Probabilistic methods have been successfully used to analyze the risk associated with different existing nuclear power plant designs. Such tools can be also used at earlier stages in the design process. Defining a methodology for generation of the probabilistic lifetime reliability for a specific tokamak in-vessel component, the divertor, is the objective of this thesis. The divertor plates establish an interface between the plasma and the material surface of the tokamak device.The design of the divertor cooling system is a most demanding task since it endures the largest power density during operation. Even more severe consequences would appear under some transient conditions.In the present analysis, the divertor conditions for the technology phase of operation of the International Thermonuclear Experimental Reactor (ITER) as specified at the conceptual design stage are used as the reference design. The methodology developed to analyze the ITER divertor reliability consists of the following steps:1. description of normal operating conditions;2. thermal-hydraulic analysis for normal operation;3. identification and classification of transient events; 4. estimation of frequency of occurrence of transient events; 5. thermal analysis for transient event conditions; 6. defining the failure modes / failure criteria models; 7. assessment and propagation of uncertainties; 8. evaluation of the probability of avoiding failure of the divertor plate. The transient events considered in this work are grouped in two categories as follows:-transients that do not affect the divertor temperature distribution prior to shutdown, such as: auxiliary heating system disturbances, magnet system disturbances, main coolant disturbances, balance of plant disturbances, internal plasma disturbances; -transients that affect the divertor temperature distribution prior to shutdown, such as those on the coolant side: Loss of Coolant Accident (LOCA), Loss of Flow Accident (LOFA), and Loss of Heat Sink (LHS), and overpower (OP) transient on the plasma side. In applying this methodology, one failure mode is assumed to be the predominant failure mode: that of surface material loss due to sputtering, melting and evaporation. Taking no credit for redeposition of that material partially balances the fact that other failure modes are not accounted for. 2 The heat conduction computer code HEATING 7.2 is used for the steady-state and transient thermohydraulic analyses as well as for estimating the material loss during transients. The development of a reliability function requires defining a probability distribution function for the material loss during transients. A second order response surface (Response Surface Methodology) is derived for the material loss as a function of the uncertain parameters. The -uncertainties of the frequency of occurrence of transients and material loss parameters are propagated through the reliability function by a Monte Carlo simulation.Using the limited data available leads to the conclusion that there is a high prob...
