Modeling, simulation and PI gain-scheduling controller optimization of water desalination plant with liquid jet vacuum ejector

Abstract

This paper gives a novel approach to modeling, simulation and control of a water desalination plant. The process in desalination plant is based on a vapour compression acquired by a liquid jet vacuum ejector. Optimal plant design for a nominal operating point is obtained using the combination of the multi-objective optimization with the multi-criteria decision making. Furthermore, the dynamic steady-state model which includes a liquid jet vacuum ejector, pump, valve and pressure vessel is developed. Moreover, for the different values of inlet temperature and water mass flow rate (disturbances) the optimal regimes of system actuators are found by the antlion algorithm, with respect to the profit maximization. Also, the control-oriented model of the system has been developed and linearized around different operating points to design a gain-scheduling nonlinear proportional integral controller. Additionally, it is demonstrated that the proposed controller is capable of guiding the system through the optimal states under the influence of disturbances. A map of the controller gains is determined by extensive simulation minimizing the appropriate cost functions. High-fidelity model has been numerically simulated in order to demonstrate that the developed method provides the desired design specifications while minimizing operational costs.