Structural integrity analyses are used to guarantee the reliability of critical engineering components under certain conditions of interest. In general, the involved parameters have statistical distributions. Choosing a single set of values for the parameters of interest does not show the real statistical distribution of the output parameters. In particular, offshore pipes installation by reeling is a matter of concern due to the severe conditions of the process. Since it is necessary to guarantee the integrity of the pipes, a probabilistic fracture mechanics reliability analysis seems to be the most adequate approach. In this work, a probabilistic fracture mechanics assessment approach to perform the structural reliability analysis of tubes subjected to a reeling process was developed. This procedure takes into account the statistical distributions of the material properties and pipe geometry, using a fracture mechanics approach and the Monte Carlo method. Two-parameter Weibull distributions were used to model the variability of the input parameters. The assessment procedure was implemented as a self-contained executable program. The program outputs are: the statistical distribution of critical crack size, amount of crack extension, final crack size and the cumulative probability of failure for a given crack size. A particular case of interest was studied; a seamless tube - OD 323.9 × wt 14.3 mm, was analyzed. Tolerable defect size limits (defect depth vs. defect length curves) for different probability of failure levels were obtained. A sensitivity analysis was performed; the effect of material fracture toughness and misalignment was studied.
Robust H∞ Deep Neural Network-based Filter Design of Nonlinear Stochastic Signal Systems
