The estimates of the reference temperature T 0 , obtained for the base metal and the weld-seam metal of the Cr - Ni - Mo - V type (shell 200 mm thick) on the basis of statistical modeling by the Monte Carlo method are presented. T 0 was determined according to the ASTM E1921 standard taking into account the inhomogeneity of the material. The sample size of the fracture toughness values KJC for T 0 modeling was 12, 24 and 70. The Monte Carlo method was used for analysis of the correctness of metal identification (homogeneous/inhomogeneous). It is shown that sampling of 12 samples do not provide a reliable determination whether the metal is homogeneous or inhomogeneous (incorrect results were obtained in 50% of cases for the base metal and in 37% of cases for the weld-seam metal). When the sample size increased to 24 samples, the incorrect results were obtained in 5% of cases. The T 0 values with allowance for the material inhomogeneity were determined by two ways: using a screening procedure and proceeding from the actual bimodal representation of the fracture toughness distribution (parameters of the bimodal distribution were determined by the maximum likelihood method). It is shown that both methods give close results for the base and weld-seam metal, the magnitude of the shift towards positive values in the average T 0 values determined with allowance for the inhomogeneity being about 22°C. Using the obtained T 0 estimates, the lower envelopes of the temperature curves of the fracture toughness are constructed (master curves for 5 % failure probability).
Read full abstract