Analysis of the long-term performance of an on-site greywater treatment plant using novel statistical approaches

Abstract

This paper analyses the performance of three greywater treatment systems: RBC, MBR and stand-alone sand filter. Pollutants' concentrations in raw greywater exhibited a long-term increase. The RBC and MBR exhibited high removal efficiency, while the sand filter performance was much poorer. In order to apply control measures, dynamic regression SPC (DRSPC), SPC (statistical process control) with variable control limits, was developed. DRSPC often better described underlying long-term trends. For RBC effluent, DRSPC had the highest advantage for TKN. For MBR effluent, it had the highest advantage for turbidity and COD_t. For the stand-alone sand filter effluent, the DRSPC advantage was much lower. Constant density curves showed low negative correlation between inflow and outflow VSS, and no correlation between inflow and outflow BOD_t for the RBC. For the MBR, no slope was identified for VSS and low negative slope for BOD_t. The BOD_t of the sand filter effluent revealed positive correlation with inflow.

Keywords:

• greywater treatment
• MBR
• RBC
• sand filtration
• process control
• SPC