Abstract
The functional design of submerged breakwaters is still developing, particularly with respect to modelling of the nearshore wave field behind the structure. This paper describes a method for predicting the wave transmission coefficients behind submerged breakwaters using machine learning algorithms. An artificial neural network using the radial-basis function approach has been designed and trained using laboratory experimental data expressed in terms of non-dimensional parameters. A wave transmission coefficient calculator is presented, based on the proposed radial-basis function model. Predictions obtained by the radial-basis function model were verified by experimental measurements for a two dimensional breakwater. Comparisons reveal good agreement with the experimental results and encouraging performance from the proposed model. Applying the proposed neural network model for predictions, guidance is given to appropriately calculate wave transmission coefficient behind two dimensional submerged breakwaters. It is concluded that the proposed predictive model offers potential as a design tool to predict wave transmission coefficients behind submerged breakwaters. A step-by-step procedure for practical applications is outlined in a user-friendly form with the intention of providing a simplified tool for preliminary design purposes. Results demonstrate the model’s potential to be extended to three dimensional, rough, permeable structures.
Similar content being viewed by others
References
Huang W-P, Chou C-R, Yim J-Z (2005) On reflection and diffraction due to a detached breakwater. Ocean Eng 32(14–15):1762–1779
Ilic S, van der Westhuysen AJ, Roelvink JA, Chadwick AJ (2007) Multidirectional wave transformation around detached breakwaters. Coast Eng 54(10):775–789
Du Y, Pan S, Chen Y (2010) Modelling the effect of wave overtopping on nearshore hydrodynamics and morphodynamics around shore-parallel breakwaters. Coast Eng 57(9):812–826
Kristensen S-E, Drønen N, Deigaard R, Fredsoe J (2013) Hybrid morphological modelling of shoreline response to a detached breakwater. Coast Eng 71:13–27
Hur D-S (2004) Deformation of multi-directional random waves passing over an impermeable submerged breakwater installed on a sloping bed. Ocean Eng 31(10):1295–1311
Jeng D-S, Schacht C, Lemckert C (2005) Experimental study on ocean waves propagating over a submerged breakwater in front of a vertical seawall. Ocean Eng 32(17–18):2231–2240
Tsai C-P, Chen H-B, Lee F-C (2006) Wave transformation over submerged permeable breakwater on porous bottom. Ocean Eng 33(11–12):1623–1643
Wang BX, Otta AK, Chadwick AJ (2007) Transmission of obliquely incident waves at low-crested breakwaters: theoretical interpretations of experimental observations. Coast Eng 54(4):333–344
Zanuttigh B, Martinelli L, Lamberti A (2008) Wave overtopping and piling-up at permeable low crested structures. Coast Eng 55:484–498
Calabrese M, Vicinanza D, Buccino M (2008) 2D wave setup behind submerged breakwaters. Ocean Eng 35(10):1015–1028
Zhang J-S, Jeng D-S, Liu PL-F, Zhang C, Zhang Y (2012) Response of a porous seabed to water waves over permeable submerged breakwaters with Bragg reflection. Ocean Eng 43:1–12
Sharif Ahmadian A, Simons RR (2014) A 3D numerical model of nearshore wave field behind submerged breakwaters. Coast Eng 83:190–204. doi:10.1016/j.coastaleng.2013.10.016
Lee T-L, Jeng D-S (2002) Application of artificial neural networks in tide-forecasting. Ocean Eng 29(9):1003–1022
Huang W, Murray C, Kraus N, Rosati J (2003) Development of a regional neural network for coastal water level predictions. Ocean Eng 30(17):2275–2295
Kim D-H, Park W-S (2005) Neural network for design and reliability analysis of rubble mound breakwaters. Ocean Eng 32(11–12):1332–1349
Rao S, Mandal S (2005) Hindcasting of storm waves using neural networks. Ocean Eng 32(5–6):667–684
Lee T-L (2006) Neural network prediction of a storm surge. Ocean Eng 33(3–4):483–494
