP - 276

JO - Journal of Hydrology

JF - Journal of Hydrology

SN - 0022-1694

ER -