Improving knowledge about permeability in membrane bioreactors through sensitivity analysis using artificial neural networks.

Membrane bioreactor (MBR) has been widely employed for industrial effluent treatment, as its higher efficiency in removing pollutants makes effluent reuse more feasible. However, membrane fouling remains as a limiting factor for its greater diffusion. This work performed a sensitivity analysis study to investigate the effects of analytical and operating variables on membrane permeability. The case study is a MBR treating oil refinery effluents. After the identification and validation of a predictive neural model for permeability, sensitivity analysis methods based on both connection weights and variable disturbances were used to quantify and rank the variables influence. A comprehensive analysis showed that Suspended solids and Days between cleanings exerted greater effects on permeability, whereas sludge filterability and sludge temperature were less significant. In sequence, a specific analysis revealed distinct dynamics in MBR operation given different solids concentrations. For instance, from higher solids concentrations, among all the evaluated parameters, only COD presented low significance to the permeability. This evidence suggests that permeability is more susceptible to variations when operating with higher concentrations of Suspended solids. The global result of this study contributes to more efficient MBR operations since distinct relations with permeability imply different effects on membrane fouling.