used to investigate each of these safety hazards are enumerated in the table's test description column. The relationship of each of these specific tests to postulated accident environments is discussed in the purpose column. The conditions column presents the values for the primary independent variable in each test. Finally, there is a description of the effect each independent variable value had on the fuel-containment capability of the RTG/RHU component under examination. An examination of Table II reveals some important trends in RTG/RHU component responses to safety testing. These trends include: 1. complete fuel containment under static overpressures characteristic of a shuttle launch vehicle explosion; 2. fuel containment in most face-on fragment collisions and fuel release in edge-on fragment collisions; 3. complete fuel containment in launch vehicle fires; and 4. fuel containment in surface impacts on sand and water; some fuel release in surface impacts on granite, concrete and steel. The DOE ESAR analyzed the accident scenarios such that a probabilistic risk assessment could be performed for each mission phase. References [3] and [4] present this risk assessment in detail for the RTGs and RHUs, respectively. The probability of fuel release for a given accident depends on the probability of the accident and its severity. For the expectation analysis the most probable releases occur in Phases 1 and 4, with values of 3.6 X 10~4 for each phase. The least probable releases occur in Phases 0 and 5, at 5 X 10-7 for each phase. For Phase 5, the probability of release is taken as the probabilty of the event occurring, while the probability of release is only about 1/lOth of the probability of an accident in Phases 1 and 4. For the other phases a release is expected to occur in less than one-one hundredth of the accidents. The expectation cases for each mission phase are used to quantify mission risk. The risk of an event is defined as the product of the probability of that event and its consequences. The risk from a mission phase considers all the accidents in that phase as represented by the expectation source term. Similarly, the mission risk is the sum of the risks of the phases. The expectation consequences for each phase result in less than one predicted cancer-related fatality within the potentially exposed population, in addition to those which would normally occur. Thus, it is reasonable to interpret the results as zero. Nevertheless, there would be radiological impact involved in any accident releasing fuel, and the product of probability and consequence (collective dose or fractional health effect) gives a measure of non-lethal relative risks of the individual phases. Phase 1 involves the highest risk of 4.3 X 10-7 primarily because of the relatively high accident probability. Phase 4 is the next highest risk at 9 X 10-8, again because of the high accident probability. Phase 5 is the third highest risk, at 6x 10“8 because of the higher release and collective dose, even though the probability is low. The lifetime average individual risk for each mission phase ranges from 5X10-12 to 9X10-9. Lifetime average individual risk from the normal incidence of cancer fatality is approximately 0.2. The NCRP 91 negligible individual risk level is 1.0 X 10-7 annually or approximately 1 X 10-6 over a lifetime. Thus we see that the individual risks from the Galileo expectation analysis fall in the NCRP negligible range. To add additional perspective, there are natural events, such as the impact of a large asteroid with the Earth, which have a probability of the order of 10-6 per year. This event would have much more severe consequences than those summarized in this paper.
RkJQdWJsaXNoZXIy MTU5NjU0Mg==