Search

This paper presents a new transport model for reverse osmosis (RO) systems, which combines irreversible thermodynamics, together with solution-diffusion theory. The simplifications adopted by the classical theory for solution-diffusion mechanisms have been found to be quite lacking when it comes to predicting the separation of multicomponent mixtures. The presented model accounts for multicomponent computations through the application of thermodynamic property models, as a means to predict the various interactions amongst the species that are present in solution. The developed transport model is relatively easy to implement, and can be utilized alongside existing equipment and thermodynamic property models. The applicability of the model presented in this paper has been tested on three different case studies, including a case that investigates single component behavior and a case that investigates multicomponent behavior The proposed model shows very good agreement with experimental results.

This paper utilizes a stochastic optimization approach using genetic algorithms, for conducting rigorous pipe size sensitivity assessments onto the design of pressure relief networks. By sampling high performance candidates, only the finest options can survive. The pressure relief network system that was investigated in this work was previously reported in literature. The problem is constrained and involves minimizing a cost objective function that evaluates the overall network performance, in which the best pipe size combination should be selected for each segment within the network. The overall goal of this paper was to seek cost-effective designs for the pressure relief piping system by exploring different ranges of pipe diameters that are available for each segment in the network and comparing how the overall design of the system is affected, when the number of pipe size options to select from is varied.