A dynamic optimization approach for a multi-effect desalination (MED) integrated with thermosolar storage system

Abstract

This work aims to propose a more robust and realistic dynamic optimization (DO) approach to maximize the profit of distilled water production in a multi-effect desalination (MED) plant integrated with thermosolar storage system. Unlike the literature, the underlying model comprehends the integrated system composed of solar collectors, thermal energy storage (TES) tanks and MED. To carry out this study, a non-linear phenomenological model representative of these conjugated systems was developed and validated with dynamic data from the literature, with maximum relative deviations below 5%. A sensitivity analysis identified irradiance, ambient temperature, and the water flow to the solar collectors and to the first MED’s evaporator as the main variables. These last two variables were used as decision variables in the optimization problem, which was solved using the orthogonal collocations on finite elements. This method proved to be stable and computationally efficient, with a simulation time of 54 s. The results demonstrate that the consideration of two decision variables, instead of just one in the objective function, provides a 5% gain in profitability and almost 24% in distillate production. Simulations over longer time horizon (48 h) confirm that the TES can supply the heat demand for desalination even after sunset, yet a receding horizon architecture is required to prevent tank’s drying or overflow. The optimal profiles obtained for the water flow to the solar collectors and to the MED’s first effect can be used for implementing control schemes.