Prediction of Wave Transmission Characteristics of Low-Crested Structures with Comprehensive Analysis of Machine Learning
Abstract
:1. Introduction
2. Methodology
2.1. Machine Learning Model
2.1.1. Linear Regression Model
2.1.2. Lasso Regression
2.1.3. Ridge Regression
2.1.4. SVM
2.1.5. Gaussian Process Regression (GPR)
2.1.6. Ensemble Method
- (1)
- Random Forest (RF)
- (2)
- Boosting method
2.2. Analysis of Machine Learning Model
2.2.1. Performance Measurement
2.2.2. Analysis Method of Feature Importance
- (1)
- eXplainable Artificial Intelligence (XAI)
- (2)
- Shapley Additive exPlanations (SHAP)
2.3. Empirical Formula of Wave Transmission Coefficient
2.4. Model Design Condition and Method
2.4.1. Machine Learning Automatic Pipeline Model
2.4.2. Machine Learning Model Configuration and Input Conditions
- 81 data on rubble mound emerged/submerged breakwater [Seelig];
- 95 data on tetrapod submerged breakwater [Daemrich and Kahle];
- 31 data on rubble mound emerged/submerged breakwater [Van der Meer];
- 53 data on rubble mound emerged/submerged breakwater [Daemen].
3. Results and Discussion
3.1. Comparison of Machine Learning Model and Model Selection
3.2. Model Performance Analysis
3.2.1. Results of Splitting a Dataset
3.2.2. 10-Fold Validation Analysis
3.2.3. Evaluation of GBR Model Using Another Data Set
- 20 data on rubble mound LCS (submerged) [Delft Hydraulics].
- 21 data on rubble mound LCS (emerged) [Allsop].
3.2.4. Feature Importance Analysis
3.2.5. Influence of Input Variable Number
3.3. Comparison of Wave Transmission Coefficient Using Empirical Formulas and Machine Learning Models
4. Conclusions
Author Contributions
Funding
Conflicts of Interest
Appendix A
Parameter | Definition | Average | Max | Min |
---|---|---|---|---|
H0 (m) | Wave height | 0.123 | 0.231 | 0.021 |
T0 (s) | Wave period | 2.017 | 3.660 | 0.910 |
RC (m) | Crest freeboard | −1.081 | 0.196 | −0.420 |
B (m) | Crest width | 0.429 | 1.000 | 0.200 |
Dn50 (m) | Nominal diameter | 0.110 | 0.161 | 0.028 |
Slope of structure | 0.507 | 0.667 | 0.250 |
References
- Mimura, N.; Kawaguchi, E. Responses of Coastal Topography to Sea-Level Rise. In Proceedings of the 25th International Conference on Coastal Engineering, Orlando, FL, USA; 1997; pp. 1349–1360. [Google Scholar]
- Moghaddam, E.I.; Hakimzadeh, H.; Allahdadi, M.N.; Hamedi, A.; Nasrollahi, A. Wave-induced currents in the northern gulf of Oman: A numerical study for Ramin prot along the Iranian Coast. Am. J. Fluid Dyn. 2018, 8, 30–39. [Google Scholar]
- Hseih, T.C.; Ding, Y.; Yeh, K.C.; Jhong, R.K. Investigation of Morphological Changes in the Tamsui River Estuary Using an Integrated Coastal and dEstuarine Processes Model. Water 2020, 12, 1084. [Google Scholar] [CrossRef]
- Pilarczyk, K.W. Design of low-crested submerged structures—An overview. In Proceedings of the 6th International Conference on Coastal and Port Engineering in Developing Countries, Pianc-Copedec, Colombo, Sri-Lanka, 15–19 September 2003; pp. 1–16. [Google Scholar]
- Seabrook, S.R.; Hall, K.R. Wave transmission at submerged rubble mound breakwaters. In Proceedings of the 26th International Conference on Coastal Engineering, Copenhagen, Denmark, 22–26 June 1998; pp. 2000–2013. [Google Scholar]
- Tanaka, N. Wave deformation and beach stabilization capacity of wide-crested submerged breakwaters. In Proceedings of the 23rd National Conference on Coastal Engineering, Fukuoka, Japan, 25–26 November 1976; pp. 152–157. (In Japanese). [Google Scholar]
