Summary and conclusions
A simulation of a brewery CIP system was proposed using
reference nets extended with external java functions to investigate
the effect of an intelligent water quality management system, the
‘water switch’. The ‘water switch’ utilises a fuzzy logic model for
separating process water according to monitored water quality
parameters measured by a brewery CIP system. The change in
process water characteristics was described using a timedependent
equation integrated into the reference net and the
final energy availability in the process wastewater was approximated
using values suggested for a combined heat and power
system in an industrial sized facility.
The reference net modelling and simulation system provides a
visual representation of the flow of processes in the brewery CIP
system for analysis and modification. The Java plugin management
system allows for data extraction and integration of complex
modelling techniques outside the scope of a standard Petri net
model. Combining these techniques allowed for a series of experimental
simulations to be developed and the impact of changes to
the brewery CIP system quickly prototyped.
The continuous functions, such as the change in process water
over time, were simplified in this model to allow for easy data
manipulation within the reference net. However, increasingly
complicated continuous functions can be integrated into the
reference net through the plugin management system. The reference
net continues to model the batch processes while the Java
functions handle all continuous and complex processes. A visual
overview of the entire system is retained by the reference net
model and the complex processes are hidden underneath in object
nets and Java functions. This system also allows for the quick
integration of independent process descriptions modelled using
external programs through Java interfaces.
The integration of a database management system allows for
real data from a brewery CIP automation system to be simulated in
the reference net without requiring changes to the model. Optimisation
of the reference net and the online CIP system can be run
in parallel with results from one influencing changes in the other.
This simulation focused on the cleaning of the tanks required for
the boiling and fermentation stages of the beer brewing process.
However, a significant portion of the concentrated wastewater
produced by a typical brewery originates from the bottle cleaning
and filling processes. Further research could focus on implementing
the procedures outlined here on a plant wide scale, with the system
net in this study becoming another object net in a larger tree of
nets. The advantage of the reference net system is the potential for
easy integration of new processes based on the same formalism.
This study shows that the reference net formalism can combine
external models and databases into a brewery CIP system simulation.
The reference net can be used to provide quick, visual feedback
on the effects of implementing a new technology, such as the
‘water switch’, into an existing brewery CIP system. The methods
used in this study could be applied to target any similar industrial
process where water and energy usage are of importance.
Optimisation of water usage in a brewery clean-in-place system using
reference nets