A simplified yet comprehensive fatigue design procedure for tubular joints of offshore structures is developed. The allowable hot spot stress range for the fatigue design wave is derived as a function of a parameter which defines the structural response. A procedure for calibrating the structural parameter to detailed fatigue analysis results is illustrated.
For typical offshore structures that are not dynamically sensitive, i.e., structures in relatively shallow waters with low natural periods, it is often unnecessary to perform a detailed fatigue analysis of the tubular joints. These joints can normally be checked for fatigue by limiting the stress levels due to the design wave. For typical Gulf of Mexico template structures with natural periods less than three seconds, the fatigue design provisions of API RP2A1 specify an allowable peak hot spot stress of 60 ksi. This recommendation does not account for differences due to water depth, type of joint, location of joint in structure, choice of design wave height, etc. A more discriminating procedure is thus desirable.
An API sponsored research project performed by Professor Paul Wirsching2 recommended a fatigue design procedure based on a long-term Weibull distribution of stress range. The allowable stress levels from this reliability-based procedure were derived in terms of a single parameter (s) which describes the long term stress range distribution. Marshall and Luyties3 recommended preliminary values for ~ based on calibration to detailed fatigue analysis. However, the procedure has not been adopted in RP2A due to insufficient calibration. A difficulty with calibrating this procedure is that the single parameter ? must account for both structural response variations (joint type, location of joint in structure, natural period of structure) and the environmental loading parameters representing the effects of multiple wave populations such as hurricane waves and day-to-day waves. Calibrated values of ~ are thus restricted to a particular description of the wave climate, thus limiting the general applicability of the calibrated results.
The API Task Group on Fatigue Criteria was formed in 1983 to review the fatigue provisions in RP2A. Developing a more discriminating fatigue design procedure was one of the main objectives of the group. The writers proposed a fatigue design procedure to the group as described in this paper. The basic premise of this procedure is to separate the effects of environmental loading from those due to structural response. The environmental parameters are treated separately as described herein, and the calibration process involves only a structural response parameter. This procedure is presently being calibrated to the results of detailed fatigue analyses of various platforms by API task group members. Since the calibrated results are independent of the environmental parameters, allowable stress ranges for other environmental conditions can easily be derived.
The procedure described in this paper is based on a closed form expression for fatigue damage developed by Nolte and Hansford4 This fatigue expression assumes that the S-N curve relating number of cycles to failure (N) to the stress range (8) can be expressed as where m and K are constants.
