(351e) Bifurcation Analysis of Wastewater Treatment Operation

Bifurcation Analysis of Stickney Water Reclamation Plant

Wastewater treatment plants (WWTPs) are an integral part of
modern day communities, ensuring a clean environment and providing clean water.
For almost a century, several technical enhancements to WWTPs have been made to
improve the treatment process. However, as with many industrial processes, the
necessity of sustainable operation has emerged in the last decade, as a result
of the embraced sustainability policies throughout the world.

A major portion of the energy consumption in WWTPs is
attributed to the need for forced aeration in the activated sludge process. Aeration
is critical for maintaining biomass activity, increasing the dissolved oxygen
concentration in the wastewater, and ensuring proper mixing in the tank. Most
WWTPs operate at unnecessarily high aeration conditions to ensure adherence to
specification limits. It should be obvious that a better understanding of the complex
behavior of the process would naturally lead to optimized and more
cost-effective operation.

Mathematical modeling offers deep insight into the workings
of a process, especially when coupled with a bifurcation analysis approach, as
we present in this study. A thorough bifurcation analysis leads to the
classification and quantification of the entire operating space, thus allowing
for a global view of operational constraints and risks. It can also serve as a
decision tool for designing novel control strategies, such as an "agent-based
control framework".

In this study, we focus on Stickney Water Reclamation Plant
(WRP) in West Chicago, which is the world's largest WWTP. We use a recently
developed model of the Stickney WRP, which was validated with historical data
from the plant from 2001 to 2009. The process is modeled as a 1-D continuous
reactor (discretized as a train of 48 CSTR tanks) followed by a settler/clarifier
with return activated solids recycling. The kinetic model is based on the widely
used Activated Sludge Model no. 1 (ASM1). A custom continuation algorithm was
used to carry out the bifurcation analysis on the resulting 480 algebraic
(originally ordinary differential) equations. Attention was given to effluent
concentrations of ammonia, dissolved oxygen and the biomass in evaluating the
analysis results. This is the first instance of a bifurcation analysis of such
a comprehensive model for the activated sludge process. Previous studies were limited
to simplified models that were not validated against existing installations. When
the total wastewater flow rate into the plant was used as a bifurcation
parameter, multiple steady states can be broadly located, even in the vicinity
of the nominal flow rate used by the process. The steady state set consists of
a steady state where the coexistence of two biomass species (autotrophic and
heterotrophic) is possible, states where only heterotrophs or autotrophs can
exist,  and
a washout steady state at which no biomass exists. This finding has important
implications on the operation of the WWTP. Similar observations are made when
ammonia loading and water temperature are considered as bifurcation parameters.

We also used the bifurcation approach to analyze the behavior
of the plant during major storm events that result in sustained increased
inflow owing to the nature of the combined flow situation into the sewer
network of Chicago and most major cities. The results show
that permit violation may occur during the long transients of storm
events. Possible remedial actions are proposed.

One chief aim of this study was the assessment of air usage
at different conditions; hence a bifurcation analysis on airflow was performed
indirectly by varying the mass transfer coefficient of dissolved oxygen. Results
show that it is possible to reduce the airflow by about 40-60% of the nominal
value, without causing an increase in ammonia or a decrease in dissolved
oxygen. This represents significant energy savings for this process. Therefore,
it is possible to achieve the sustainability goals of WWTPs, should proper
control strategies are implemented. We are currently working on encapsulating
the knowledge gained from the bifurcation study into an Agent-Based system
(ABS) for monitoring and control of the cyber-physical system resulting from
the combination of the treatment process, a sensor network, and the ABS.