Estimation of the Foundation Condition of a Fixed Platform by Measurement of Dynamic Response

Abstract

ABSTRACT We analyzed the recorded dynamic response of the Monopod platform, in Cook Inlet, and simulated the response with computer models in an attempt to discover the reason for observed changes in the fundamental frequency. Using a 3-dimensional linear dynamic structural analysis program we-modeled the recorded response to ice impacts and simulated model response to an assumed earthquake. We matched the results of the dynamic program with a non-linear finite-difference soil pile analysis program to model the soil structure interaction. By varying model parameters we found that for the Monopod, dynamic response was insensitive to-changes in deck load mass or virtual water mass, but was strongly influenced by soil stiffness. Other investigators have found mass and hydrodynamic effects more important for other structures. By comparing numerical model output with recorded structural behavior, we were able to estimate the condition of the soil and the pile foundation. This analysis revealed that the change in the fundamental vibration period was related to changes in the soil around the piles near the mudline. We conclude that, for small deflections, the upper 20 feet of the oil does not contribute to the dynamic response Of the Soil-structure system. For the larger motions, associated with an assumed earthquake, this zone would contribute to dynamic response. Monopod Platform Has Acceleration and Strain Recording System When the Monopod platform (FIGURES 1 and 2) was installed in 1966, a strain gage system was installed to indirectly measure ice forces (FIGURE 3). With this system, structural vibrations were also observed. An accelerometer system and recorder were installed in 1972. as described by Durning and Engle, 1976. The system consists of three triaxial accelerometers, two on the production deck and one at the mudline in an unused well conductor. The locations of the accelerometers are shown in FIGURE 4. Data from tile accelerometers and from the previously installed strain gages are recorded on cassette tape decks located in the production office on the platform. The recording system remains inactive until it is activated by a seismic starter located on the production deck. Acceleration and strain data are recorded automatically, and the system shuts off ten seconds after the vibrations cease. The analog tape records were later digitized, with an 0.02 second sampling interval, for computer analysis, such as calculation of spectra. Change in Vibration Period The platform's fundamental period was determined to be 1.1 seconds from the strain gage measurements of 1967. After the accelerometer system was installed in 1972, the fundamental period was 1.3 seconds. This 18% increase in the fundamental period caused concern. In 1974, we undertook research to investigate and verify the available dynamic and nonlinear analysis methods, and to determine the cause of the observed change in the Monopod's fundamental vibration period.