- Buccino, M.; Calabrese, M. Conceptual approach for prediction of wave transmission at Low-Crested Breakwaters. J. Waterw. Port Coast. Ocean. Eng. 2007, 133, 213–224. [Google Scholar] [CrossRef]
- Ahmadian, A.S.; Simons, R.R. Estimaion of nearshore wave transmission for submerged breakwaters using a data-driven predictive model. Neural Comput. Appl. 2016, 29, 705–719. [Google Scholar] [CrossRef] [Green Version]
- Losada, I.J.; Losada, M.A.; Martin, F.L. Harmonic generation past a submerged porous step. Coast. Eng. 1997, 31, 281–304. [Google Scholar] [CrossRef]
- Martinelli, L.; Zanuttigh, B.; Lamberti, A. Hydrodynamic and morphodynamic response of isolated and multiple low crested structures: Experiments and simulations. Coast. Eng. 2006, 53, 363–379. [Google Scholar] [CrossRef]
- Ai, C.; Jin, S.; Lv, B. A new fully non-hydrostatic 3D free surface flow model for water wave motions. Int. J. Numer. Meth. Fluids 2011, 66, 1354–1370. [Google Scholar] [CrossRef]
- Ning, Y.; Liu, W.; Sun, Z.; Zhao, X.; Zhang, Y. Parametric study of solitary wave propagation and runup over fringing reefs based on a Boussinesq wave model. J. Mar. Sci. Technol. 2018, 24, 512–525. [Google Scholar] [CrossRef]
- Hur, D.S.; Lee, W.D.; Cho, W.C. Three-dimensional flow characteristics around permeable submerged breakwaters with open inlet. Ocean Eng. 2012, 44, 100–116. [Google Scholar] [CrossRef]
- Higuera, P.; Lara, J.L.; Losada, I.J. Realistic wave generation and active wave absorption for Navier-Stokes models: Application to OpenFOAM. Coast. Eng. 2013, 71, 102–118. [Google Scholar] [CrossRef]
- Mitchell, T.M. Machine Learning; McGraw-Hill, Inc.: New York, NY, USA, 1997. [Google Scholar]
- Salehi, H.; Burgueño, R. Emerging artificial intelligence methods in structural engineering. Eng. Struct. 2018, 171, 170–189. [Google Scholar] [CrossRef]
- Kundapura, S.; Arkal, V.H.; Pinho, J.L. Below the data range prediction of soft computing wave reflection of semicircular breakwater. J. Mar. Sci. Appl. 2019, 18, 167–175. [Google Scholar] [CrossRef]
- Kuntoji, G.; Manu, R.; Subba, R. Prediction of wave transmission over submerged reef of tandem breakwater using PSO-SVM and PSO-ANN techniques. ISH J. Hydraul. Eng. 2018, 26, 283–290. [Google Scholar] [CrossRef]
- Kim, D.H.; Park, W.S. Neural network for design and reliability analysis of rubble mound breakwaters. Ocean Eng. 2005, 32, 1332–1349. [Google Scholar] [CrossRef]
- Van der Meer, J.W. Rock Slopes and Gravel Beaches under Wave Attack. Ph.D. Thesis, Delft University of Technology, Delft, The Netherlands, 1990. [Google Scholar]
- Koc, M.L.; Balas, C.E.; Koc, D.I. Stability assessment of rubble-mound breakwaters using genetic programming. Ocean Eng. 2016, 111, 8–12. [Google Scholar] [CrossRef]
- Boser, B.E.; Guyon, I.M.; Vapnik, V.N. A Training Algorithm for Optimal Margin Classifiers. In 5th Annual ACM Workshop on COLT; Haussler, D., Ed.; ACM Press: Pittsburg, PA, USA, 1992. [Google Scholar]
- Steve, R.G. Support Vector Machines for Classification and Regression; Technical Report; University of Southampton: Southampton, UK, 1998. [Google Scholar]
- Bruges, C.J.C. A Tutorial on Support Vector Machines for Pattern Recognition. Data Min. Knowl. Discov. 1998, 2, 121–167. [Google Scholar] [CrossRef]
- Rasmussen, C.E. Gaussian processes in machine learning. In Advanced Lectures on Machine Learning; Bousquet, O., von Luxburg, U., Ratsch, G., Eds.; Springer: Berlin/Heidelberg, Germany, 2004; pp. 63–71. [Google Scholar]
- Rasmussen, C.E.; Williams, C.K.I. Gaussian Processes for Machine Learning; The MIT Press: Cambridge, UK, 2010; 266p. [Google Scholar]
