Study of the design of a novel inherently safe low-pressure water reactor

Abstract

Current nuclear reactors used for commercial electrical power generation are mainly watercooled reactors with outputs of ~1000MW(e) that operate at high pressures. Postulated accidents in such plants, due for example to a leakage of primary coolant, or failure of core cooling systems, can result in core meltdown with potentially catastrophic environmental consequences. To address safety concerns, advanced reactor designs have been developed that have a reduced the risk of core meltdown and improved resistance to hazards such as earthquakes and aircraft impact. However, the increased complexity of these designs has resulted in escalating construction costs and schedule delays in current projects, which has deterred further investment in Western countries. To improve affordability of nuclear power generation, several smaller reactor designs have therefore been proposed but these are often scaled-down versions of existing larger reactors. It is unclear that the problem of escalating design complexity due to safety demands will not result in similarly high and uncertain investment costs. This thesis describes a study of the safety and performance characteristics of the LPWR, which is a novel low-pressure water-cooled underground reactor design that has been developed at Imperial College London in this thesis study, as an alternative to current power reactor technologies. The LPWR is conceived as an inherently-safe natural circulation reactor that, though achieving a significant power output of around 300MW(e), retains a simple design with minimal dependence on complex safety features. The design aim is to achieve reduced costs of construction, operation and maintenance at the same time as a significantly reduced environmental risk due to reactor accidents. This thesis describes the LPWR design and presents an initial thermal-hydraulic analysis of the plant performance that shows that a significant electrical output is potentially achievable, despite the plant having a relatively low thermodynamic efficiency and using gravity-driven natural circulation rather than conventionally electrically driven circulating pumps to provide core flow: the low operating pressure is found to be beneficial in giving a margin to the critical heat flux limit (DNBR) that is much greater than in a conventional high-pressure reactor even though the fuel linear power densities are similar. A safety analysis of the LPWR is performed by modelling the transient behaviour of the reactor in bounding accident conditions. Results 2 | P a g eshow that there would be a large margin to core dry-out and fuel overheating in any fault of hazard condition that would need to be included in the design basis (i.e. sequences with a frequency of occurrence above one in one million years). This contrasts favourably with the safety performance of current high-pressure reactor designs where core dry-out and fuel failures occur in several accident sequences within the plant design basis. An initial probabilistic safety analysis of the LPWR indicates core melt frequency of ~10-8/reactor year, which is three orders of magnitude lower than that typically achieved by current high-pressure reactor designs. Such a low level of risk is negligible and may make the design more acceptable for wide deployment than current high-pressure reactor designs. An economic analysis shows that despite the increased fuel costs due to its reduced thermal efficiency, the potentially reduced construction and operating costs of an LPWR offer the potential for electricity costs that are lower than those for current large reactor designs and that are competitive with existing energy technologies.