IAHR, founded in 1935, is a worldwide independent member-based organisation of engineers and water specialists working in fields related to the hydro-environmental sciences and their practical application. Activities range from river and maritime hydraulics to water resources development and eco-hydraulics, through to ice engineering, hydroinformatics, and hydraulic machinery.
Log On
About IAHRDirectoryCommitteesMy IAHRNews & JournalseLibraryeShopEventsJoin IAHRWorld CongressDonate
spacer.gif
spacer.gif eLibrary
spacer.gif eLibrary
You are here : eLibrary : IAHR World Congress Proceedings : 36th Congress - The Hague (2015) ALL CONTENT : Flood risk management and adaptation : A new system for bridge scour monitoring and prediction
A new system for bridge scour monitoring and prediction
Author : MANOUSOS VALYRAKIS(1), PANAYIOTIS MICHALIS(2) & HANQING ZHANG(1)
ABSTRACT
Earth's surface is continuously shaped due to the action of geophysical flows. Erosion due to the flow of water in river
systems has been identified as a key problem in preserving ecological health but also a threat to our built environment and
critical infrastructure, worldwide. As an example, it has been estimated that a major reason for bridge failure is due to
scour. Even though the flow past bridge piers has been investigated both experimentally and numerically, and the
mechanisms of scouring are relatively understood, there still lacks a tool that can offer fast and reliable predictions. Most
of the existing formulas for prediction of bridge pier scour depth are empirical in nature, based on a limited range of data or
for piers of specific shape. In this work, the use of a novel methodology is proposed for the prediction of bridge scour.
Specifically, the use of an Adaptive Neuro-Fuzzy Inference System (ANFIS) is proposed to estimate the scour depth
around bridge piers. In particular, various complexity architectures are sequentially built, in order to identify the optimal for
scour depth predictions, using appropriate training and validation subsets obtained from the USGS database (and preprocessed
to remove incomplete records). The model has five variables, namely the effective pier width (b), the approach
velocity (v), the approach depth (y), the mean grain diameter (D50) and the skew to flow. Simulations are conducted with
data groups (bed material type, pier type and shape) and different number of input variables, to produce reduced
complexity and easily interpretable models. Analysis and comparison of the results indicate that the developed ANFIS
model has high accuracy and outstanding generalization ability for prediction of scour parameters. The effective pier width
(as opposed to skew to flow) is amongst the most relevant input parameters for the estimation. Training of the system to
new bridge geometries and flow conditions can be achieved by obtaining real time data, via novel electromagnetic sensors
monitoring scour depth. Once the model is trained with data representative of the new system, bridge scour prediction can
be performed for high/design flows or floods.
File Size : 481,476 bytes
File Type : Adobe Acrobat Document
Chapter : IAHR World Congress Proceedings
Category : 36th Congress - The Hague (2015) ALL CONTENT
Article : Flood risk management and adaptation
Date Published : 18/08/2015
Download Now