Modelling wave-structure-foundation interaction for marine gravitystructures

Direction Dr.-Ing. Hisham Elsafti
Team Dr.-Ing. Hisham Elsafti
Funding DFG, EL 865/1-1
Duration 01.11.2016 - 2019

Brief description

1. Motivation and Objectives

Figure 1: Concept of wave-structure-foundation interaction for marine gravity structures (Elsafti, 2015)

Precast marine gravity structures (such as caissons) are advantageous in terms of construction time, quality control, maintenance, multi-purpose use and environmental aspects. Nevertheless, they are more vulnerable to foundation failures, especially to stepwise failures that can render structures well designed against extreme conditions non-functional under relatively moderate, repetitive wave loads. Moreover, the stepwise residual build-up of displacements of marine gravity structures may increase their vulnerability to extreme loads.

The stepwise failure mechanism is a result of highly complex processes associated with wave-structure-foundation interaction (Figure 1). Despite Many transnational projects carried out to enhance the design tools for marine gravity structures as well as the understanding of relevant physical processes associated with wave-structure-foundation interaction, no reliable tools exist for the prediction of the stepwise failure mechanism and no physically sound interpretation of this failure mechanism is provided. The latter statement is particularly valid for the tendency of a marine gravity structure to tilt in the shoreward or the seaward direction when subject to water wave loading. Moreover, the stepwise failure of marine gravity structures is not yet addressed in design manuals and guidelines. In this project, a new fully coupled CFD-CSD modelling tool will be developed based on the earlier work in the Leichtweiss-Institute in order to overcome the shortcomings of previous models and to substantially enhance and refine concepts developed in Elsafti (2015).

Figure 2: The load eccentricity concept and examples for calculating the relative load eccentricity from CFD simulations (Elsafti, 2015)

The following requirements for coupling the prospective CFD-CSD model system are drawn: (1) Necessity of two-way coupling in order to reproduce correctly the uplift pressures on marine gravity structures and for simulating the added fluid mass and fluid damping effects on the structure. Additionally, the coupling should provide an approach for deformable porous media as opposed to non-deformable porous media in CFD models and (2) Simulation of fluid as multiphase with proper momentum exchange between phases (inside and outside porous media) in order to greatly enhance the understanding of the role of air in both breaking wave impact and seabed response to cyclic loading (changes in the degree of soil saturation as a result of loading). Further, this will help to better understand the role of entrapped air in breaking waves in reducing the degree of saturation of the seabed soil under the seaward edge of the structure.

Elsafti (2015) developed the "load eccentricity concept" for the analysis of the stepwise failure of marine gravity structures. The load eccentricity is defined as the maximum shoreward eccentricity of the vertical force resultant Fv from the mid-point of the structure-foundation interface (Figure 3a). The relative load eccentricity (e/B) includes all information about the load and the properties of the gravity structure in a single parameter. Examples of e/B reproduced from the GWK tests by CFD simulations of Elsafti (2015) are illustrated in Figure 3b.

Figure 3: Idealisation of caisson-induced stresses/forces under both caisson edges for different load eccentricity regimes (Elsafti, 2015)

