Predicting the geometry of stable alluvial channels: combination of data mining and meta-heuristic optimization

One of the important topics in river engineering is the design of stable alluvial channel geometry in the regime mode (dynamic balance between erosion and sedimentation) including the width, depth and slope. In this research, the ANFIS and ANFIS-PSO models were used to model the geometry parameters...

Full description

Saved in:
Bibliographic Details
Main Authors: Fatemeh Lanjabi Sharahi, Seyed Morteza Seyedian, Mahmood Akbari, Ali Heshmatpour
Format: Article
Language:English
Published: Shahid Chamran University of Ahvaz 2023-10-01
Series:Journal of Hydraulic Structures
Subjects:
Online Access:https://jhs.scu.ac.ir/article_18746_3417863f698f502002a6644a63e41dbf.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850127677321641984
author Fatemeh Lanjabi Sharahi
Seyed Morteza Seyedian
Mahmood Akbari
Ali Heshmatpour
author_facet Fatemeh Lanjabi Sharahi
Seyed Morteza Seyedian
Mahmood Akbari
Ali Heshmatpour
author_sort Fatemeh Lanjabi Sharahi
collection DOAJ
description One of the important topics in river engineering is the design of stable alluvial channel geometry in the regime mode (dynamic balance between erosion and sedimentation) including the width, depth and slope. In this research, the ANFIS and ANFIS-PSO models were used to model the geometry parameters of stable channels. To achieve this objective, we utilized a comprehensive dataset comprising 410 data series sourced from 15 different channels, encompassing various types such as straight and meandering, as well as natural and laboratory. In each measurement, information on the flow rate (Q), average particle diameter (d), shear stress (τ), top width of the channel (W), average depth of flow (h) and longitudinal slope of the channel (S) was collected. Randomly, 70% of the data was used for training, and the remaining 30% was used for validation of the ANFIS and ANFIS-PSO models. Totally, 42 models were derived from the combination of 7 input data sets (Q, d, and τ) and employed both ANFIS and ANFIS-PSO, models to estimate the W, h, and S as the three types of outputs. In modeling of the W and h parameters, the best input was the Q, which the R2, CRM and NRMSE for all data with the ANFIS model were equal to 0.954, -0.029, 0.567 and with ANFIS-PSO model were 0.912, -0.042, and 0.487, respectively. Also, to estimate the S, the modeling results had error. In general, the modeling results with the ANFIS-PSO model were more accurate than the results of the ANFIS model.
format Article
id doaj-art-5dee18813ded4601bba574cf36fef019
institution OA Journals
issn 2345-413X
2345-4156
language English
publishDate 2023-10-01
publisher Shahid Chamran University of Ahvaz
record_format Article
series Journal of Hydraulic Structures
spelling doaj-art-5dee18813ded4601bba574cf36fef0192025-08-20T02:33:36ZengShahid Chamran University of AhvazJournal of Hydraulic Structures2345-413X2345-41562023-10-01938810210.22055/jhs.2023.45445.127618746Predicting the geometry of stable alluvial channels: combination of data mining and meta-heuristic optimizationFatemeh Lanjabi Sharahi0Seyed Morteza Seyedian1Mahmood Akbari2Ali Heshmatpour3Department of Range and Watershed Management, Faculty of Agriculture and Natural Resources, Gonbad Kavous University, Gonbad Kavous, Golestan, Iran.Department of Range and Watershed Management, Faculty of Agriculture and Natural Resources, Gonbad Kavous University, Gonbad Kavous, Golestan, Iran.Department of Water Science and Engineering, Faculty of Agriculture and Environment, Arak University, Arak, Iran.Department of Range and Watershed Management, Faculty of Agriculture and Natural Resources, Gonbad Kavous University, Gonbad Kavous, Golestan, Iran.One of the important topics in river engineering is the design of stable alluvial channel geometry in the regime mode (dynamic balance between erosion and sedimentation) including the width, depth and slope. In this research, the ANFIS and ANFIS-PSO models were used to model the geometry parameters of stable channels. To achieve this objective, we utilized a comprehensive dataset comprising 410 data series sourced from 15 different channels, encompassing various types such as straight and meandering, as well as natural and laboratory. In each measurement, information on the flow rate (Q), average particle diameter (d), shear stress (τ), top width of the channel (W), average depth of flow (h) and longitudinal slope of the channel (S) was collected. Randomly, 70% of the data was used for training, and the remaining 30% was used for validation of the ANFIS and ANFIS-PSO models. Totally, 42 models were derived from the combination of 7 input data sets (Q, d, and τ) and employed both ANFIS and ANFIS-PSO, models to estimate the W, h, and S as the three types of outputs. In modeling of the W and h parameters, the best input was the Q, which the R2, CRM and NRMSE for all data with the ANFIS model were equal to 0.954, -0.029, 0.567 and with ANFIS-PSO model were 0.912, -0.042, and 0.487, respectively. Also, to estimate the S, the modeling results had error. In general, the modeling results with the ANFIS-PSO model were more accurate than the results of the ANFIS model.https://jhs.scu.ac.ir/article_18746_3417863f698f502002a6644a63e41dbf.pdfstable channelgeometryanfisanfis-pso
spellingShingle Fatemeh Lanjabi Sharahi
Seyed Morteza Seyedian
Mahmood Akbari
Ali Heshmatpour
Predicting the geometry of stable alluvial channels: combination of data mining and meta-heuristic optimization
Journal of Hydraulic Structures
stable channel
geometry
anfis
anfis-pso
title Predicting the geometry of stable alluvial channels: combination of data mining and meta-heuristic optimization
title_full Predicting the geometry of stable alluvial channels: combination of data mining and meta-heuristic optimization
title_fullStr Predicting the geometry of stable alluvial channels: combination of data mining and meta-heuristic optimization
title_full_unstemmed Predicting the geometry of stable alluvial channels: combination of data mining and meta-heuristic optimization
title_short Predicting the geometry of stable alluvial channels: combination of data mining and meta-heuristic optimization
title_sort predicting the geometry of stable alluvial channels combination of data mining and meta heuristic optimization
topic stable channel
geometry
anfis
anfis-pso
url https://jhs.scu.ac.ir/article_18746_3417863f698f502002a6644a63e41dbf.pdf
work_keys_str_mv AT fatemehlanjabisharahi predictingthegeometryofstablealluvialchannelscombinationofdataminingandmetaheuristicoptimization
AT seyedmortezaseyedian predictingthegeometryofstablealluvialchannelscombinationofdataminingandmetaheuristicoptimization
AT mahmoodakbari predictingthegeometryofstablealluvialchannelscombinationofdataminingandmetaheuristicoptimization
AT aliheshmatpour predictingthegeometryofstablealluvialchannelscombinationofdataminingandmetaheuristicoptimization