3 Hours ahead of time flood water level prediction using NNARX structure: Case study pahang

Noor Ashikin Rohaimi, Fazlina Ahmat Ruslan, Ramli Adnan

Research output: ResearchConference contribution

Abstract

Flood is defined as an overflow of large amount of water beyond its normal limits. Therefore, it has become threat to people's life and can cause damages to properties. However, in Malaysia, the only existing flood warning system are the alarming system which only notify residents nearby flood location to evacuate only when flood occur. Thus, flood water level prediction is very much needed in order to prevent flood disaster to happen. One of the effective techniques which frequently used to solve nonlinear cases such as flood is Artificial Neural Network (ANN). Therefore, this paper proposed 3 hours flood water level prediction using Neural Network Autoregressive model with Exogenous Input (NNARX) technique. The area involved in this study was along Pahang river basin where the flood location is situated at Mentakab. Four input parameters were fed in to the NNARX model to predict flood 3 hours ahead of time. The inputs were carefully selected during flood events. The samples used for training, validation and testing stage are 1553, 1997 and 4000 samples respectively. The NNARX flood prediction model developed using Matlab Neural Network Toolbox. Result shows satisfactory performance with low error measures.

LanguageEnglish
Title of host publication2016 7th IEEE Control and System Graduate Research Colloquium, ICSGRC 2016 - Proceeding
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages98-103
Number of pages6
ISBN (Electronic)9781509011759
DOIs
StatePublished - Jan 10 2017
Event2016 7th IEEE Control and System Graduate Research Colloquium, ICSGRC 2016 - Shah Alam, Selangor, Malaysia
Duration: Aug 8 2016 → …

Other

Other2016 7th IEEE Control and System Graduate Research Colloquium, ICSGRC 2016
CountryMalaysia
CityShah Alam, Selangor
Period8/8/16 → …

Fingerprint

Water
Prediction
predictions
water
Water levels
Neural networks
Neural Networks
flood predictions
Malaysia
river basins
warning systems
disasters
education
damage
causes
Overflow
Autoregressive Model
Disaster
Prediction Model
Network Model

All Science Journal Classification (ASJC) codes

  • Engineering (miscellaneous)
  • Control and Optimization
  • Modelling and Simulation
  • Instrumentation

Cite this

Rohaimi, N. A., Ruslan, F. A., & Adnan, R. (2017). 3 Hours ahead of time flood water level prediction using NNARX structure: Case study pahang. In 2016 7th IEEE Control and System Graduate Research Colloquium, ICSGRC 2016 - Proceeding (pp. 98-103). [7813309] Institute of Electrical and Electronics Engineers Inc.. DOI: 10.1109/ICSGRC.2016.7813309

3 Hours ahead of time flood water level prediction using NNARX structure : Case study pahang. / Rohaimi, Noor Ashikin; Ruslan, Fazlina Ahmat; Adnan, Ramli.

2016 7th IEEE Control and System Graduate Research Colloquium, ICSGRC 2016 - Proceeding. Institute of Electrical and Electronics Engineers Inc., 2017. p. 98-103 7813309.