- Breiman, L. Random Forests. Mach. Learn. 2001, 45, 5–32. [Google Scholar] [CrossRef] [Green Version]
- Ali, J.; Khan, R.; Ahmad, N.; Maqsood, I. Random Forests and Decision Trees. Int. J. Comput. Sci. Issues 2012, 9, 272–278. [Google Scholar]
- Zhou, W.; Eckler, S.; Barszczyk, A.; Waese-Perlman, A.; Wang, Y.; Gu, X.; Feng, Z.-P.; Peng, Y.; Lee, K. Waist circumference prediction for epidemiological research using gradient boosted trees. BMC Med. Res. Methodol. 2021, 21, 47. [Google Scholar] [CrossRef] [PubMed]
- Lundberg, S.A.; Lee, S.I. Unified Approach to Interpreting Model Prediction. In Proceedings of the 31st Conference on Neural Information Processing System(NIPS), Long Beach, CA, USA, 4–9 December 2017; pp. 1–10. [Google Scholar]
- Van der Meer, J.W.; Briganti, R.; Zanuttigh, B.; Wang, B. Wave transmission and reflection at low-crested structures: Design formulae, oblique wave attack and spectral change. Coast. Eng. 2005, 52, 915–929. [Google Scholar] [CrossRef]
- D’Angremond, K.; Van der Meer, J.W.; De Jong, R.J. Wave transmission at low-crested structures. In Proceedings of the 25th Coastal Engineering Conference, Orlando, FL, USA, 2–6 September 1996; pp. 2418–2426. [Google Scholar]
- Bleck, M.; Oumeraci, H. Hydraulic Performance of Artificial Reefs: Global and Local Description. In Proceedings of the 28th International Conference on Coastal Engineering, Cardiff, UK, 7–12 July 2002; pp. 1778–1790. [Google Scholar]
- Seelig, W.N. Two Dimensional Tests of Wave Transmission and Reflection Characteristics of Laboratory Breakwaters; Technical Report 80-1; Coastal Engineering Research Center: Fort Belvoir, VA, USA; Army Corps of Engineers Waterways Experiment Station: Vicksburg, MS, USA, 1980; p. 187. [Google Scholar]
- Daemrich, K.; Kahle, W. Schutzwirkung von Unterwasser Wellen Brechern unter dem Einfluss Unregelmassiger Seegangswellen; Technical Report, Report Heft 61; Franzius-Instituts fur Wasserbau und Kusteningenieurswesen: Hannover, Germany, 1985. [Google Scholar]
- Daemen, I.F.R. Wave Transmission at Low-Crested Structures. Master’s Thesis, Delft University of Technology, Delft, The Netherlands, 1991. [Google Scholar]
- Shin, S.W.; Bae, I.R.; Lee, J.I. Physical Modelling of the Wave Transmission over a Tetrapod Armored Artificial Reef. J. Coast. Res. 2019, 91, 126–130. [Google Scholar] [CrossRef]
- Gandomi, M.; Dolatshahi Pirooz, M.; Varjavand, I.; Nikoo, M.R. Permeable Breakwaters Performance Modeling: A Comparative Study of Machine Learning Techniques. Remote Sens. 2020, 12, 1856. [Google Scholar] [CrossRef]
- Hydraulics, D. AmWaj Island Development, Bahrain: Physical Modelling of Submerged Breakwaters; Report H4087; Delft Hydraulics: Delft, The Netherlands, 2002. [Google Scholar]
- Allsop, N.W.H. Low-Crested Breakwaters, Studies in Random Waves. In Proceedings of the Coastal Structures’83, Arlington, VA, USA, 9–11 March 1983; pp. 94–107. [Google Scholar]
Parameter | Definition | Average | Max | Min | |
---|---|---|---|---|---|
X1 | RC/H0 | Relative freeboard | −0.494 | 4.0 | −8.696 |
X2 | B/H0 | Relative crest width | 4.525 | 43.478 | 0.889 |
X3 | surf similarity parameter | 4.145 | 10.541 | 1.181 | |
X4 | B/L0 | Ratio of the crest width to wave length | 0.09 | 0.424 | 0.012 |
X5 | RC/h | the relative freeboard to water depth ratio | −0.065 | 0.734 | −0.56 |
X6 | Dn50/hc | Ratio of the nominal diameter to structure height | 0.166 | 0.336 | 0.065 |
X7 | hc/h | The relative structure height | 0.935 | 1.734 | 0.44 |
