Neural networks for prediction of stream flow based on snow accumulation

Sansiri Tarnpradab, Kishan Mehrotra, Chilukuri K Mohan, David G Chandler

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

This study aims to improve stream-ow forecast at Reynolds Mountain East watersheds, which is located at the southernmost of all watersheds in Reynolds Creek Experimental Watershed Idaho, USA. Two separate models, one for the annual data and the other for the seasonal (April-June) data from 1983-1995 are tested for their predictability. Due to the difculties in collecting data during winter months, in particular the snow water equivalent (SWE), this study evaluates the impact of excluding this variable. Our results show that multilayer perceptrons (MLP) and support vector machines (SVM) are more suitable for modeling the data. The results also reveal that the difference between stream-ow forecast via annual and seasonal models is insignicant and for longer term predictions SWE is a strong driver in the stream-ow forecast. Principal Component Analysis (PCA) and Particle Swarm Optimization (PSO) are also used in this study to identify useful features. The results from PCA derived models show that PCA helps reduce prediction error and the results are more stable than using models without PCA. PSO also improved results; however, the set of selected attributes by PSO is less believable than given by PCA. The best prediction is achieved when MLP model is implemented with attributes generated by PCA.

Original languageEnglish (US)
Title of host publicationIEEE SSCI 2014 - 2014 IEEE Symposium Series on Computational Intelligence - CIES 2014: 2014 IEEE Symposium on Computational Intelligence for Engineering Solutions, Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages88-94
Number of pages7
ISBN (Print)9781479945108
DOIs
StatePublished - Jan 15 2015
Event2014 IEEE Symposium on Computational Intelligence for Engineering Solutions, CIES 2014 - Orlando, United States
Duration: Dec 9 2014Dec 12 2014

Other

Other2014 IEEE Symposium on Computational Intelligence for Engineering Solutions, CIES 2014
CountryUnited States
CityOrlando
Period12/9/1412/12/14

Fingerprint

Stream flow
Snow
Principal component analysis
Neural networks
Watersheds
Particle swarm optimization (PSO)
Multilayer neural networks
Support vector machines
Water

ASJC Scopus subject areas

  • Computer Science Applications
  • Artificial Intelligence
  • Computational Theory and Mathematics

Cite this

Tarnpradab, S., Mehrotra, K., Mohan, C. K., & Chandler, D. G. (2015). Neural networks for prediction of stream flow based on snow accumulation. In IEEE SSCI 2014 - 2014 IEEE Symposium Series on Computational Intelligence - CIES 2014: 2014 IEEE Symposium on Computational Intelligence for Engineering Solutions, Proceedings (pp. 88-94). [7011836] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CIES.2014.7011836

Neural networks for prediction of stream flow based on snow accumulation. / Tarnpradab, Sansiri; Mehrotra, Kishan; Mohan, Chilukuri K; Chandler, David G.

