Influence of the design parameters on the stress state of saddle-supportedpipelines: an artificial neural network approach

A neural network methodology is herein applied to numerically treat the sen sitivity analysis problem of above-ground pipelines under static loading by taking into account the possibility of development of uplifting phenomena at the pipe-saddle interfaces. Assuming classical frictionless unilateral c ontact to mathematically describe the pipeline support conditions coupled b y an appropriate finite element scheme, the discrete problem is put in the form of an inequality constrained quadratic optimization problem with respe ct to either displacements or stresses. In order to investigate the structu ral response and the stress states of the above-ground pipeline at hand wit h respect to the variation of critical design parameters which are the pipe thickness and the support conditions, the sensitivity analysis problem is formulated as a quadratic programming problem with the design parameters ap pearing in the quadratic term. The feasibility of using appropriately desig ned neural networks to model the complicated nonlinear relationship between the several input parameters associated with above-ground pipelines and th eir support conditions is thus demonstrated. (C) 1999 Elsevier Science Ltd. All rights reserved.