Investigation of Meta-heuristics Algorithms in ANN Streamflow Forecasting
The deterministic approach, which utilizes the gradient information in the search process, is prone to trapping at local minima, primarily due to the presence of saddle points and local minima in the non-convex objective function of an artificial neural network (ANN). This study investigated the eff...
Saved in:
Main Authors: | , , , , , |
---|---|
Other Authors: | |
Format: | Article |
Published: |
Korean Society of Civil Engineers
2024
|
Subjects: | |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
my.uniten.dspace-34190 |
---|---|
record_format |
dspace |
spelling |
my.uniten.dspace-341902024-10-14T11:18:21Z Investigation of Meta-heuristics Algorithms in ANN Streamflow Forecasting Wei Y. Hashim H. Chong K.L. Huang Y.F. Ahmed A.N. El-Shafie A. 58169750800 56800153400 57208482172 55807263900 57214837520 16068189400 Machine learning Meta-heuristic algorithms Optimization Statistical tests Time series forecasting Wavelet transform Design of experiments Forecasting Heuristic algorithms Heuristic methods Machine learning Neural networks Statistical tests Statistics Stochastic models Stochastic systems Stream flow Wavelet transforms Deterministic approach Gradient based Local minimums Machine-learning Meta-heuristics algorithms Optimisations Performance Streamflow forecasting Time series forecasting Wavelets transform Optimization The deterministic approach, which utilizes the gradient information in the search process, is prone to trapping at local minima, primarily due to the presence of saddle points and local minima in the non-convex objective function of an artificial neural network (ANN). This study investigated the efficacy of a hybrid model that adopted a meta-heuristic algorithm (MHA) as an optimizer to extend the training ANN method, from a gradient-based to a stochastic population-based approach for streamflow forecasting. In the latter, parameter tuning utilizing the design of experiment (DOE) technique, has become an integral element in the optimization process due to reliance on their parameters. For model convenience, a wavelet transform was employed to decompose the series into sub-series. The empirical studies of MHA performance showed that the hybrid MHA-ANN was superior for streamflow forecasting, especially with the firefly algorithm that had an average RMSE = 96.06, an improvement of approximately 17% over the gradient-based ANN (RMSE = 113.92). However, among the adopted MHAs, not all are compatible with optimizing the ANN for streamflow forecasting, thus requiring a thorough study as performance varies from case to case. Two additional statistical tests, such as the Kruskal-Wallis H test and the Mann-Whitney U test, further validated such disparity in the MHA�s performance. � 2023, Korean Society of Civil Engineers. Final 2024-10-14T03:18:21Z 2024-10-14T03:18:21Z 2023 Article 10.1007/s12205-023-0821-6 2-s2.0-85151518306 https://www.scopus.com/inward/record.uri?eid=2-s2.0-85151518306&doi=10.1007%2fs12205-023-0821-6&partnerID=40&md5=070472e797481fef213fc82d40c161d7 https://irepository.uniten.edu.my/handle/123456789/34190 27 5 2297 2312 Korean Society of Civil Engineers Scopus |
institution |
Universiti Tenaga Nasional |
building |
UNITEN Library |
collection |
Institutional Repository |
continent |
Asia |
country |
Malaysia |
content_provider |
Universiti Tenaga Nasional |
content_source |
UNITEN Institutional Repository |
url_provider |
http://dspace.uniten.edu.my/ |
topic |
Machine learning Meta-heuristic algorithms Optimization Statistical tests Time series forecasting Wavelet transform Design of experiments Forecasting Heuristic algorithms Heuristic methods Machine learning Neural networks Statistical tests Statistics Stochastic models Stochastic systems Stream flow Wavelet transforms Deterministic approach Gradient based Local minimums Machine-learning Meta-heuristics algorithms Optimisations Performance Streamflow forecasting Time series forecasting Wavelets transform Optimization |
spellingShingle |
Machine learning Meta-heuristic algorithms Optimization Statistical tests Time series forecasting Wavelet transform Design of experiments Forecasting Heuristic algorithms Heuristic methods Machine learning Neural networks Statistical tests Statistics Stochastic models Stochastic systems Stream flow Wavelet transforms Deterministic approach Gradient based Local minimums Machine-learning Meta-heuristics algorithms Optimisations Performance Streamflow forecasting Time series forecasting Wavelets transform Optimization Wei Y. Hashim H. Chong K.L. Huang Y.F. Ahmed A.N. El-Shafie A. Investigation of Meta-heuristics Algorithms in ANN Streamflow Forecasting |
description |
The deterministic approach, which utilizes the gradient information in the search process, is prone to trapping at local minima, primarily due to the presence of saddle points and local minima in the non-convex objective function of an artificial neural network (ANN). This study investigated the efficacy of a hybrid model that adopted a meta-heuristic algorithm (MHA) as an optimizer to extend the training ANN method, from a gradient-based to a stochastic population-based approach for streamflow forecasting. In the latter, parameter tuning utilizing the design of experiment (DOE) technique, has become an integral element in the optimization process due to reliance on their parameters. For model convenience, a wavelet transform was employed to decompose the series into sub-series. The empirical studies of MHA performance showed that the hybrid MHA-ANN was superior for streamflow forecasting, especially with the firefly algorithm that had an average RMSE = 96.06, an improvement of approximately 17% over the gradient-based ANN (RMSE = 113.92). However, among the adopted MHAs, not all are compatible with optimizing the ANN for streamflow forecasting, thus requiring a thorough study as performance varies from case to case. Two additional statistical tests, such as the Kruskal-Wallis H test and the Mann-Whitney U test, further validated such disparity in the MHA�s performance. � 2023, Korean Society of Civil Engineers. |
author2 |
58169750800 |
author_facet |
58169750800 Wei Y. Hashim H. Chong K.L. Huang Y.F. Ahmed A.N. El-Shafie A. |
format |
Article |
author |
Wei Y. Hashim H. Chong K.L. Huang Y.F. Ahmed A.N. El-Shafie A. |
author_sort |
Wei Y. |
title |
Investigation of Meta-heuristics Algorithms in ANN Streamflow Forecasting |
title_short |
Investigation of Meta-heuristics Algorithms in ANN Streamflow Forecasting |
title_full |
Investigation of Meta-heuristics Algorithms in ANN Streamflow Forecasting |
title_fullStr |
Investigation of Meta-heuristics Algorithms in ANN Streamflow Forecasting |
title_full_unstemmed |
Investigation of Meta-heuristics Algorithms in ANN Streamflow Forecasting |
title_sort |
investigation of meta-heuristics algorithms in ann streamflow forecasting |
publisher |
Korean Society of Civil Engineers |
publishDate |
2024 |
_version_ |
1814061045051817984 |
score |
13.209306 |