DETECTION OF BREAKWATER FAILURE MODES FROM PHOTOGRAMMETRIC DERIVED POINT CLOUDS – A CASE STUDY FOR BUNBURY PORT, WESTERN AUSTRALIA
Abstract
The concept of a digital twin is increasingly applied to a wide range of physical assets as it offers spatial contextualisation, temporal contextualisation, simulation, and optimisation. For the Southern Ports Authority (SPA) of Western Australia one important asset class is breakwater walls. SPA has implemented a spatial contextualisation for their breakwater facilities using a Geographic Information System (GIS) including the breakwater wall condition. This proof-of-concept study aims to validate that timely mapping of breakwater conditions using Unmanned Aerial Vehicles (UAV) and the information derived from the images captured by UAVs is possible. Three failure modes have been investigated using single and multiple epoch data. The failure modes are slope defects, breach or loss of crest elevation and armour movement or loss of armour interlocking. The highest potential for an automatic assessment was concluded for the Breach or loss of crest elevation failure mode. Armour movement or loss is likely to be detected as well, as long as the point spacing of the point clouds in different epochs is comparable. Further investigations are required for the loss of armour interlocking as well as for multi-epoch slope defect assessment. Hence, while it is a proof-of-concept study only, it is the first step to develop a more automated assessment of breakwater walls. Consequently, this data can then be used for simulation and optimisation and the integration of the data in business processes, i.e., the maintenance cycle for the breakwater walls.

Authors
P. Helmholz1, L. Boyle2, B.E. Tily-Laurie3, D. May4, G. Singh5, J. Hogben6, Y. Ren7, Q. Li8, D. Belton9, S. Khaksar10, S.J. Snyman11
Curtin University, Australia1,2,3,4,5,7,8,9,10, Southern Ports Authority, Australia6,11

Keywords
Breakwater, Failure Modes, UAV, Point Cloud, Point-To-Point Comparison, Point-To-Model Comparison
Yearly Full Views
JanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecember
000000010000
Published By :
ICTACT
Published In :
ICTACT Journal on Soft Computing
( Volume: 13 , Issue: 3 , Pages: 2948 - 2959 )
Date of Publication :
April 2023
Page Views :
191
Full Text Views :
2

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.