Optimization of the dynamic behavior of a solar distillation cell by Model Predictive Control

The solar distillation cell is mainly composed by two parallel plates, one heated by sun radiation on which a water film falls, the other one maintained at a lower temperature and used for condensing water vapor. The equations describing the process behavior are nonlinear and coupled partial differential equations for mass, energy and momentum balances, with initial and boundary conditions. Solved by the finite volume method, they provide the temperature, concentration and velocity fields in the gas phase, in the liquid film and the temperature field in the heated plate. This model originally built in steady state to study the influence of operating parameters such as the heating flux, the feed temperature and flow rate, the temperature of the condensation plate and the form factor of the cell, on the flow rate of distilled water, has been extended to describe the fully nonlinear transient behavior. A reduced dynamic model, only characterized by inputs and outputs, has been identified for control and then used to dynamically optimize the operating parameters of the cell behavior by means of Model Predictive Control in order to maximize the mean flow rate of distilled water confronted to a variable solar radiation.

► New dynamic model based and solved from an existing steady-state model.
► Reduction of the full model to four continuous transfer functions.
► Dynamic optimization replaced by a Model Predictive Control method.
► Control valid both in open and closed loop.
► Gives maximum distillate flow rate even with disturbances such as solar heat flux.