Development and application of a decision group Back-Propagation Neural Network for flood forecasting

Chang Shian Chen, Boris Po Tsang Chen, Nai-Fang Chou, Chao Chung Yang

Research output: Contribution to journalArticle

34 Citations (Scopus)

Abstract

Applying a deterministic model to a system simulation often requires an assumption that the system input-output relationship should hold the whole event. However, this assumption does not apply for nonlinear time-variant systems which possible different outputs can be found due to the various input structures. In this study, a Decision Group Back-Propagation Network (DGBPN) is proposed to avoid the prediction risks inherent in deterministic Back-Propagation Neural Network (BPN) models. The DGBPN was combined with many BPN models. The number of BPN models was dependant on the length of the learning data. The operation process of the DGBPN allows each given input vector of model learning to stand for the scheme of the vector space where it is located, and then uses the output estimation ability to select a suitable BPN model for each given input vector scheme from the BPN model candidates as a representative BPN model (also a member of a decision group). The similarity of input vectors between the forecasting pattern (verification pattern) and model learning patterns was employed to select a BPN model from the DGBPN for forecasting. A flood hydrograph forecast was completed using this method. The model developed in this study not only provides the flexibility necessary for a deterministic BPN model to cope with the volatility of the rainfall-runoff process, but also improves adaptability to reduce estimation error when a system input vector over the range of modeling is encountered. The study concludes with implications for theory, research, and practice. The DGBPN performed well and produced fair forecasting results in verification tests at the Wu-Shi watershed in Taiwan.

Original languageEnglish
Pages (from-to)173-182
Number of pages10
JournalJournal of Hydrology
Volume385
Issue number1-4
DOIs
Publication statusPublished - 2010 May 7

Fingerprint

flood forecasting
back propagation
learning
decision
hydrograph

All Science Journal Classification (ASJC) codes

  • Water Science and Technology

Cite this

Chen, Chang Shian ; Chen, Boris Po Tsang ; Chou, Nai-Fang ; Yang, Chao Chung. / Development and application of a decision group Back-Propagation Neural Network for flood forecasting. In: Journal of Hydrology. 2010 ; Vol. 385, No. 1-4. pp. 173-182.
@article{a5cb5e11dc8d440f8567f407d617b259,
title = "Development and application of a decision group Back-Propagation Neural Network for flood forecasting",
abstract = "Applying a deterministic model to a system simulation often requires an assumption that the system input-output relationship should hold the whole event. However, this assumption does not apply for nonlinear time-variant systems which possible different outputs can be found due to the various input structures. In this study, a Decision Group Back-Propagation Network (DGBPN) is proposed to avoid the prediction risks inherent in deterministic Back-Propagation Neural Network (BPN) models. The DGBPN was combined with many BPN models. The number of BPN models was dependant on the length of the learning data. The operation process of the DGBPN allows each given input vector of model learning to stand for the scheme of the vector space where it is located, and then uses the output estimation ability to select a suitable BPN model for each given input vector scheme from the BPN model candidates as a representative BPN model (also a member of a decision group). The similarity of input vectors between the forecasting pattern (verification pattern) and model learning patterns was employed to select a BPN model from the DGBPN for forecasting. A flood hydrograph forecast was completed using this method. The model developed in this study not only provides the flexibility necessary for a deterministic BPN model to cope with the volatility of the rainfall-runoff process, but also improves adaptability to reduce estimation error when a system input vector over the range of modeling is encountered. The study concludes with implications for theory, research, and practice. The DGBPN performed well and produced fair forecasting results in verification tests at the Wu-Shi watershed in Taiwan.",
author = "Chen, {Chang Shian} and Chen, {Boris Po Tsang} and Nai-Fang Chou and Yang, {Chao Chung}",
year = "2010",
month = "5",
day = "7",
doi = "10.1016/j.jhydrol.2010.02.019",
language = "English",
volume = "385",
pages = "173--182",
journal = "Journal of Hydrology",
issn = "0022-1694",
publisher = "Elsevier",
number = "1-4",

}

Development and application of a decision group Back-Propagation Neural Network for flood forecasting. / Chen, Chang Shian; Chen, Boris Po Tsang; Chou, Nai-Fang; Yang, Chao Chung.

In: Journal of Hydrology, Vol. 385, No. 1-4, 07.05.2010, p. 173-182.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Development and application of a decision group Back-Propagation Neural Network for flood forecasting

AU - Chen, Chang Shian

AU - Chen, Boris Po Tsang

AU - Chou, Nai-Fang

AU - Yang, Chao Chung

PY - 2010/5/7

Y1 - 2010/5/7

N2 - Applying a deterministic model to a system simulation often requires an assumption that the system input-output relationship should hold the whole event. However, this assumption does not apply for nonlinear time-variant systems which possible different outputs can be found due to the various input structures. In this study, a Decision Group Back-Propagation Network (DGBPN) is proposed to avoid the prediction risks inherent in deterministic Back-Propagation Neural Network (BPN) models. The DGBPN was combined with many BPN models. The number of BPN models was dependant on the length of the learning data. The operation process of the DGBPN allows each given input vector of model learning to stand for the scheme of the vector space where it is located, and then uses the output estimation ability to select a suitable BPN model for each given input vector scheme from the BPN model candidates as a representative BPN model (also a member of a decision group). The similarity of input vectors between the forecasting pattern (verification pattern) and model learning patterns was employed to select a BPN model from the DGBPN for forecasting. A flood hydrograph forecast was completed using this method. The model developed in this study not only provides the flexibility necessary for a deterministic BPN model to cope with the volatility of the rainfall-runoff process, but also improves adaptability to reduce estimation error when a system input vector over the range of modeling is encountered. The study concludes with implications for theory, research, and practice. The DGBPN performed well and produced fair forecasting results in verification tests at the Wu-Shi watershed in Taiwan.

AB - Applying a deterministic model to a system simulation often requires an assumption that the system input-output relationship should hold the whole event. However, this assumption does not apply for nonlinear time-variant systems which possible different outputs can be found due to the various input structures. In this study, a Decision Group Back-Propagation Network (DGBPN) is proposed to avoid the prediction risks inherent in deterministic Back-Propagation Neural Network (BPN) models. The DGBPN was combined with many BPN models. The number of BPN models was dependant on the length of the learning data. The operation process of the DGBPN allows each given input vector of model learning to stand for the scheme of the vector space where it is located, and then uses the output estimation ability to select a suitable BPN model for each given input vector scheme from the BPN model candidates as a representative BPN model (also a member of a decision group). The similarity of input vectors between the forecasting pattern (verification pattern) and model learning patterns was employed to select a BPN model from the DGBPN for forecasting. A flood hydrograph forecast was completed using this method. The model developed in this study not only provides the flexibility necessary for a deterministic BPN model to cope with the volatility of the rainfall-runoff process, but also improves adaptability to reduce estimation error when a system input vector over the range of modeling is encountered. The study concludes with implications for theory, research, and practice. The DGBPN performed well and produced fair forecasting results in verification tests at the Wu-Shi watershed in Taiwan.

UR - http://www.scopus.com/inward/record.url?scp=77950867287&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77950867287&partnerID=8YFLogxK

U2 - 10.1016/j.jhydrol.2010.02.019

DO - 10.1016/j.jhydrol.2010.02.019

M3 - Article

VL - 385

SP - 173

EP - 182

JO - Journal of Hydrology

JF - Journal of Hydrology

SN - 0022-1694

IS - 1-4

ER -