Performance-based seismic analysis of an anchored sheet pile quay wall

Abstract

Ports are civil works which have a major societal and economic importance. Quay structures are infrastructural elements of primary significance for the functioning of a port system. The ability to economically design quay structures with sufficient seismic resistance is therefore of great importance when situated in areas that are prone to earthquakes. Conventional seismic design is force-based i.e. that structures are designed to have sufficient capacity to withstand a pseudo-static seismic design force. This methodology is associated with no insight in the performance of the structure when exceeding the pseudo-static limit equilibrium state and uneconomic design due to the demand that the structure can resist a very high seismic design force without deforming. A more advanced alternative is Performance-Based Design (PBD) methodology. In this methodology the key design parameters for the seismic performance of structures are stress states and deformations of soil and structure, rather than just a seismic design force. Furthermore it recognizes that varying amounts of permanent deformations associated with different degrees of (repairable) damage are allowable. The present study is embedded in the topic of performance-based seismic design of quay structures. Typical quay types are gravity-based quay walls, sheet pile quay walls and pile-deck structures. The observed trend in seismic quay design is that gravity and sheet pile type structures (i.e. retaining walls) are associated with areas with zero to low seismicity while pile-deck structures are generally the preferred solution in areas with higher seismicity. This can be explained by more favourable seismic performance (i.e. more deformation capacity) of pile-deck structures compared to retaining walls. In line with this trend it is found that PBD methodology is developed to significant lesser extent for retaining walls (especially anchored sheet pile walls) than for pile-deck structures. Therefore the present study focuses on performance-based seismic design of anchored sheet pile quay walls. In the seismic design methodology there are generally three levels of seismic analysis available, i.e. simplified analysis (pseudo-static), simplified dynamic analysis and dynamic analysis. Simplified analysis of anchored sheet pile quay walls is associated with conventional design methodology. Simplified dynamic analysis can be used to obtain a first estimate of permanent-displacement of a structure after exceeding limit equilibrium, based on an assumed failure mode. This type of analysis has to be made more suitable for anchored sheet pile quay walls. In dynamic analysis the seismic behaviour of a structure can be simulated by means of finite element software. Experience has shown that it is desirable to consider sheet pile quay walls in a less conservative way in (preliminary) seismic design for which pseudo-static methodology is commonly applied. Therefore the general objective of the present study is to propose improvements on (simplified) seismic design methodologies for anchored sheet pile quay walls by considering deformation behaviour. For this purpose a research methodology is developed in which pseudo-static, permanent-displacement and FE analysis are employed, calibrated with an experimental reference case that considers a typical anchored sheet pile quay wall. The reference case is taken from a conference paper. It reports on a shake table test under centrifugal gravity which is performed on a scale model of an existing sheet pile quay wall with a batter pile anchor. The quay is situated in homogeneous soil that consists of coarse densified sand. Due to the soil condition liquefaction effects are prevented. Sequential seismic loading of increasing severity is applied during the shake table testing. Measurement results that are reported in the reference case paper comprise bending moments in the sheet pile wall, normal forces in the anchor rod and horizontal displacements of the sheet pile wall. For simplified analysis a calibrated D-SHEET PILING model of the reference case anchored sheet pile quay wall is created. Through an iterative pseudo-static calculation procedure in which D-SHEET PILING and reference case dynamic bending moment results are fitted, it is attempted to find a deformation-based seismic load reduction for structural forces in the sheet pile wall that can be applied in pseudo-static design methodology. For simplified dynamic analysis an analytical limit equilibrium model is developed, based on the failure behaviour of the reference case. The goal of this model is that it can compute the critical acceleration of the anchored quay structure and estimate the sheet pile forces at this critical state. These abilities are validated with PLAXIS 2D and checked with the reference case measurements respectively. Six accelerograms in the reference case soil column, obtained with equivalent linear site-response analysis (with SHAKE2000), are combined with the computed critical acceleration for permanent-displacement (sliding-block) analysis. For dynamic analysis a calibrated PLAXIS 2D model of the reference case anchored sheet pile quay wall is created. Dynamic performance of the PLAXIS 2D model is validated with SHAKE2000 by comparing site-response analysis results of both models. Pseudo-static and pseudo-dynamic calculations are applied to obtain the critical acceleration. Dynamic calculations with six bedrock motions are carried out to simulate the reference case experiment. PLAXIS 2D calculation results are used to validate simplified and simplified dynamic analysis results and to gain insight in the seismic failure behaviour of the anchored sheet pile quay wall. Approaches for (simplified) performance-based seismic analysis of a typical anchored sheet pile quay wall are proposed as a result of the research. For pseudo-static methodology a deformation-based seismic load reduction for structural forces in the sheet pile wall is proposed. For the present reference case it is concluded that a reduction in the range of 45% to 50% is allowable. For simplified dynamic analysis a limit equilibrium model is proposed to compute the critical acceleration of the present quay structure and to estimate sheet pile forces at this critical state. It is concluded that the ability of the limit equilibrium model is satisfactory. Although subjected to uncertainty, permanent-displacement analysis results indicate that the sliding-block analysis, originally developed for embankments, is possibly not suitable for anchored sheet pile quay walls. For dynamic analysis it is concluded that PLAXIS 2D is able to compute the reference case failure behaviour reasonably well, despite some computational setbacks. Complementary is the conclusion that PLAXIS 2D pseudo-static approach proves to be suitable to determine the critical acceleration of an anchored sheet pile structure in contrast to pseudo-dynamic approach which appears less suitable for that matter. In addition the performance-based design principle is linked to the present study so that an idea about the seismic performance limits of anchored sheet pile quay walls in quantitative terms can be provided. As a result of the present study findings it is recommended to perform more extensive research on the ability of permanent-displacement analysis to evaluate the amount of sliding displacement of an anchored sheet pile quay wall. In line with this recommendation it is found that further research on site-response analysis is desirable in the application of simplified dynamic and dynamic analysis. In general it is recommended to create more seismic test cases with different setups for a broader validity of the present results, to develop a seismic test case for the Groningen earthquake situation, to add measurement instrumentation to new and existing structures for verification of research results and to make such (raw) measurement data available to the public.