The response of monolithic breakwaters was classified in four load eccentricity regimes: Low, medium, high and extreme. For the low eccentricity regime, the structure does not loose contact with the foundation (e/B ≤ 1/6), for which the own-weight eccentricity plays a crucial role in assessing whether the structure will tilt in seaward or in shoreward direction (Figure 3a,c). Other load eccentricity regimes are associated with breaking wave impact and cause momentary (mostly partial) loss of contact between the structure and the underlying foundation. During the impact and when the structure is trying to restore its initial position before impact, the structure-induced stresses on the foundation can be idealised as forces (Figure 3b and 2b). For the medium load eccentricity regime, the inertial restoration force (Fsea ) is smaller than the impact induced force (Fshore , Figure 3d). However, for the high and extreme regimes Fsea > Fshore (Figure 3e). This means that for the medium eccentricity regime, the structure is most likely to tilt in the shoreward direction, whereas for the high regime, the structure is most likely to tilt in the seaward direction. Therefore, the extreme load eccentricity regime, the structure is subject to excessive sliding coupled with rotation, possibly causing erosion in the rubble foundation shoreside as well as partial embedment of the shoreward edge of the caisson in the foundation. For the extreme regime, the structure is more likely to tilt shoreward. However, the extreme regime may be regarded like a catastrophic failure as opposed to the stepwise failure mechanism. A well designed structure should generally not experience the extreme eccentricity regime. The boundaries between the four eccentricity regimes are tentatively defined by Elsafti (2015), but still need to be specified more precisely based on simulations using an improved CFD-CSD model.

Elsafti (2015) introduced a nonlinear 3-DOF mass-spring-dashpot model in OpenFOAM named "caissonFoam", in which an elastoplastic spring idealized for cyclic loading is developed (changing stiffness, separation of loading and unloading p−y relation and soil densification each load cycle. The model parameters were calibrated for different relative load eccentricities and soil relative densities. The caissonFoam model includes additional nonlinear features such as: simulation of structure-foundation separation/reattachment by deactivating and reactivating vertical supports, a horizontal slider to simulate stepwise sliding of caisson and changing location of pivot according to soil-structure interaction. The caissonFoam model is applied successfully to reproduce the large-scale GWK tests.

Mainly based on lessons learned from previous work in LWI, the overall objective of the proposed project is to develop a strongly coupled CFD-CSD model system for the analysis of coastal and offshore structures, with the main focus on reproducing the stepwise failure mechanism of marine gravity structures. The proposed model system will be validated using the large-scale GWK tests from the EU-LIMAS project (Kudella et al., 2006).

2. Work Programme and Methodology

Figure 4: Validation of the CFD-CSD model system using large-scale GWK tests for regular breaking waves (H = 0.7m. and T = 6.5 s.) (Elsafti, 2015)

The semi-coupled CFD-CSD model system of Elsafti (2015) was used successfully to reproduce the large-scale GWK caisson breakwater tests from Kudella et al. (2006) (Figure 4). The model system can reproduce stepwise (per wave event) residual displacements and consequent pore pressure build-up in the seabed underneath a marine gravity structure. It was found that the numerically computed pore pressure amplitude is overestimated under the seaward edge (Figure 4b) but underestimated under the shoreward edge (Figure4d). Possible reasons for the difference are: (1) difference in the degree of saturation in the soil underneath both caisson edges due to loading (effect of seabed response on changes in the air content on the pore fluid) and (2) damping effects provided by the fluid around the structure. Both factors are not accounted for in the model and need thus to be considered in its further development.

The validated model system will then be applied to perform a systematic parameter study which is aimed at substantially improving the understanding of the highly complex physical processes associated with wave-structure-foundation interaction with a particular focus on marine gravity structures. More specifically, the results of the parameter study will allow us to answer questions raised by the results of the analysis of the current knowledge (e.g. the role of breaking waves induced loading on gravity structures in changing the underlying soil degree of saturation). Moreover, the results will also allow to further develop the aforementioned "load eccentricity concept" by including the effects of own-weight eccentricity and structural configurations such as side berms, and to establish a more profound understanding and a more precise specification of the limits between the four load eccentricity regimes. Likewise, the simplified caissonFoam model will be extended in order to cope with the considered more realistic structural configurations.

