Optimal design of nuclear desalination multiple effect evaporation process plant

Full text: The interest in the use of nuclear reactors as an energy source for the desalination of seawater started in the Nineteen Sixties. The urgent necessity to provide potable water has evoked a number of reports. The International Atomic Energy Agency (IAEA) has shown intense interest in nuclear desalination since 1989. In particular, water scarcity has become a reality in the arid and semi-arid parts of the northeastern region of Brazil, and its variability from season to season, makes much of water supply unavailable when it is most needed. This is the reason why nuclear desalination may be an alternative for fresh water supply and power generation in this region of the country. There exist three concepts of nuclear desalination having technico-economic feasibility; i.e., multiple-effect evaporation (MEE), reverse osmosis (RO), and multiple-stage flashing (MSF). There is extensive literature, both academic and industrial, covering diverse aspects of the aforementioned concepts. However, it appears that the question of the optimal design of a nuclear desalination process plant has not been addressed so far. It is the objective of the work reported in this paper to start to fill this gap. The focus of the present work is on the optimal design of a nuclear desalination process plant consisting of a pressurised water reactor (PWR) and a multi-effect evaporator (MEE). In particular, the PWR-MEE conventional coupling scheme is considered, whereby vapour extracted from one (on more) turbine stage is used to preheat seawater. The optimal design problem is formulated and solved as a mathematical programming model, in a somewhat similar fashion to work reported for a combined cycle coupled to a MSF system. Results obtained so far are encouraging, although vapour consumption rates are 2-3 times higher than those suggested by EURODESAL work. Work is in progress to ascertain that operating conditions are identical to those of EURODESAL with a view to validating the mathematical programming model developed in the present work. (author)