Y | Kt | Transmission coefficient | 0.482 | 0.922 | 0.049 |
Performance Measures | ||||
---|---|---|---|---|
ML Regressor | MSE | I | SI | R2 |
M1. AdaBoost | 0.0013 | 0.995 | 0.073 | 0.979 |
M2. Gradient boost | 0.0010 | 0.996 | 0.065 | 0.983 |
M3. ANN | 0.0013 | 0.995 | 0.078 | 0.979 |
M4. RF | 0.0013 | 0.995 | 0.073 | 0.979 |
M5. SVM | 0.0022 | 0.991 | 0.101 | 0.965 |
M6. GPR | 0.0038 | 0.984 | 0.132 | 0.941 |
M7. Lasso | 0.0049 | 0.981 | 0.150 | 0.924 |
M8. Ridge | 0.0050 | 0.980 | 0.152 | 0.922 |
M9. KR | 0.0028 | 0.989 | 0.113 | 0.957 |
M10. Linear | 0.0121 | 0.944 | 0.235 | 0.814 |
Splitting Ratio | MSE | I | SI | R2 |
---|---|---|---|---|
7:3 | 1.0 × 10−3 | 0.996 | 0.065 | 0.983 |
8:2 | 0.8 × 10−3 | 0.997 | 0.058 | 0.988 |
9:1 | 0.8 × 10−3 | 0.997 | 0.051 | 0.988 |
Folds | Performance Measures | |||
---|---|---|---|---|
MSE | R2 | MAPE | MAE | |
Fold 1 | 1.4 × 10−3 | 0.971 | 0.073 | 0.029 |
Fold 2 | 1.5 × 10−3 | 0.969 | 0.060 | 0.027 |
Fold 3 | 1.0 × 10−3 | 0.987 | 0.076 | 0.024 |
Fold 4 | 1.4 × 10−3 | 0.979 | 0.080 | 0.024 |
Fold 5 | 1.1 × 10−3 | 0.982 | 0.056 | 0.023 |
Fold 6 | 2.1 × 10−3 | 0.965 | 0.114 | 0.030 |
Fold 7 | 1.5 × 10−3 | 0.979 | 0.082 | 0.026 |
Fold 8 | 2.7 × 10−3 | 0.958 | 0.070 | 0.034 |
Fold 9 | 1.6 × 10−3 | 0.961 | 0.108 | 0.033 |
Fold 10 | 1.3 × 10−3 | 0.978 | 0.085 | 0.027 |
Mean | 1.6 × 10−3 | 0.973 | 0.080 | 0.027 |
STD | 0.4 × 10−3 | 9.0 × 10−3 | 18.7 × 10−3 | 3.8 × 10−3 |
Combinations | Performance Measures | ||||||
---|---|---|---|---|---|---|---|
MSE | I | SI | R2 | ||||
Test | Train | Test | Train | Test | Train | ||
1: X1, X2, X3, X4, X5, X6, X7 | 0.8 × 10−3 | 0.997 | 0.999 | 0.058 | 0.032 | 0.988 | 0.999 |
2: X2, X3, X4, X5, X6, X7 | 1.3 × 10−3 | 0.995 | 0.999 | 0.073 | 0.040 | 0.981 | 0.998 |
3: X1, X4, X5, X6, X7 | 1.1 × 10−3 | 0.996 | 0.999 | 0.066 | 0.070 | 0.984 | 0.995 |
4. X1, X2, X3, X4, X6 | 1.6 × 10−3 | 0.994 | 0.999 | 0.082 | 0.041 | 0.977 | 0.998 |
5. X1, X4, X5, X7 | 1.3 × 10−3 | 0.995 | 0.999 | 0.072 | 0.071 | 0.982 | 0.995 |
6. X2, X3, X4, X6 | 22.5 × 10−3 | 0.894 | 0.967 | 0.303 | 0.351 | 0.675 | 0.845 |
7. X2, X3, X6 | 22.9 × 10−3 | 0.889 | 0.962 | 0.307 | 0.370 | 0.668 | 0.820 |
8. X1, X5, X7 | 3.7 × 10−3 | 0.986 | 0.987 | 0.123 | 0.230 | 0.947 | 0.948 |
Methods | MSE | I | SI | R2 |
---|---|---|---|---|
Van der Meer (2005) | 9.0 × 10−3 | 0.974 | 0.221 | 0.810 |
D’Angremond (1996) | 6.0 × 10−3 | 0.983 | 0.161 | 0.840 |
Bleck and Oumeraci (2001) | 17.0 × 10−3 | 0.960 | 0.238 | 0.710 |
Gradient boosting | 0.8 × 10−3 | 0.997 | 0.058 | 0.988 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Kim, T.; Kwon, S.; Kwon, Y. Prediction of Wave Transmission Characteristics of Low-Crested Structures with Comprehensive Analysis of Machine Learning. Sensors 2021, 21, 8192. https://doi.org/10.3390/s21248192
Kim T, Kwon S, Kwon Y. Prediction of Wave Transmission Characteristics of Low-Crested Structures with Comprehensive Analysis of Machine Learning. Sensors. 2021; 21(24):8192. https://doi.org/10.3390/s21248192
Chicago/Turabian StyleKim, Taeyoon, Soonchul Kwon, and Yongju Kwon. 2021. "Prediction of Wave Transmission Characteristics of Low-Crested Structures with Comprehensive Analysis of Machine Learning" Sensors 21, no. 24: 8192. https://doi.org/10.3390/s21248192
APA StyleKim, T., Kwon, S., & Kwon, Y. (2021). Prediction of Wave Transmission Characteristics of Low-Crested Structures with Comprehensive Analysis of Machine Learning. Sensors, 21(24), 8192. https://doi.org/10.3390/s21248192