Seismic Analyses of Concrete Dam, Comparison Between Finite-Element Analyses and Seismic Records

Abstract eng:
Currently, in engineering practice, the main approach for dynamic analysis of concrete dams with finite-element method is still based on coarse simplification: fluid-structure interaction is generally taken into account with Westergaard addedmasses and soil-structure interaction is considered using massless foundation. But an increasing number of authors show that these simplified approaches tend to overestimate stresses. Consequently, dam-foundation rock interaction with radiative damping and dam-water interaction considering compressibility of water, partial absorption of hydrodynamic pressure waves by sediments at the lakebed, and semiunbounded extent of the reservoir and foundation rock domains must be taken into account. After a brief presentation of a more advanced model taking into account the mass of the foundation, viscous-spring boundary input model and potential-based fluid finite elements with absorptive boundaries, two existing Japanese dams, a gravity dam and an arch dam, are extensively studied. Earthquakes were recorded on each dam, allowing back analyses through comparison of simulated and recorded data at the base and crest of the dams. The dynamic response of each dam is simulated using data recorded at the foundation and the response is compared to the data recorded at sensor located at the dam’s crest. Methods are also compared in a stability assessment manner, as done in professional practice. Comparisons between analyses and records confirm that the conventional massless foundation/added masses approaches with 5% concrete damping greatly overestimates the response of the dam, particularly in the case of gravity dams.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket