A Fuzzy Support Vector Regression Model for Estimation of Collapse Moment of Wall-Thinned Pipes

The pipe bends and elbows are regarded as critical components in piping systems of nuclear power plants because they are incorporated into piping systems to allow modification of the isometric routing and more importantly pipe bends are usually incorporated to reduce anchor reaction forces. Also, the pipe bends and elbows are capable of absorbing considerably large thermal expansion and seismic movement through the energy dissipation as a result of local plastic deformation so that they maintain the integrity of piping system under transiently loading conditions. Significant care must be taken to avoid their collapse moment. Therefore, it is important to accurately assess the safety margin for a collapse of pipe bends and elbows under various operating conditions. The wall-thinned defect is mainly caused by flow accelerated corrosion, and it reduces failure pressure, load-carrying capacity, deformation ability, and fatigue resistance of pipe bends and elbows. Therefore, it is necessary to investigate the effect of wall-thinned defects on the failure behavior of pipe bends and elbows and to accurately estimate the collapse loads of wall thinned bends and elbows under various loading conditions