Efficient data-driven machine learning models for scour depth predictions at sloping sea defences
Seawalls are critical defence infrastructures in coastal zones that protect hinterland areas from storm surges, wave overtopping and soil erosion hazards. Scouring at the toe of sea defences, caused by wave-induced accretion and erosion of bed material imposes a significant threat to the structural...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2024-02-01
|
| Series: | Frontiers in Built Environment |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fbuil.2024.1343398/full |
| _version_ | 1797319467569512448 |
|---|---|
| author | M. A. Habib S. Abolfathi John. J. O’Sullivan M. Salauddin |
| author_facet | M. A. Habib S. Abolfathi John. J. O’Sullivan M. Salauddin |
| author_sort | M. A. Habib |
| collection | DOAJ |
| description | Seawalls are critical defence infrastructures in coastal zones that protect hinterland areas from storm surges, wave overtopping and soil erosion hazards. Scouring at the toe of sea defences, caused by wave-induced accretion and erosion of bed material imposes a significant threat to the structural integrity of coastal infrastructures. Accurate prediction of scour depths is essential for appropriate and efficient design and maintenance of coastal structures, which serve to mitigate risks of structural failure through toe scouring. However, limited guidance and predictive tools are available for estimating toe scouring at sloping structures. In recent years, Artificial Intelligence and Machine Learning (ML) algorithms have gained interest, and although they underpin robust predictive models for many coastal engineering applications, such models have yet to be applied to scour prediction. Here we develop and present ML-based models for predicting toe scour depths at sloping seawall. Four ML algorithms, namely, Random Forest (RF), Gradient Boosted Decision Trees (GBDT), Artificial Neural Networks (ANNs), and Support Vector Machine Regression (SVMR) are utilised. Comprehensive physical modelling measurement data is utilised to develop and validate the predictive models. A Novel framework for feature selection, feature importance, and hyperparameter tuning algorithms are adopted for pre- and post-processing steps of ML-based models. In-depth statistical analyses are proposed to evaluate the predictive performance of the proposed models. The results indicate a minimum of 80% prediction accuracy across all the algorithms tested in this study and overall, the SVMR produced the most accurate predictions with a Coefficient of Determination (r2) of 0.74 and a Mean Absolute Error (MAE) value of 0.17. The SVMR algorithm also offered most computationally efficient performance among the algorithms tested. The methodological framework proposed in this study can be applied to scouring datasets for rapid assessment of scour at coastal defence structures, facilitating model-informed decision-making. |
| first_indexed | 2024-03-08T04:07:18Z |
| format | Article |
| id | doaj.art-21e941b8993c4acfa7e44c9276f150dd |
| institution | Directory Open Access Journal |
| issn | 2297-3362 |
| language | English |
| last_indexed | 2024-03-08T04:07:18Z |
| publishDate | 2024-02-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Built Environment |
| spelling | doaj.art-21e941b8993c4acfa7e44c9276f150dd2024-02-09T04:40:32ZengFrontiers Media S.A.Frontiers in Built Environment2297-33622024-02-011010.3389/fbuil.2024.13433981343398Efficient data-driven machine learning models for scour depth predictions at sloping sea defencesM. A. Habib0S. Abolfathi1John. J. O’Sullivan2M. Salauddin3UCD School of Civil Engineering, UCD Dooge Centre for Water Resources Research and UCD Earth Institute, University College Dublin, Dublin, IrelandSchool of Engineering, University of Warwick, Coventry, United KingdomUCD School of Civil Engineering, UCD Dooge Centre for Water Resources Research and UCD Earth Institute, University College Dublin, Dublin, IrelandUCD School of Civil Engineering, UCD Dooge Centre for Water Resources Research and UCD Earth Institute, University College Dublin, Dublin, IrelandSeawalls are critical defence infrastructures in coastal zones that protect hinterland areas from storm surges, wave overtopping and soil erosion hazards. Scouring at the toe of sea defences, caused by wave-induced accretion and erosion of bed material imposes a significant threat to the structural integrity of coastal infrastructures. Accurate prediction of scour depths is essential for appropriate and efficient design and maintenance of coastal structures, which serve to mitigate risks of structural failure through toe scouring. However, limited guidance and predictive tools are available for estimating toe scouring at sloping structures. In recent years, Artificial Intelligence and Machine Learning (ML) algorithms have gained interest, and although they underpin robust predictive models for many coastal engineering applications, such models have yet to be applied to scour prediction. Here we develop and present ML-based models for predicting toe scour depths at sloping seawall. Four ML algorithms, namely, Random Forest (RF), Gradient Boosted Decision Trees (GBDT), Artificial Neural Networks (ANNs), and Support Vector Machine Regression (SVMR) are utilised. Comprehensive physical modelling measurement data is utilised to develop and validate the predictive models. A Novel framework for feature selection, feature importance, and hyperparameter tuning algorithms are adopted for pre- and post-processing steps of ML-based models. In-depth statistical analyses are proposed to evaluate the predictive performance of the proposed models. The results indicate a minimum of 80% prediction accuracy across all the algorithms tested in this study and overall, the SVMR produced the most accurate predictions with a Coefficient of Determination (r2) of 0.74 and a Mean Absolute Error (MAE) value of 0.17. The SVMR algorithm also offered most computationally efficient performance among the algorithms tested. The methodological framework proposed in this study can be applied to scouring datasets for rapid assessment of scour at coastal defence structures, facilitating model-informed decision-making.https://www.frontiersin.org/articles/10.3389/fbuil.2024.1343398/fullrandom forestgradient boosted decision treesSupport Vector Machine Regressionmarine and coastal managementcoastal hazards mitigationtoe scouring |
| spellingShingle | M. A. Habib S. Abolfathi John. J. O’Sullivan M. Salauddin Efficient data-driven machine learning models for scour depth predictions at sloping sea defences Frontiers in Built Environment random forest gradient boosted decision trees Support Vector Machine Regression marine and coastal management coastal hazards mitigation toe scouring |
| title | Efficient data-driven machine learning models for scour depth predictions at sloping sea defences |
| title_full | Efficient data-driven machine learning models for scour depth predictions at sloping sea defences |
| title_fullStr | Efficient data-driven machine learning models for scour depth predictions at sloping sea defences |
| title_full_unstemmed | Efficient data-driven machine learning models for scour depth predictions at sloping sea defences |
| title_short | Efficient data-driven machine learning models for scour depth predictions at sloping sea defences |
| title_sort | efficient data driven machine learning models for scour depth predictions at sloping sea defences |
| topic | random forest gradient boosted decision trees Support Vector Machine Regression marine and coastal management coastal hazards mitigation toe scouring |
| url | https://www.frontiersin.org/articles/10.3389/fbuil.2024.1343398/full |
| work_keys_str_mv | AT mahabib efficientdatadrivenmachinelearningmodelsforscourdepthpredictionsatslopingseadefences AT sabolfathi efficientdatadrivenmachinelearningmodelsforscourdepthpredictionsatslopingseadefences AT johnjosullivan efficientdatadrivenmachinelearningmodelsforscourdepthpredictionsatslopingseadefences AT msalauddin efficientdatadrivenmachinelearningmodelsforscourdepthpredictionsatslopingseadefences |