Abstract
When a small crack is detected in a pressure vessel or piping, we can estimate the fatigue life of the vessel or piping by applying the classical law of fracture mechanics for crack growth if we are certain that the crack growth exponent is correct and the crack geometry is a simple plane. Unfortunately, for an ageing vessel or piping, the degradation will, in practice, change not only the crack growth exponent but the crack shape from a simple plane to a zig-zag pattern. To validate the crack growth exponent for an ageing vessel or piping, we present the design of an Intelligent PYTHON (IP) code to convert the information of the growing crack geometry measured by monitoring a small crack that was initially detected and subsequently continuously monitored over a period of time such that the IP-based analysis code will use the realistic zig-zag crack geometry as a series of re-meshed finite-element meshes for finding the correct crack growth exponent. Using a numerical example, we show that such an IP-assisted continuous monitoring program, using PYTHON as the management tool, TRUEGRID as the topological crack meshing tool, and two finite-element analysis codes for verifiable stress analysis, is feasible for predicting more accurately the fatigue life of a cracked vessel or piping because the material model has a field-validated crack growth exponent. Significance and limitations of this IP-assisted approach are discussed.
Universal fluctuation of polygonal crack geometry in solidified lava
