Development of Performance-Based Testing Specifications for the Arkutun-Dagi Friction Pendulum Bearings

Abstract

Friction Pendulum bearings (FPBs) are seismic isolation devices that reduce the lateral accelerations of the supported structure in an earthquake. This paper describes the application of probabilistic methods to establish criteria for qualification testing of reduced-size prototype bearings. The methodology is used to determine demands to impose during testing (displacements, velocities, etc.) that are representative of desired performance standards (probabilities of exceedence). Previous test programs used in the offshore and onshore industry have used deterministic analysis based on lower bound and upper bound properties to establish demands for the test program. These follow from the deterministic analysis approach used in the global structural design. These methods establish maximum demands to target in the test, but do not give information on their actual probabilities of exceedence. Use of the methodology described in this paper allows engineers to test the FPBs at levels having probabilities of exceedence aligned with other major platform components, ensuring that the test conditions are not excessively conservative or unconservative. Rather, the test conditions comply with a specified probability of non performance of 2% in 50 years achieved with a particular level of confidence (90% generally and 99% for more critical parameters). The performance-based (response-based) approach described in this paper can be applied to a broad range of seismically isolated platforms. These methods were originally developed for onshore seismic design of buildings. Hence, the underlying probabilistic framework is general and can be applied to structural design and qualification testing of other components. Introduction Friction Pendulum Bearings (FPBs) are specialized devices that will be used to isolate the topsides of the Arkutun-Dagi platform from earthquakes. The Arkutun-Dagi platform is part of the Sakhalin-I development which is located in an area of high seismicity. Offshore platforms located in the region require highly specialized design to balance the competing requirements of the harsh demands from earthquake, wave and ice loading. Structural design strategies that have been used in the region to protect structures from earthquakes include a tuned mass damper to protect the drilling derrick on the Orlan platform and Friction Pendulum seismic isolation on the Lunskoye-A and Piltun-Astokhskoye-B platforms.

Implications of strong earthquake ground motion duration on the response and testing of seismic isolation systems

“…Constantinou et al 17 and Fenz et al 33 have used the history of temperature at the sliding interfaces of isolators to determine testing protocols for sliding isolators in specific applications of very large isolators operating in extreme environmental conditions. Fenz et al 33 also expanded the procedure to determine test conditions that comply with a specified probability of non‐performance achieved with a particular level of confidence. However, the approach is complex and does not always lead to a single simple test as often desired.…”

“…While the cumulative energy has been used in the evaluation of general testing criteria of isolators in codes and standards, alternative procedures have been proposed and used to design experiments in applications of very large isolators used in offshore platforms under extreme conditions of operation. 17,33 These isolators were of the SFP type 27 and could not be dynamically tested anywhere as they carried loads in excess of 100MN. Principles of scaling and similarity were used to design reduced size isolators for dynamic testing.…”

Characterizing friction in sliding isolation bearings

“…Based on recent applications [17][18][19][20][21], the static axial pressure on sliders of FP (and TFP) bearings is much less than 100 MPa and is not expected to exceed 100 MPa under earthquake loading. The parameters of (2) are therefore fit to the experimental data for axial pressures less than 100 MPa.…”

Assessment of Scale Effects in the Experimental Evaluation of the Coefficient of Friction of Sliding Isolators

“…Constantinou and co-workers developed a scaling methodology based on calculating the temperature history at the sliding surface of the production isolator subjected to bidirectional motion and replicating it at the sliding surface of the scaled isolator under an equivalent unidirectional protocol [Constantinou et al, 2007]. Another example for designing qualification testing of a large isolator used in offshore structures carrying load of about 10000 tons was presented by Fenz [Fenz et al, 2011]. However, in both cases, the influence of temperature on friction was unknown, and the coefficient of friction was assumed as temperature-independent, which is not realistic.…”