Water level prediction using artificial neural network with particle swarm optimization model
© 2017 IEEE. Flash flood is a natural disaster that causes great losses. It happens mostly in rural areas when heavy rainfall is gathered into the main river in watershed areas. Lots of water comes into the river. This causes a great volume of water flows down to the downstream river area. The water...
Saved in:
Main Authors: | , , |
---|---|
Format: | Conference Proceeding |
Published: |
2018
|
Subjects: | |
Online Access: | https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85039942646&origin=inward http://cmuir.cmu.ac.th/jspui/handle/6653943832/57052 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Chiang Mai University |
id |
th-cmuir.6653943832-57052 |
---|---|
record_format |
dspace |
spelling |
th-cmuir.6653943832-570522018-09-05T03:34:22Z Water level prediction using artificial neural network with particle swarm optimization model Pornnapa Panyadee Paskorn Champrasert Chuchoke Aryupong Computer Science © 2017 IEEE. Flash flood is a natural disaster that causes great losses. It happens mostly in rural areas when heavy rainfall is gathered into the main river in watershed areas. Lots of water comes into the river. This causes a great volume of water flows down to the downstream river area. The water level at the downstream river should be predicted to issue the warning messages to the villagers in the floodplains before the flood arrival. Thus, a flash flood early warning system is a solution to reduce damage from flash floods. Although the artificial neural network (ANN) can be applied as the prediction model, the accuracy of the prediction results depends on the parameter values (e.g., the number of previous data, the period of previous data). This paper proposes to apply the particle swarm optimization technique to tune up the parameter values in the ANN. The proposed model, called W-POpt model, consists of two components, which are 1) PSO is applied as optimizer to search for the optimal parameter values for the ANN training process, and 2) ANN is applied to find the predicted water level. The evaluation results show that PSO yields the optimal parameter values. Applying PSO can reduce the training process time in ANN. The predicted water level from the W-POpt model is acceptable for applying in flash flood early warning systems. 2018-09-05T03:34:22Z 2018-09-05T03:34:22Z 2017-10-18 Conference Proceeding 2-s2.0-85039942646 10.1109/ICoICT.2017.8074670 https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85039942646&origin=inward http://cmuir.cmu.ac.th/jspui/handle/6653943832/57052 |
institution |
Chiang Mai University |
building |
Chiang Mai University Library |
country |
Thailand |
collection |
CMU Intellectual Repository |
topic |
Computer Science |
spellingShingle |
Computer Science Pornnapa Panyadee Paskorn Champrasert Chuchoke Aryupong Water level prediction using artificial neural network with particle swarm optimization model |
description |
© 2017 IEEE. Flash flood is a natural disaster that causes great losses. It happens mostly in rural areas when heavy rainfall is gathered into the main river in watershed areas. Lots of water comes into the river. This causes a great volume of water flows down to the downstream river area. The water level at the downstream river should be predicted to issue the warning messages to the villagers in the floodplains before the flood arrival. Thus, a flash flood early warning system is a solution to reduce damage from flash floods. Although the artificial neural network (ANN) can be applied as the prediction model, the accuracy of the prediction results depends on the parameter values (e.g., the number of previous data, the period of previous data). This paper proposes to apply the particle swarm optimization technique to tune up the parameter values in the ANN. The proposed model, called W-POpt model, consists of two components, which are 1) PSO is applied as optimizer to search for the optimal parameter values for the ANN training process, and 2) ANN is applied to find the predicted water level. The evaluation results show that PSO yields the optimal parameter values. Applying PSO can reduce the training process time in ANN. The predicted water level from the W-POpt model is acceptable for applying in flash flood early warning systems. |
format |
Conference Proceeding |
author |
Pornnapa Panyadee Paskorn Champrasert Chuchoke Aryupong |
author_facet |
Pornnapa Panyadee Paskorn Champrasert Chuchoke Aryupong |
author_sort |
Pornnapa Panyadee |
title |
Water level prediction using artificial neural network with particle swarm optimization model |
title_short |
Water level prediction using artificial neural network with particle swarm optimization model |
title_full |
Water level prediction using artificial neural network with particle swarm optimization model |
title_fullStr |
Water level prediction using artificial neural network with particle swarm optimization model |
title_full_unstemmed |
Water level prediction using artificial neural network with particle swarm optimization model |
title_sort |
water level prediction using artificial neural network with particle swarm optimization model |
publishDate |
2018 |
url |
https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85039942646&origin=inward http://cmuir.cmu.ac.th/jspui/handle/6653943832/57052 |
_version_ |
1681424806865534976 |