Automatic control applied to osmotic membrane processes in the water industry: A review

Abstract

The need to convert salt water into fresh water is clearly a matter of growing urgency. In this regard, osmotic processes can ensure a continuous supply of water suitable for human consumption. In this paper, osmotic membrane processes such as reverse osmosis (RO), forward osmosis (FO) and pressure-retarded osmosis (PRO) are addressed. Mathematical models based on differential equations of all the processes are described. In addition, different approaches of dynamic multi-input multi-output experimental models relating the most relevant RO variables are reviewed, as the FO and PRO processes do not yet have a model in this form. Predictive control techniques proved to be more capable of addressing disturbances and uncertainties than classical control (P, PI and PID). Nevertheless, PI (often implemented along with robust control) is one of the most widely used strategies. Latest techniques such as intelligent control (fuzzy-based control, neural networks and genetic algorithms) are also reviewed in the present work. This review concludes that efforts should be made to improve current control strategies, for example by incorporating expert controllers or artificial intelligence-based controllers.