Modélisation et Simulation de Systèmes de Dessalement d’Eau de Mer par Osmose Inverse et Energies Renouvelables.

Abstract

The objective of this work is twofold. It aims first at proposing mathematical models which can describe the operation of a sea water desalination module by reverse osmosis. The calculation of the various flow rates and the required pressures for this operation allows to compute the specific energy consumption: the energy required to produce one cubic meter of drinking water. The second objective of this work was to study the possibility of substituting conventional electrical energy by a renewable energy such as wind energy. For that purpose, the meteorological data of some coastal regions are studied to determine their wind potential. In this work, two mathematical models are developed to describe the operation of reverse osmosis modules. The first mathematical model is simple and does not take into account the flow of the permeate and the retentate in the separation module. It is a system of non linear equations. The resolution of such a system does not present any particularityThe second mathematical model takes into account the flow of the permeate and the retentate inside the module. The two most used modes of circulations are studied: co-current and counter-current circulation modes. The obtained mathematical model is a set of ordinary differential equations. The resolution of the mathematical model developed for the counter-current flow pattern is subjected to the split boundary value problem. To solve this problem, a robust and efficient procedure based on orthogonal collocation on finite element method was developed. Two checkings of both developed mathematical models, with their variants (with and without polarization concentration phenomenon), were carried out. The first checking was done with a real case which is the sea water desalination plant of Bousfer, Oran, situated in the west of Algeria. The second checking was carried out with previous work using reverse osmosis in a closed loop (concentration system). The results obtained in both cases show a good adequacy between the values obtained by both developed mathematical models and real ones. The second part of this work was the assessment of wind resources of coastal areas for the study of a possible substitution of the electrical energy, produced by power plants, by electrical energy produced by wind turbines. The first step dealt with collecting meteorological data over a long period. The processing of these data is an important and tedious task. It involves the calculation of correlation matrices for specific data sets to determine the most corresponding data in order to calculate the lacking ones by linear regression. For low correlation coefficients, the method of interpolation by the cubic spline functions was used to perform this operation. Statistical analysis begins with the study of time variation of wind speeds in the monthly, annual and multiannual scale. It is followed by the determination of the parameters of Weibull distribution. For that purpose, two methods were used: linear and non-linear regression. The Weibull distribution parameters enabled the calculation of the available specific powers at the level of every studied site.