van Gent MRA, van den Boogaard HFP, Pozueta B, Medina JR (2007) Neural network modelling of wave overtopping at coastal structures. Coast Eng 54(8):586–593
Cha D, Zhang H, Blumenstein M (2011) Prediction of maximum wave-induced liquefaction in porous seabed using multi-artificial neural network model. Ocean Eng 38(7):878–887
Kim D-H, Kim Y-J, Hur D-S (2014) Artificial neural network based breakwater damage estimation considering tidal level variation. Ocean Eng 87:185–190
Bhaskaran PK, Rajesh Kumar R, Barman Rahul, Ravichandran M (2010) A new approach for deriving temperature and salinity fields in the Indian Ocean using artificial neural networks. J Mar Sci Technol 15(2):160–175
Bhaskaran PK, Rajesh Kumar R, Dube SK, Murty TS, Gangopadhyay A, Chaudhuri A, Rao AD (2008) Travel time atlas and the role of neural networks for an early warning system for tsunamis in the Indian Ocean. In: Conference paper, December 2008
Bhaskaran PK, Rajesh Kumar R, Dube SK, Rao AD, Murty T, Gangopadhyay A, Chaudhuri A (2008) Tsunami early warning system-an Indian Ocean perspective. J Earthq Tsunami 2(3):197–226
Barman R, Prasad Kumar B, Pandey PC, Dube SK et al (2006) Tsunami travel time prediction using neural networks. Geophys Res Lett 33:L16612. doi:10.1029/2006GL026688
van der Meer J, Briganti R, Wang B, Zanuttigh B (2004) Wave transmission at low-crested structures, including oblique wave attack. In: Proceedings of 29th international conference on coastal engineering, ASCE vol 41, pp 52–64
van der Meer JW, Briganti R, Zanuttigh B, Wang B (2005) Wave transmission and refection at low-crested structures: design formulae, oblique wave attack and spectral change. Coast Eng 52:915–929
Panizzo A, Briganti R (2007) Analysis of wave transmission behind low-crested breakwaters using neural networks. Coast Eng 54(9):643–656
Buccino M, Calabrese M (2007) Conceptual approach for prediction of wave transmission at low-crested breakwaters. J Waterw Port Coast Ocean Eng 133(3):213–224
Goda Y, Ahrens JP (2008) New formulation of wave transmission over and through low-crested structures. In: Proceedings of the 31st international conference of coastal engineering, vol 4
Kurban T, Beşdok E (2009) A comparison of RBF neural network training algorithms for inertial sensor based terrain classification. Sensors 9:6312–6329. doi:10.3390/s90806312
Adams C, Sonu C (1987) Wave transmission across submerged near-surface breakwaters. Coast Eng 1986:1729–1738. doi:10.1061/9780872626003.126
d’Angremond K, van der Meer J, De Jong R (1996) Wave transmission at low-crested structures. In: Proceedings of 25th international conference on coastal engineering, ASCE, vol 24, pp 18–27
Seabrook S, Hall K (1998) Wave transmission at submerged rubblemound breakwaters. In: Proceedings of 26th international conference on coastal engineering, vol 2, pp 2000–2013
van Oosten R, Peixo Marco J, van der Meer JW, Verhagen H (2006) Wave transmission at low-crested structures using neural networks. In: International conference on coastal engineering, San Diego, ASCE, pp 4932–4944
Sharif Ahmadian A, Simons RR (2012) 3-D wave field around submerged breakwater. ASCE, New York. doi:10.9753/icce.v33.structures.13
Sharif Ahmadian A (2016) Numerical models for submerged breakwaters. Coastal hydrodynamics and morphodynamics, 1st edn. Elsevier, Amsterdam
Stucky A, Bonnard D (1937) Contribution to the experimental study of marine rock fill dikesbull. Technical report, Technique de Suizze Romande
Beach-Erosion-Board (1940) A model study of the effect of submerged breakwaters on wave action. Technical memorandum No. 1. Chief of Engineers, U.S. War Department
Morison J (1949) Model study of wave action on underwater barriers. Technical report, Report HE-116-304, Inst. Eng. Res., Univ. Calif., Berkley
Johnson JW, Fuchs RA, Morison JR (1951) The damping action of submerged breakwaters. Trans Am Geophys Union 32(5):704–718
