comparing two start-up strategies for MBRs: experimental study and mathematical modelling

Abstract

The performance of a membrane bioreactor (MBR), and mechanisms of fouling formation, may differ due to the start-up. Therefore, the start-up can constitute an aspect that critically influences MBR performance during its lifespan. Indeed, the start-up can influence the mechanisms of membrane fouling, which is of paramount importance in an MBR. In order to gain insights on the effects of the start-up, both experimental and mathematical modelling studies were carried out on an MBR pilot plant. The MBR pilot plant constituted of a hollow fibre membrane module, in a submerged configuration, was fed by real wastewater. Two experimental periods were carried out, lasting 65 days each, characterised by two different MBR start-ups: without inoculum of activated sludge, and with inoculum of activated sludge. An integrated MBR mathematical model was applied to the two periods in order to combine the insights gained from the experimental survey. The MBR model simulates both biological and physical processes simultaneously, taking into account the interactions between them. Two MBR model calibrations were carried out for the two different experimental start-ups. Model calibration was carried out by means of an innovative calibration protocol, based on the global sensitivity analysis, and the Generalised Likelihood Uncertainty Estimation. Thus, two sets of model parameters were assessed and compared. Both experimental and mathematical model results were analysed to quantify the effects of the two start-ups on membrane fouling. The results showed only slight differences in terms of total pollutant removal efficiency. However, relevant differences appeared with respect to the fouling phenomenon between the two start-ups. The results demonstrated that the best choice, in terms of the start-up of MBR systems, comes from a compromise between minimisation of the fouling issues and high system performance.