Pressure Vessel Safety Assessment Of Material Properties And Defect Detection Uncertainty

AbstractSafe assessment criterion and technology for the flawed pressure vessel (PV) wasfirstly reviewed. Several problems starVing fOr solving in pressure vessel safetyassessment (PVSA) were indicated and the artificial intelligent method was discussedand introduced to solve uncertainty problems in PVSA. The risk analysis theory ofmodem manage science was creatively indoduced to PVSA in this thesis.The fuzzy clustering method was introduced to substituting mechanical properties ofmaterials in PVSA. According to chemistry ingredient and general mechanical propertiesof materials, the substiforional data of fractUre touglmess was decided by this method.The results by the hay clustering were identical to those determined by examinationand coincident to thOse results chosen by other principles and methods, hahermore,which dealt with more factors and decreased the influence of artificial factors.Based on fuzzy clustering method, the fhazy neural netWork (FNN) was introducedto identifying the mechanical properties of materials in PVSA, which resolved theproblem that the mechanical properties of materials was unable to be determinedaccurately in PVSA. On the basis of the known experimental data and the lmown dataused in engineering assessmeni, the forecastCd data of fracfore toughness was calculatCdafter those Anown data were classified by fhaZy clustering. The results calculated by thefuzzy neural netWork were identical to those determined by exaxnination and coincidentto those chosen by other principles and methods. With the advanages of fhazymathematics and wtificial neural network, the FN'N was aPplied to identifying themechanical properties of materials in PVSA, which determined more accuratelynecessny data for PVSA.The fuzzy set theory and the probability statistics method are combined andintroduced to NondestrUctive Testing (NDT) reliability analysis. According as fhazysynthesis judge method, a fuzzy synthesis judge is given to principal factors of NDTreliability and the values of uPPer and lower limits of NDT fuzzy section. On the basis ofthe fuzzy definition of NDT reliability and flaw detect probability, NDT fuzzy reliabilityand fuzzy POD was analyzed and calculated, which dealt logically with the fhazycharacteristic and randomicity and prepared for safety assessment and risk estimation onflawed structure.IIThe risk analysis theory of modem manage science was creatively introduced to PVSA, which consummated the system of PVSA and made monitoring on-line, defect assessment, invalidation risk estimation, equipment maintain and replacing become one flesh. The method of risk analysis and decision on the invalidation of pressure vessel & piping and its common process in engineering applications were indicated. Furthermore, the invalidation risk of 400m3 liquid hydrocarbon global tank of a rendering plant was analyzed and the result estimated by risk analysis was consistent to that by safety assessment.Finally, the entire research production was summarized. The next work required by a consummate and intellectualized system for PVSA was indicated.