384268 Optimization of Direct Contact Membrane Distillation Based on Regressive Rotation Design
Membrane distillation (MD) is a promising separation technology for desalination. It is of significance for the practical application of MD to improve process efficiency and to reduce energy consumption through process design and parameter optimization. From literature, the study on the modelling and optimization of MD process considering both permeate flux and thermal efficiency integrating both operating conditions and module dimension parameters is rare. The purpose of this study is to try to search for both optimum operating conditions and module dimension parameters as well as to know their interaction effect on MD performance to achieve high permeate flux, large water production and low energy cost. The statistical experimental design and response surface methodology were applied for the modeling and optimization of direct contact membrane distillation (DCMD) process. The method is useful to understand the interaction effects between factors and to reduce the total number of experimental runs.
In DCMD process, polyvinylidene fluoride (PVDF) hollow fiber membranes were employed to desalting 3.5wt% sodium chloride aqueous solution and all of salt rejection in this study is above 99.9%. The operating conditions including feed inlet temperature, permeate inlet temperature and velocity of feed and the module configuration parameters including module packing density and length-diameter ratio were taken into account. The average permeate flux , water productivity per unit volume of module, water production per unit exergy loss and an comprehensive index (CI) of them were taken as optimization objectives and their regression models were developed.
It is concluded that the optimal set of operating conditions and membrane module parameters has been determined and the significant increase of MD performance was achieved. The coupling and interaction effect between the operating conditions and the module parameters were significant on the performance of DCMD. The models for the prediction of DCMD performance were established. Good agreement was observed between the simulated and the experimental results.