Goda Y, Takeda H, Moriya Y (1967) Laboratory investigation on wave transmission over breakwaters. Report of the port and Harbour Research Institute 13, 38
Goda Y (1969) Re-analysis of laboratory data on wave transmission over breakwaters. Report of the port and Harbour Research Institute, vol 18(3), pp 3–18
Seelig W (1980) Two-dimensional tests of wave transmission and reflection characteristics of laboratory breakwaters. Technical report, CERC, Fort Belvoir, Report No. 80-1
Allsop N (1983) Low-crested breakwaters, studies in random waves. In: Proceedings of coastal structure 83 ASCE, pp 94–107
Ahrens J (1987) Characteristics of reef breakwaters. Technical report CERC-87-17, 45
Davies B, Kriebel D (1992) Model testing on wave transmission past low crested breakwaters. In: Proceedings of 23rd international conference on coastal engineering ASCE, pp 1115–1128
Daemrich K, Kahle W (1985) Schutzwirkung von unterwasser wellen brechern unter dem einfluss unregelmassiger seegangswellen. Technical Report, Franzius Instituts fur Wasserbau und Kusteningenieurswesen, Report Heft 61 (in Germany)
Powell K, Allsop N (1985) Low-crested breakwaters, hydraulic performance and stability. Technical report SR 57, HR Wallingford
van der Meer J (1988) Rock slopes and gravel beaches under wave attack. Ph.D. thesis, Delft University of Technology, Delft Hydraulics Report No. 396
Daemen I (1991) Wave transmission at low crested structures. Ph.D. thesis, M.Sc. thesis, Delft University of Technology, Faculty of Civil Engineering, Delft Hydraulics Report H 462
Seabrook S, Hall K (1997) Effect of crest width and geometry on submerged breakwater performance. In: Proceedings of Canadian coastal conference CCSEA, pp 58–72
Bleck M, Oumeraci H (2002) Hydraulic performance of artificial reefs: global and local description. In: Proceedings of 28th international conference on coastal engineering ASCE, pp 1778–1790
Calabrese M, Vicinanza D, Buccino M (2002) Large-scale experiments on the behaviour of low crested and submerged breakwaters in presence of broken waves. In: Proceedings of 28th international conference on coastal engineering ASCE, pp 1900–1912
Kramer M, Zanuttigh B, Vandermeer J, Vidal C, Gironella F (2005) Laboratory experiments on low-crested breakwaters. Coast Eng 52:867–885
van der Meer J (1991) Stability and transmission at low-crested structures. Technical report, Delft Hydraulic, Report No. H 453, 33p
van der Meer J, Daemen I (1994) Stability and wave transmission at low-crested rubble-mound structures. J Waterw Port Coast Ocean Eng. doi:10.1061/(ASCE)0733-950X(1994)120:1(1)
Gironella X, Sanchez-Arcilla A, Brigantti R, Sierra JP, Moreno L (2002) Submerged detached breakwaters: towards a functional design. In: Proceedings of 28th coastal engineering conference, Cardiff, Wales, UK, pp 1768–1777
Garcia N, Lara J, Losada I (2004) 2-d numerical analysis of near-field flow at low-crested permeable breakwaters. J Coast Eng 51:991–1020
Melito I, Melby J (2002) Wave runup, transmission, and reflection for structures armored with core-loc. J Coast Eng 42:33–52
Hirose N, Watanuki A, Saito M (2002) New type units for artificial reef development of ecofriendly artificial reefs and the effectiveness thereof. In: Proceedings of 30th international navigation congress. PIANC, p Cd ROM
Tanaka N (1976) Wave deformation and beach stabilization capacity of wide-crested submerged breakwaters. In: Proceedings of 23rd Japanese conference coastal engineering, pp 152–157 (in Japanese)
Stein R (1993) Selecting data for neural networks. AI Expert 8(2):42–47
Haykin S (1999) Neural networks, a comprehensive foundation, 2nd edn. Prentice Hall, New Jersey
Hagan MT, Demuth HB, Beale MH (1996) Neural network design. PWS Publication, Boston
Wilmott CJ (1981) On the validation of models. Prog Phys Geogr 2:184–194
Haller MC, Darlymple RA, Svendsen IA (2002) Experimental study of nearshore dynamics on a barred beach with rip channels. J Geophys Res 107(C6):14-1–14-21
Draper NR (1984) The Box–Wetz criterion versus R2. J Stat Soc 147:100–103