3. Prospective Results

The specific objectives of the proposed project may be summarized as follows: (i) Development of a strongly coupled CFD-CSD model system (including enhancements to overcome the shortcomings of the semi-coupled model system of Elsafti (2015); e.g. multiphase pore fluid), (ii) Enhancement of the speed of the geotechnical model, (iii) Validation of the model system with large-scale caisson breakwater experiments, (iv) Enhancement of the understanding of physical processes associated with gravity structures response to breaking wave impacts and further development of the load eccentricity concept, (v) Further development of the simplified caissonFoam model by Elsafti (2015) for more realistic conditions and (vi) Documentation and publication of the developed tools under the General Public License (GPL), thus providing a platform for further development and a wider range of applications of the prospective model system through academic and other types of collaborations. Please visit (www.geotechfoam.com) for code updates.


The financial support of the German Research Foundation (Deutsche Forschungsgemeinschaft; DFG) for the WaSFI project (EL 865-1/1) is gratefully acknowledged.



Elsafti, H. (2017): Analysis of stepwise failure of marine gravity structures and implications for design practice. The 2nd place winning paper of the international PIANC De-Paepe Willems award. To be published in the PIANC Yearbook 2017.

Elsafti, H. and Oumeraci, H. (2017): A semi-coupled CFD-CSD model for wave-structure-foundation interaction of marine gravity structures. Applied Ocean Research (submitted), Elsevier.

Elsafti, H. and Oumeraci, H. (2017): A simplified model for stepwise failure of monolithic breakwaters. Journal of Waterway, Port, Coastal, and Ocean Engineering. In review, American Society of Civil Engineers (ASCE).

Elsafti, H. and Oumeraci, H. (2017): Analysis and classification of stepwise failure of monolithic breakwaters. Coastal Engineering. Accepted, Elsevier.

Elsafti, H. and Oumeraci, H. (2016): A numerical hydro-geotechnical model for marine gravity structures. Computers and Geotechnics 79, pp. 105-129.

Elsafti, H. (2015): Modelling and Analysis of Wave-Structure-Foundation Interaction for Monolithic Breakwaters. PhD thesis, Leichtweiss-Institute for Hydraulic Engineering and Water Resources, TU-Braunschweig, Germany, 2015. Available online: http://www.digibib.tu-bs.de/?docid=00060996.

El Safti, H., Bonakdar, L., Oumeraci, H. (2014): A Hybrid 2D-3D CFD Model System for Offshore Pile Groups Subject to Wave Loading. 33rd International Conference on Ocean, Offshore and Arctic Engineering (OMAE), San Francisco, USA.

El Safti, H., and Oumeraci, H. (2014): A Numerical Wave-Structure-Soil Interaction Model with Application to Monolithic Breakwaters Subject to Breaking Wave Impact. 11th International Conference on Hydroscience & Engineering (ICHE), Hamburg, Germany.

El Safti, H., and Oumeraci, H. (2013): Modelling Sand Foundation Behaviour underneath Caisson Breakwaters Subject to Breaking Wave Impact, Proceedings of the 32nd International Conference on Ocean, Offshore and Arctic Engineering, OMAE 2013, Nantes, France

El Safti, H. (2013): A Numerical Wave-Structure-Soil Interaction Model for Monolithic Breakwaters Subject to Breaking Wave Impact, student paper competition, proceedings of the PORTS'13 Confetence, Seattle, USA

El Safti, H., Kuddela, M. and Oumeraci, H. (2012): Modelling wave-induced residual pore pressure and deformation of sand foundations underneath caisson breakwaters. proceedings of the 33rd international conference on Coastal Engineering (ICCE 2012), Santander, Spain.

Kudella, M.; Oumeraci, H.; de Groot, M.B.; Meijers, P. (2006): Large-scale experiments on pore pressure generation underneath a caisson breakwater. ASCE, Journal of Waterway, Port, Coastal and Ocean Engineering 132(4), pp. 310-324.

De Groot, M.B.; Kudella, M.; Meijers, P.; Oumeraci, H. (2006): Liquefaction phenomena underneath marine gravity structures subjected to wave loads. ASCE, Journal of Waterway, Port, Coastal and Ocean Engineering, Special issue on Liquefaction Around Marine Structures 132(4), pp. 325-335.