2024 : 11 : 14
Rasoul Daneshfaraz

Rasoul Daneshfaraz

Academic rank: Professor
ORCID:
Education: PhD.
ScopusId:
HIndex:
Faculty: 1
Address:
Phone:

Research

Title
Estimation of Scour Depth of Rectangular Bridge Piers Using Artificial Neural Network, Adaptive Neural-Fuzzy Inference System, and Support Vector Machine
Type
Presentation
Keywords
ANN and SVM method, Collision angles, Rectangular pier, Scour depth.
Year
2021
Researchers ، Mehran Seifollahi ، Rasoul Daneshfaraz ، amir ghaderi

Abstract

In the present study, three situations were studied: a rectangular pier without a cable, a rectangular pier with a cable that has a diameter equal to 10% of the pier diameter, and a cable twist angle of 15 degrees, and a rectangular pier with a cable 15% of the pier diameter and an angle of twist of 12 degrees. Tests were carried out with different flow-approach angles: zero, 5, 10, and 15 degrees. In this study, due to the limitation in the collection of experimental data with an angle of incidence of 20 degrees due to the effect of the flume wall on the scouring rate, ANN, ANFIS, and SVM have been used to estimate 20-degree piers. In the present study, zero, 5, and 10-degree collision angles have been used for training and 15° collision angles have been used for prediction as well as comparison with experimental results. In the next step, the depth of scouring with the collision angle of 20 degrees is estimated. The results of the present study show that the ANN and SVM method has very good accuracy in estimating scour depth. However, the ANFIS is not recommended for estimation due to its low accuracy.