The simulation of seafloor-steel catenary interaction and prediction of riser fatigue life required an accurate characterization of seafloor stiffness as well as realistic description of riser load-deflection (P-y) response. This thesis presents two load-deflection (P-y) models (non-degradating and degradating models) to simulate seafloor-riser interaction. These two models considered the seafloor-riser system in terms of an elastic steel pipe supported on non-linear soil springs with vertical motions. These two models were formulated in terms of a backbone curve describing self-embedment of the riser, bounding curves describing P-y behavior under extremely large deflections, and a series of rules for describing P-y behavior within the bounding loop. The non-degradating P-y model was capable of simulating the riser behavior under very complex loading conditions, including unloading (uplift) and re-loading (downwards) cycles under conditions of partial and full separation of soils and riser. In the non-degradating model, there was a series of model parameters which included three riser properties, two trench geometry parameters and one trench roughness parameter, two backbone curve model parameters, and four bounding loop model parameters. To capture the seafloor stiffness degradation effect due to cyclic loading, a degradating P-y model was also developed. The degradating model proposes three degradation control parameters, which consider the effects of the number of cycles and cyclic unloading-reloading paths. Accumulated deflections serve as a measure of energy dissipation. The degradating model was also made up of three components. The first one was the backbone curve, same as the non-degradating model. The bounding loops define the P-y behavior of extreme loading deflections. The elastic rebound curve and partial separation stage were in the same formation as the non-degradating model. However, for the re-contact and re-loading curve, degradation effects were taken into the calculation. These two models were verified through comparisons with laboratory basin tests. Computer codes were also developed to implement these models for seafloor-riser interaction response.

The simulation of seafloor-steel catenary interaction and prediction of riser fatigue life required an accurate characterization of seafloor stiffness as well as realistic description of riser load-deflection (P-y) response. This thesis presents two load-deflection (P-y) models (non-degradating and degradating models) to simulate seafloor-riser interaction. These two models considered the seafloor-riser system in terms of an elastic steel pipe supported on non-linear soil springs with vertical motions. These two models were formulated in terms of a backbone curve describing self-embedment of the riser, bounding curves describing P-y behavior under extremely large deflections, and a series of rules for describing P-y behavior within the bounding loop. The non-degradating P-y model was capable of simulating the riser behavior under very complex loading conditions, including unloading (uplift) and re-loading (downwards) cycles under conditions of partial and full separation of soils and riser. In the non-degradating model, there was a series of model parameters which included three riser properties, two trench geometry parameters and one trench roughness parameter, two backbone curve model parameters, and four bounding loop model parameters. To capture the seafloor stiffness degradation effect due to cyclic loading, a degradating P-y model was also developed. The degradating model proposes three degradation control parameters, which consider the effects of the number of cycles and cyclic unloading-reloading paths. Accumulated deflections serve as a measure of energy dissipation. The degradating model was also made up of three components. The first one was the backbone curve, same as the non-degradating model. The bounding loops define the P-y behavior of extreme loading deflections. The elastic rebound curve and partial separation stage were in the same formation as the non-degradating model. However, for the re-contact and re-loading curve, degradation effects were taken into the calculation. These two models were verified through comparisons with laboratory basin tests. Computer codes were also developed to implement these models for seafloor-riser interaction response.

Realistic predictions of service life of steel catenary risers (SCR) require an accurate characterization of seafloor stiffness in the zone where the riser contacts the seafloor, the so- called touchdown area (TDA). This paper describes the key features of a seafloor-riser interaction model based on the previous experimental model tests. The seafloor is represented in terms of non-linear load-deflection (P-y) relationships, which are also able to account for soil stiffness degradation due to vertical cyclic loading. The P-y approach has some limitations, but simulations show good agreement with experimental data. Hence, stiffness degradation and rate effects during penetration and uplift motion (suction force increase) of the riser are well captured through comparison with previous experimental tests carried out at the Centre for Offshore Foundation Systems (COFS) and Norwegian Geotechnical Institute (NGI). The analytical framework considers the riser-seafloor interaction problem in terms of a pipe resting on a bed of springs, and requires the iterative solution of a fourth-order ordinary differential equation. A series of simulations is used to illustrate the capabilities of the model. Due to the non-linear soil springs with stiffness degradation it is possible to simulate the trench formation process and estimate deflections and moments along the riser length. The seabed model is used to perform parametric studies to assess the effects of stiffness, soil strength, amplitude of pipe displacements, and riser tension on pipe deflections and bending stresses. The input parameters include the material properties (usually pipe and soil), model parameters, and loading conditions such as the amplitude of imposed dis- placements, tension, and moment. Primary outputs from this model include the deflected shape of the riser pipe and bending moments along riser length. The code also provides the location of maximum trench depth and the position where the maximum bending moment occurs and any point where user is interested in.