Krause P, Boyle DP, Base F (2005) Comparison of different efficiency criteria for hydrological model assessment. Adv Geosci 5:89–97
Tukey JW (1977) Box-and-whisker plots. In: Explanatory data analysis. Addison-Wesley, Reading, pp 39–43
McGill R, Tukey JW, Larsen WA (1978) Variations of boxplots. Am Stat 32:12–16
Park J, Sandberg IW (1993) Approximation and radial-basis function networks. Neural Comput 5:305–316
Broomhead D, Lowe D (1988) Multivariable functional interpolation and adaptive networks. Complex Syst 2:321–355
Poggio T, Girosi F (1990) Networks for approximation and learning. Proc IEEE 78:1481–1497
Hamadneh N, Sathasivam S, Choon OH (2012) Higher order logic programming in radial basis function neural network. Appl Math Sci 6(3):115–127
Robert J, Howlett LCJ (2001) Radial basis function networks 2: new advances in design. Physica-Verlag, Herdelberg
Liu Y, Zheng Q, Shi Z, Chen J (2004) Training radial basis function networks with particle swarms. Lecture Notes Computer Science, vol 3173, pp 317–322
Simon D (2002) Training radial basis neural networks with the extended Kalman filter. Neurocomputing 48:455–475
Karayiannis NB (1999) Reformulated radial basis neural networks trained by gradient descent. IEEE Trans Neural Netw 10(3):657–671
Bullinaria J (2004) Introduction to neural networks. Lecture notes
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix 1: A simplified tool and initial prediction scheme with a step by step procedure (K t calculator)
The algorithm used in this appendix is based closely on the algorithm introduced by [12] and the later work by Sharif Ahmadian [36].
As explained in Sect. 5, the final optimized RBF network has one radial basis function layer with 15 nodes. A K t calculator including details about the model mathematics is described below, based on the proposed non-dimensional RBF model. A step-by-step procedure for practical applications is outlined in a user-friendly form [12, 36] with the intention of providing an initial prediction scheme according to the proposed ANN model described in the paper. For more details and theoretical aspects see [63] or [64].
Referring first to the non-dimensional parameters used throughout the paper, an input vector I should be considered:
The normalization process according to the Eq. (7) is performed:
where vectors I min, I max are characterized by values:
Considering the center and weight matrices defined below, the input signals are passed through the network. The center matrix c and weight vector w are applied in the transfer functions Eqs. (26) and (30) respectively. The values in the last column of the center matrix c and the weight vector w are the bias values for the hidden and output layers respectively.
Activation functions of radial basis type are used to determine the values of hidden neurons, such as:
where
and constant value d is:
where sp = 1.0.
The Euclidean distance between the vectors a o and c, is defined as follows:
where c i and a o,i are the coordinates of c and a o,i in dimension i.
Whereas a linear activation function has been applied, the scaled model outcome can be computed as follows:
where b 2 = 0.3438 is the bias neuron for the output layer.
Finally, the model output needs to be rescaled. Thus, K t is calculated as follows:
where O min = 0.1187 and O max = 0.8416.
The main aim is to introduce a simplified tool for preliminary design purposes which can be used in submerged breakwater design. Although the model is calibrated and designed using regular wave data, similar predictions can be made using significant wave height (H s )/zero moment wave height (H mo) and peak wave period, T p as inputs. However, some considerations might be required in applying regular wave results to random sea‐states [12, 36].
Appendix 2: Radial basis function (RBF) network
This appendix gives a concise description of the training algorithm adopted in this paper for the RBF network.
