Abstract
•Nonlinear dynamic model for membrane distillation is developed and validated.•Different model structures, formulations, and complexity are compared.•The spatial model structure outperformed the lumped structure.•Model considering membrane heat transfer mechanism found superior to others.•Nonlinearity is essential for best model predictions.
In this study, a nonlinear dynamic model is developed to capture the transient response of the outlet temperatures of a direct contact membrane distillation (DCMD). Different structures and options of the dynamic model are also proposed. The effectiveness of the model is validated against experimental data generated from a pilot plant. Incorporation of nonlinearity in the DCMD model enhanced its prediction over that obtained by the standard transfer functions and/or the linearized model. It is also observed that the performance of the spatial version of the model outperforms that of the lumped formulation. Judicious and minor modification of the model parameters expanded the capability of the nonlinear spatial model in matching the plant responses over a wide range. In addition, the results illuminated the effectiveness and role of the overall heat transfer coefficient in enhancing the model accuracy and predictiveness over a wide range of operating conditions.