Abstract
Several types of structures are used in Coastal Engineering with the aim of preventing shoreline erosion, such as groynes, detached breakwaters, submerged breakwaters, etc. Submerged breakwaters have the advantage of their minimal visual impact, which has made them ever more popular (Chang & Liou, 2007). When the incoming waves impinge on a submerged breakwater, a process of energy transformation occurs. Many laboratory and numerical studies have been carried out in order to investigate this process (Kobayashi & Wurjanto, 1989) (Losada, Losada & Martin, 1995) (Losada, Silva & Losada, 1996) (Liu, Lin, Hsu, Chang, Losada, Vidal & Sakakiyama, 2000). The energy of the incident wave is transformed as follows: (i) one part of this energy is transmitted above the crest of the structure and ? in the case of permeable submerged breakwaters ? through its interior; (ii) another part is dissipated by wave breaking and by friction with the structure during the transmission process and finally, (iii) the remaining energy is reflected seaward. The reflection level is related with the scour in front of the structure. Therefore, a good knowledge about the reflection process may be helpful in order to avoid or at least mitigate the possible problems in the structure foundations. However, due to the complexity of the problem, the influence of all the relevant parameters (the structure slope and submergence, the water depth, the wave period and height, etc.) is not entirely understood yet and new approaches are needed. In this work, an Artificial Neural Network (ANN) has been applied to a series of results obtained from a previous study of Taveira-Pinto (2001), in which several physical models were tested. Once trained and validated, the ANN has been used to estimate the wave reflection coefficient.
TopBackground
ANNs have proved to be a very powerful and versatile Artificial Intelligence technique (Orchad, 1993) (Haykin, 1999). In fact, they have been successfully applied to a great number of areas, including system identification and control, pattern recognition, data processing, time series prediction, modelling, etc (Rabuñal, Dorado, Pazos, Pereira & Rivero, 2004) (Rabuñal & Dorado, 2005).
In Civil Engineering, ANNs have been used most notably in Hydrology (Govindaraju & Rao, 2000) (Maier & Dandy, 2000) (Dawson & Wilby, 2001) (Cigizoglu, 2004). With regard to Ocean Engineering, ANN’s have been applied to breakwater stability (Mase, Sakamoto & Sakai, 1995) (Medina, Garrido, Gómez-Martín & Vidal, 2003) (Kim & Park, 2005) (Yagci, Mercan, Cigizoglu & Kabdasli, 2005), wave forecasting (Tsai, Lin, & Shen, 2002) and tide-forecasting (Lee & Jeng, 2002).
Key Terms in this Chapter
Reflection: The process by which the energy of the incoming waves is returned seaward.
Artificial Neural Networks: Interconnected set of many simple processing units, commonly called neurons, that use a mathematical model representing an input/output relation
Back-Propagation Algorithm: Supervised learning technique used by ANNs that iteratively modifies the weights of the connections of the network so the error given by the network after the comparison of the outputs with the desired one decreases.
Submerged Breakwater: Coastal protection structure crowned at, or below, the still water level
Significant Wave Height: In wave record analysis, the average height of the highest one-third of a selected number of waves
JONSWAP Spectrum: Wave spectrum typical of growing deep water waves developed from field experiments and measurements of waves and wave spectra in the Joint North Sea Wave Project.
Peak Period: The wave period determined by the inverse of the frequency at which the wave energy spectrum reaches its maximum.