An Automated Procedure For Platform Strength Assessment

Abstract

ABSTRACT Ultimate strength is one of the key parameters used in safety and economic assessments of platforms. The full ultimate strength (FUS) is typically determined from nonlinear push-over analysis. An alternate approach uses linear analysis to determine the simplified ultimate strength (SUS) which is a lower bound of FUS. This paper documents an automated analysis procedure developed to determine the SUS of platforms. By using the automated approach, platform safety assessments can be performed efficiently and economically. SUS results following thii procedure are presented for five platforms. These results are compared to FUS results and some field inspection data obtained after Hurricane Andrew. INTRODUCTION During Hurricane Andrew in August 1992, several platforms in the Gulf of Mexico (GOM), toppled or sustained significant damage. Recognizing that the Andrew experience presents a unique opportunity to better understand the performance of GOM structures in huge storms, Chevron launched an investigation of several of its platforms in the South Timbalier area. The South Timbalier area experienced waves with return periods in the order of 100 years that toppled three platforms and severely damaged several others. The overall goal of the investigation was to develop an understanding of why platforms failed and survive& which eventually may lead to the development of reliability models that more accurately predict failure probability. Full ultimate strength (FUS) of a platform is one of the necessary items in the evaluation of platform failure probability. FUS can be determined by performing a nonlinear push-over analysis. However, such nonlinear analyses often necessitate complex modeling techniques which require assumptions that may not be fully accepted by the industry. Moreover, the computer programs available to perform such analyses often require several simplifications in the modeling of loads as well as the nonlinear structural and foundation behavior. Hence, there exists the possibility that these simplifications can significantly affect the analysis results. Alternatively, a lower bound estimate of FUS is the simplified ultimate strength (SUS) which is determined based on platform components (members, joints, foundation, etc.) reaching their ultimate capacity in a linear analysis. SUS results can be used directly in the determination of an upper bound for the platform failure probability. In many applications this may suffice. One of the major attractions of the SUS approach is that it uses relatively simple and well established modeling and analysis procedures inherent to linear frame analysis. Even though the SUS approach has been used in the past a consistent analysis procedure does not exist. This paper addresses a consistent methodology for modeling and analysis of platforms to obtain their SUS. This procedure is computerized such that linear analyses necessary for platform assessments can be performed efficiently, consistently and economically. SIMPLIFIED ULTIMATE STRENGTH APPROACH Simplified ultimate strength analysis can be implemented in an efficient procedure which does not require sophisticated computational solutions. For this project it was developed within the framework of a conventional soil-structure-interaction analysis in which the jacket above mudline is linear and the soil model is nonlinear. Material non-linearities and large deflections were not considered.