• design optimization model
• MINLP
• multi-energy system
• offshore desalination
• reverse osmosis

Cogeneration of fresh water and electricity it’s a common process in MENA (Middle East and North Africa) countries. Water demand is constantly increasing, so new desalination plants must be built to guarantee drinking water to the population of the region.
In this study, we examine the possibility of building a new offshore desalination plant in the coastal area of Haifa, Israel, developing a mixed-integer non-linear programming (MINLP) model which allows us to investigate the energy cost, layout and environmental impact of such a plant. This work is a pre-feasibility study related to the design of an energy supply system for a reverse osmosis (RO) desalination plant. The aim is to come up with some different system layouts to be compared from a techno-economic point of view, underlining the pros and cons of every solution. In particular, the model gives as output the design size of power generation technology to be installed, to satisfy water demand. Another model is then implemented, just to analyze the operation features of the solutions given as output by the design model.
In the following, the structure of the thesis is briefly explained. First, the topic is introduced, then desalination technologies are depicted, with a focus on membrane-based processes and offshore solutions. Therefore, after contextualizing the problem, the programming optimization model is presented with the problem statement and model formulation. The same model is implemented in two different versions: the first does not have any constraints related to CO2 emissions (base case), while the second does. Afterward, the case study of our interest is depicted together with all the assumptions and input data related to it. In the next Section results of the optimization models are shown and commented on. Also, a sensitivity analysis is provided, to investigate the effect of input data variation on the model output. In particular, as the objective function of the optimization model is the total annualized cost (TAC), mostly economical parameters are subjected to a sensitivity analysis. Finally, conclusions and developments ahead are explained.
The base case model output is taken as a benchmark for a comparison with other solutions, as it is cheaper and simpler than the other ones, but also the solution already utilized for onshore applications in Israel. The other solutions show how, in the selected location, it is possible to reduce the environmental impact of the investment, mainly thanks to wind energy. The wind farm design size value can reach 90 MW, depending on boundary conditions, producing 38% of the total energy demand of the RO desalination plant. By the way, from an economic point of view, such a layout is not the best, as the total annualized cost is about 38% higher than the base case.