Forecasting the effect of feast and famine conditions on biological sulphate reduction in an anaerobic inverse fluidized bed reactor using artificial neural networks

The longevity and robustness of bioreactors used for wastewater treatment is determined by the activity of the microorganisms under steady and transient loading conditions. Two identical continuously operated inverse fluidized bed bioreactors (IFB), IFB R1 and IFB R2, were tested for sulphate removal under the same operating conditions for 140 d (Periods I–IV). Later, IFB R1 was used as the control reactor (Period V), while IFB R2 was operated under feast (Period V-A) and famine (Period V-B) feeding conditions for 66 d. The sulphate removal efficiency was comparable in both IFB, <20% in Period I and ∼70% during Periods II, III and IV. The robustness of the IFB was evident when the sulphate removal efficiency remained comparable during the feast Period (67 ± 15%) applied to IFB R2 compared to continuous feeding Periods (Period IV (71 ± 4%) for IFB R2 and Period V (61 ± 15%) for IFB R1). The IFB performance was modelled using a three-layered artificial neural networks (ANN) model (5-11-3) and a sensitivity analysis, the sulphate removal was found to be dependent on the COD:sulphate ratio. Besides, the robustness, resilience and adaptation time of the IFB were affected by the degree of mixing and the hydraulic retention time.