Research output: ResearchConference contribution

Rohaimi, NA, Ruslan, FA & Adnan, R 2017, 3 Hours ahead of time flood water level prediction using NNARX structure: Case study pahang. in 2016 7th IEEE Control and System Graduate Research Colloquium, ICSGRC 2016 - Proceeding., 7813309, Institute of Electrical and Electronics Engineers Inc., pp. 98-103, 2016 7th IEEE Control and System Graduate Research Colloquium, ICSGRC 2016, Shah Alam, Selangor, Malaysia, 8/8/16. DOI: 10.1109/ICSGRC.2016.7813309
Rohaimi NA, Ruslan FA, Adnan R. 3 Hours ahead of time flood water level prediction using NNARX structure: Case study pahang. In 2016 7th IEEE Control and System Graduate Research Colloquium, ICSGRC 2016 - Proceeding. Institute of Electrical and Electronics Engineers Inc.2017. p. 98-103. 7813309. Available from, DOI: 10.1109/ICSGRC.2016.7813309
Rohaimi, Noor Ashikin ; Ruslan, Fazlina Ahmat ; Adnan, Ramli. / 3 Hours ahead of time flood water level prediction using NNARX structure : Case study pahang. 2016 7th IEEE Control and System Graduate Research Colloquium, ICSGRC 2016 - Proceeding. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 98-103
@inbook{cd3fb53f3e114e3eb619e0f2a033cbe1,
title = "3 Hours ahead of time flood water level prediction using NNARX structure: Case study pahang",
abstract = "Flood is defined as an overflow of large amount of water beyond its normal limits. Therefore, it has become threat to people's life and can cause damages to properties. However, in Malaysia, the only existing flood warning system are the alarming system which only notify residents nearby flood location to evacuate only when flood occur. Thus, flood water level prediction is very much needed in order to prevent flood disaster to happen. One of the effective techniques which frequently used to solve nonlinear cases such as flood is Artificial Neural Network (ANN). Therefore, this paper proposed 3 hours flood water level prediction using Neural Network Autoregressive model with Exogenous Input (NNARX) technique. The area involved in this study was along Pahang river basin where the flood location is situated at Mentakab. Four input parameters were fed in to the NNARX model to predict flood 3 hours ahead of time. The inputs were carefully selected during flood events. The samples used for training, validation and testing stage are 1553, 1997 and 4000 samples respectively. The NNARX flood prediction model developed using Matlab Neural Network Toolbox. Result shows satisfactory performance with low error measures.",
author = "Rohaimi, {Noor Ashikin} and Ruslan, {Fazlina Ahmat} and Ramli Adnan",
year = "2017",
month = "1",
doi = "10.1109/ICSGRC.2016.7813309",
pages = "98--103",
booktitle = "2016 7th IEEE Control and System Graduate Research Colloquium, ICSGRC 2016 - Proceeding",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - CHAP

T1 - 3 Hours ahead of time flood water level prediction using NNARX structure

T2 - Case study pahang

AU - Rohaimi,Noor Ashikin

AU - Ruslan,Fazlina Ahmat

AU - Adnan,Ramli

PY - 2017/1/10

Y1 - 2017/1/10

N2 - Flood is defined as an overflow of large amount of water beyond its normal limits. Therefore, it has become threat to people's life and can cause damages to properties. However, in Malaysia, the only existing flood warning system are the alarming system which only notify residents nearby flood location to evacuate only when flood occur. Thus, flood water level prediction is very much needed in order to prevent flood disaster to happen. One of the effective techniques which frequently used to solve nonlinear cases such as flood is Artificial Neural Network (ANN). Therefore, this paper proposed 3 hours flood water level prediction using Neural Network Autoregressive model with Exogenous Input (NNARX) technique. The area involved in this study was along Pahang river basin where the flood location is situated at Mentakab. Four input parameters were fed in to the NNARX model to predict flood 3 hours ahead of time. The inputs were carefully selected during flood events. The samples used for training, validation and testing stage are 1553, 1997 and 4000 samples respectively. The NNARX flood prediction model developed using Matlab Neural Network Toolbox. Result shows satisfactory performance with low error measures.

AB - Flood is defined as an overflow of large amount of water beyond its normal limits. Therefore, it has become threat to people's life and can cause damages to properties. However, in Malaysia, the only existing flood warning system are the alarming system which only notify residents nearby flood location to evacuate only when flood occur. Thus, flood water level prediction is very much needed in order to prevent flood disaster to happen. One of the effective techniques which frequently used to solve nonlinear cases such as flood is Artificial Neural Network (ANN). Therefore, this paper proposed 3 hours flood water level prediction using Neural Network Autoregressive model with Exogenous Input (NNARX) technique. The area involved in this study was along Pahang river basin where the flood location is situated at Mentakab. Four input parameters were fed in to the NNARX model to predict flood 3 hours ahead of time. The inputs were carefully selected during flood events. The samples used for training, validation and testing stage are 1553, 1997 and 4000 samples respectively. The NNARX flood prediction model developed using Matlab Neural Network Toolbox. Result shows satisfactory performance with low error measures.

UR - http://www.scopus.com/inward/record.url?scp=85011965958&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85011965958&partnerID=8YFLogxK

U2 - 10.1109/ICSGRC.2016.7813309

DO - 10.1109/ICSGRC.2016.7813309

M3 - Conference contribution

SP - 98

EP - 103

BT - 2016 7th IEEE Control and System Graduate Research Colloquium, ICSGRC 2016 - Proceeding

PB - Institute of Electrical and Electronics Engineers Inc.

ER -