OPTIMIZATION OF A PERIODIC BIOLOGICAL PROCESS FOR NITROGEN REMOVAL FROM WASTE-WATER

The Bio-Denipho(R) process, a ''phased isolation ditch'' technology, v aries both aeration pattern and flow path in a continuous flow multi-r eactor system to force oscillation of organic and nutrient concentrati ons in process reactors. Using a six-phase cycle, desired biochemical transformations (e.g. nitrification, denitrification), are accomplishe d at different times in the same reactor. We used an industry-standard biokinetic model (IAWPRC) to develop and test three control strategie s of increasing sophistication: (a) fixed phase lengths, (b) use of co nstant set points to switch between phases thus resulting in variable phase lengths, and (c) use of switching set points which are a functio n of on-line measurements (criteria functions). These strategies were optimized for nominal diurnal operating conditions, with the objective of minimizing effluent soluble nitrogen, subject to an upper bound on the oxygen transfer coefficient. In addition, performance of the stra tegies was simulated against 1-, 2-, and 5-day sustained peak loads. U nder nominal diurnal conditions, the performance obtained with each st rategy was comparable. Under peak loading conditions, both of the swit ching set point strategies gave substantially better performance than the fixed phase length strategy. Criteria functions were marginally be tter than constant switch points.