IEEE SSCI 2014 - 2014 IEEE Symposium Series on Computational Intelligence - CIES 2014: 2014 IEEE Symposium on Computational Intelligence for Engineering Solutions, Proceedings. Institute of Electrical and Electronics Engineers Inc., 2015. p. 88-94 7011836.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Tarnpradab, S, Mehrotra, K, Mohan, CK & Chandler, DG 2015, Neural networks for prediction of stream flow based on snow accumulation. in IEEE SSCI 2014 - 2014 IEEE Symposium Series on Computational Intelligence - CIES 2014: 2014 IEEE Symposium on Computational Intelligence for Engineering Solutions, Proceedings., 7011836, Institute of Electrical and Electronics Engineers Inc., pp. 88-94, 2014 IEEE Symposium on Computational Intelligence for Engineering Solutions, CIES 2014, Orlando, United States, 12/9/14. https://doi.org/10.1109/CIES.2014.7011836
Tarnpradab S, Mehrotra K, Mohan CK, Chandler DG. Neural networks for prediction of stream flow based on snow accumulation. In IEEE SSCI 2014 - 2014 IEEE Symposium Series on Computational Intelligence - CIES 2014: 2014 IEEE Symposium on Computational Intelligence for Engineering Solutions, Proceedings. Institute of Electrical and Electronics Engineers Inc. 2015. p. 88-94. 7011836 https://doi.org/10.1109/CIES.2014.7011836
Tarnpradab, Sansiri ; Mehrotra, Kishan ; Mohan, Chilukuri K ; Chandler, David G. / Neural networks for prediction of stream flow based on snow accumulation. IEEE SSCI 2014 - 2014 IEEE Symposium Series on Computational Intelligence - CIES 2014: 2014 IEEE Symposium on Computational Intelligence for Engineering Solutions, Proceedings. Institute of Electrical and Electronics Engineers Inc., 2015. pp. 88-94
@inproceedings{d7bf3e1074a6422db50e167a0eb92f42,
title = "Neural networks for prediction of stream flow based on snow accumulation",
abstract = "This study aims to improve stream-ow forecast at Reynolds Mountain East watersheds, which is located at the southernmost of all watersheds in Reynolds Creek Experimental Watershed Idaho, USA. Two separate models, one for the annual data and the other for the seasonal (April-June) data from 1983-1995 are tested for their predictability. Due to the difculties in collecting data during winter months, in particular the snow water equivalent (SWE), this study evaluates the impact of excluding this variable. Our results show that multilayer perceptrons (MLP) and support vector machines (SVM) are more suitable for modeling the data. The results also reveal that the difference between stream-ow forecast via annual and seasonal models is insignicant and for longer term predictions SWE is a strong driver in the stream-ow forecast. Principal Component Analysis (PCA) and Particle Swarm Optimization (PSO) are also used in this study to identify useful features. The results from PCA derived models show that PCA helps reduce prediction error and the results are more stable than using models without PCA. PSO also improved results; however, the set of selected attributes by PSO is less believable than given by PCA. The best prediction is achieved when MLP model is implemented with attributes generated by PCA.",
author = "Sansiri Tarnpradab and Kishan Mehrotra and Mohan, {Chilukuri K} and Chandler, {David G}",
year = "2015",
month = "1",
day = "15",
doi = "10.1109/CIES.2014.7011836",
language = "English (US)",
isbn = "9781479945108",
pages = "88--94",
booktitle = "IEEE SSCI 2014 - 2014 IEEE Symposium Series on Computational Intelligence - CIES 2014: 2014 IEEE Symposium on Computational Intelligence for Engineering Solutions, Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Neural networks for prediction of stream flow based on snow accumulation

AU - Tarnpradab, Sansiri

AU - Mehrotra, Kishan

AU - Mohan, Chilukuri K

AU - Chandler, David G

PY - 2015/1/15

Y1 - 2015/1/15

N2 - This study aims to improve stream-ow forecast at Reynolds Mountain East watersheds, which is located at the southernmost of all watersheds in Reynolds Creek Experimental Watershed Idaho, USA. Two separate models, one for the annual data and the other for the seasonal (April-June) data from 1983-1995 are tested for their predictability. Due to the difculties in collecting data during winter months, in particular the snow water equivalent (SWE), this study evaluates the impact of excluding this variable. Our results show that multilayer perceptrons (MLP) and support vector machines (SVM) are more suitable for modeling the data. The results also reveal that the difference between stream-ow forecast via annual and seasonal models is insignicant and for longer term predictions SWE is a strong driver in the stream-ow forecast. Principal Component Analysis (PCA) and Particle Swarm Optimization (PSO) are also used in this study to identify useful features. The results from PCA derived models show that PCA helps reduce prediction error and the results are more stable than using models without PCA. PSO also improved results; however, the set of selected attributes by PSO is less believable than given by PCA. The best prediction is achieved when MLP model is implemented with attributes generated by PCA.

AB - This study aims to improve stream-ow forecast at Reynolds Mountain East watersheds, which is located at the southernmost of all watersheds in Reynolds Creek Experimental Watershed Idaho, USA. Two separate models, one for the annual data and the other for the seasonal (April-June) data from 1983-1995 are tested for their predictability. Due to the difculties in collecting data during winter months, in particular the snow water equivalent (SWE), this study evaluates the impact of excluding this variable. Our results show that multilayer perceptrons (MLP) and support vector machines (SVM) are more suitable for modeling the data. The results also reveal that the difference between stream-ow forecast via annual and seasonal models is insignicant and for longer term predictions SWE is a strong driver in the stream-ow forecast. Principal Component Analysis (PCA) and Particle Swarm Optimization (PSO) are also used in this study to identify useful features. The results from PCA derived models show that PCA helps reduce prediction error and the results are more stable than using models without PCA. PSO also improved results; however, the set of selected attributes by PSO is less believable than given by PCA. The best prediction is achieved when MLP model is implemented with attributes generated by PCA.

UR - http://www.scopus.com/inward/record.url?scp=84922824338&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84922824338&partnerID=8YFLogxK

U2 - 10.1109/CIES.2014.7011836

DO - 10.1109/CIES.2014.7011836

M3 - Conference contribution

AN - SCOPUS:84922824338

SN - 9781479945108

SP - 88

EP - 94

BT - IEEE SSCI 2014 - 2014 IEEE Symposium Series on Computational Intelligence - CIES 2014: 2014 IEEE Symposium on Computational Intelligence for Engineering Solutions, Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -