Application of reliability analysis method to fusion component testing

The term reliability here implies that a component satisfies a set of performance criteria while under specified conditions of use over a specified period of time. For fusion nuclear technology, the reliability goal to be pursued is the development of a mean time between failures (MTBF) for a component which is longer than its lifetime goal. While the component lifetime is mainly determined by the fluence limitation (i.e., damage level) which leads to performance degradation or failure, the MTBF represents an arithmetic average life of all units in a population. One method of assessing the reliability goal involves determining component availability needs to meet the goal plant availability, defining a test-analyze-fix development program to improve component reliability, and quantifying both test times and the number of test articles that would be required to ensure that a specified target MTBF is met. Statistically, constant failure rates and exponential life distributions are assumed for analyses and blanket component development is used as an example. However, as data are collected the probability distribution of the parameter of interest can be updated in a Bayesian fashion. The nuclear component testing program will be structured such that reliability requirements for DEMO can be achieved. The program shall not exclude the practice of a good design (such as reducing the complexity of the system to the minimum essential for the required operation), the execution of high quality manufacturing and inspection processes, and the implication of quality assurance and control for component development. In fact, the assurance of a high quality testing/development program is essential so that there is no question left for reliability