In engineering applications, the most popular artificial neural network is the multi-layer perceptron (MLP). Radial basis function (RBF) networks with universal non-linear approximation properties are similar to the multi-layer perceptron in that they also use memory-based learning algorithms for their design [63, 71].
RBF networks are used widely for non-linear function approximation comparable to MLP networks [72, 73], with some additional benefits, for instance, RBFs are trained usually with a relatively higher speed [74]. The concept of a RBF neural network was introduced by Broomhead and Lowe [72]. In the same way as for a MLP network, the entire data set is split into two main data sets and training and testing performed in two distinct stages [72, 73].
While the number of layers in MLP can be variable, RBF networks are always designed with three layers of neurons: input layer, hidden layer and output layer. Working as a feed-forward RBF network: the input layer receives the input data, the hidden layer computes the outcome of the RBF units and the output layer combines the outputs from the RBF units linearly. The input and output layers in a RBF network are thus the same as in MLP, while the hidden layer is not. The main difference is related to the transfer functions applied in the hidden layer. The Euclidean distance between the input vector and the RBF unit’s centre vector c is calculated in the RBF networks while an MLP unit applies a transfer function which computes the inner product of the weight and the input vectors [63].
RBF models include biases in their hidden and output layers similar to MLP. However, RBF models usually have more computational units in the hidden layer than MLP as their transfer functions respond to a smaller area of the input space than MLP. This is of benefit as it requires less training time to design RBF compared with MLP. The outputs are also weighted. Hence, an expression of the RBF network can be presented as [75]:
where \(\overset{\lower0.5em\hbox{$\smash{\scriptscriptstyle\frown}$}}{x}\) is the input vector, \(w_{1} ,w_{2} , \ldots w_{n}\) are weight factors, c k represents the centre of the kth RBF unit, b is bias value and Φ is the hidden layer transfer function (Gaussian function):
where s k is the scaling factor of the kth RBF unit [75].
Basically, the learning process in RBF networks is a form of supervised learning. Determining and making an appropriate decision on the number of computational neurons is basically an essential step since it influences the network complexity and its generalizing capability. Insufficient computational units mean the RBF network may not estimate the underlying function adequately. However, increasing the size of hidden layer (number of the computational neurons) may cause an overlearning situation [76]. In addition, carefully specifying the optimal locations of the centres and scaling factors in the computational neurons of the hidden layer would be important tasks over the training process and very essential to obtain proper results. The most preferred form of transfer functions for a computational neuron is the Gaussian function [75, 77, 78].
The next step is for the weights of the network to be computed. At last, the bias values added to each output are calculated. A derivative-based gradient descent algorithm [78] is used in this study, chosen from the various algorithms proposed in the literature for training RBF networks.
Following the training procedure and having decided on the number of hidden units, c, s, w and b parameters are determined. The sum of squared errors (SSE) is typically applied over the training process to check the manner of functioning of the RBF network and is defined as follows [79]:
where y and \(\hat{y}\) are the actual and desired outputs respectively.
The training procedure for a RBF network is organised using the following procedure [63]. The hidden layer initially has zero number of units. Data samples are presented to the network, signals pass through the network and the error is computed. The calculated output is compared to the target value of the respective sample. An RBF unit is added. The process is performed repeatedly and the connection weight parameters are then computed to meet the desired error. Basically, the calculated error, which is statistically described as the sum of squared errors (SSE), must be lower than the desired error. In addition, the maximum number of hidden neurons should be met before stopping training [63].
In the present case in this study, a 15-node hidden layer network was chosen. The training of the RBF model was terminated once the calculated error reached the desired values (in this study 0.01) or the chosen number of training iterations (here 100) had been completed.
Rights and permissions
About this article
Cite this article
Sharif Ahmadian, A., Simons, R.R. Estimation of nearshore wave transmission for submerged breakwaters using a data-driven predictive model. Neural Comput & Applic 29, 705–719 (2018). https://doi.org/10.1007/s00521-016-2587-y
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-016-2587-y