1- Shahrood University of Technology
2- Shahrood University of Technology ,adjami@shahroodut.ac.ir
2- Shahrood University of Technology ,
Abstract: (367 Views)
Coastal areas are among the most densely populated and developed regions globally. This is owing to high economic opportunities, maritime trade access, and a thriving tourism industry. These regions represent vulnerable environments facing threats such as storms and rising sea levels, predominantly stemming from human activities. Erosion stands out as one of the most pressing threats to these regions. It is crucial to continuously monitor these areas, record changes in beach profiles and shoreline alterations, and control erosion. There are various databases worldwide are actively recording Morphodynamic and hydrodynamic data and monitoring shorelines. Locations such as Narrabeen, San Diego, and others are subsets of these databases. This research looks into changes in the shoreline and coastal berm under non-storm conditions, employing machine learning algorithms to evaluate these phenomena in the Narrabeen coastal region of Australia. The Narrabeen-Collaroy embayment, spanning 3.6 kilometers, is located on the northern shores of Sydney. In areas like Narrabeen-Collaroy, energy gradients vary along the coast due to curvature. The northern region has Moderate Dissipative Energy conditions, but moving towards the south, the conditions change to moderate reflective energy because of decreased energy. Narrabeen-Collaroy is among the most crucial databases in coastal engineering. Field data from the nearshore and coastal strip have been collected in this database from 1976 to 2019. Therefore, data related to storm profiles have not been considered (a condition that requires at least one profile to remain in each month). Based on this criterion, out of the 960 recorded profiles, 73 profiles were identified in severe storm conditions. Eventually, 887 profiles from 2006 to 2019 have been scrutinized. Initially, variations in the shoreline were investigated, focusing on the continuous area consistently interacting with waves. This region can play a significant role in evaluating the performance of models and algorithms. Following that, the geometric changes of the coastal berm, situated within the Shoreface zone, were investigated and analyzed. In this research, wrapper backward future selection algorithm has been used to filter the effective parameters consist of hydrodynamics and Morphodynamic. Also, the objective functions include shoreline changes, variations in berm crest elevation, and the horizontal position of the berm crest. In this step the parameters that has been identified from wrapper backward future selection algorithm, a feed-forward neural network algorithm was employed to predict the objective functions in the final stage. The values obtained from the neural network model for each of the three objective functions demonstrated an appropriate arrangement of parameters. When predicting shoreline changes in scenario DS4, the inclusion of the wave-breaking index parameter led to more logical and acceptable outcomes compared to scenario DTS3 and the resulting R2 is 92% with an RMSE of 3.03 meters. Predicting variations of the berm crest elevation in scenario DY4 illustrates acceptable results: R2 of 75% and RMSE of 0.35 meters. Furthermore, predicting the horizontal position of the berm crest in scenario DX7 shows that the wave-breaking index parameter, improved results compared to scenario DX6 and the final results indicate an R2 of 85.80% and an RMSE of 9.28 meters. To validate the obtained results, the differences in error between the objective function values and output data indicate that the selected scenarios predict data with minimal error. According to the results, it is crucial to accurately identify the relevant hydrodynamic and Morphodynamic parameters and correctly extracted them to achieve more precise predictions. Moreover, the results indicate that the neural network algorithm can accurately predict changes in beach morphology.
Article Type: Original Research |
Subject:
Marine Structures
Received: 2024/04/30 | Accepted: 2024/07/10
Received: 2024/04/30 | Accepted: 2024/07/10
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |