This paper presents a method of applying the probabilistic fracture mechanics techniques to the selection of excellent acceptance criteria for CTOD test, mainly focus1ng on the required fracture toughness and the number of test specimens. Several factors which are stochastic in nature are treated probabilistically. The required fracture toughness and the number of specimens to be tested in acceptance testing are considered in terms of its risk level and fracture probability vs. probability of acceptance respectively.
The current acceptance criteria in general determine δ REQ using traditional deterministic fracture mechanics analysis. The deterministic analysis can not account for the reliability of δ REQ determined and in general selects the worst value which may be the unrealistic one. In dealing with problems such as these effectively a probabilistic approach is logical. The aim of this paper is to give a method of determining δ REQ and the appropriate number of specimens for CTOD test using probabilistic fracture mechan1cs techniques. The physical and mechan1cal concepts of fracture are primarily based on those of WES 2805. Reference structural components are the welded joints of marine structures subjected to sea wave induced stresses. such as ships and offshore structures. In carrying out the analysis stresses caused by sea waves, crack size and fracture toughness having a large scatter peculiarity for welds are taken as random variables which have their specific probabilistic distributions. The discussions are largely carried out based on numerical results.
δ Probabilistic Distribution of Applied Strain ε In WES 2805, ε is divided into three categories by its constituent elements namely ε caused by boundary force ε I. ε caused by welding residual stress ε 2. and t caused by welding joint profile responsible for localized strain concentration ε 3.