predicted to be necessary in order to obtain adequate temperature control
[146].
In these mathematical models most attention has been paid to the heat
transfer problem. Similar approaches need to be developed for the water
balance to ensure that substrate beds do not dry out to levels which will
decrease bioreactor performance.
These quantitative methods are potentially powerful tools in guiding the
design and operation of large-scale bioreactors. They can be used in a rapid
evaluation of designs before they are built, identifying those with the greatest
potential to overcome overheating problems, and eliminating ideas which seem
promising but in fact have no or poor potential. Use of these methods is more cost
effective than relying solely on experimental data from large-scale bioreactors
built on the basis of best guesses, although of course the most fruitful approach
will be to run simultaneous modeling and experimental programs. Unfortunately,
many of the mass and heat transfer coefficients are poorly characterized for SSF
bioreactors. Rather, they are often estimated from the non-SSF literature. There is
an urgent for experimental work, not only to determine these parameters for SSF
bioreactors, but also to test these theoretical approaches by applying them in the
development of real large-scale SSF bioreactors.
7
Measurement and Control Within SSF Bioreactors
The variables associated with bioreactor operation, whether SLF or SSF, can be
divided into two groups – state variables and operating variables. Operating
variables are those which can be varied directly through external control. State
variables represent the state of the system or some part of it. They can only be
controlled indirectly, through manipulation of the operating variables. State
variables can be further divided into those which can be directly measured online, for example temperatures, and those which cannot be measured on-line,
such as biomass concentrations. The basic goal of an on-line control system is
to take the values of some state variables measured on-line, and to use these
values to manipulate the values of the operating variables in order to influence
the values of various state variables (which may or may not be directly measurable) in such a way as to optimize growth and product formation.
The key objective of a control system for an SSF bioreactor is typically to
control the bed temperature and water content at values which will lead to
optimal growth and product formation [167]. The operating variables which
can be manipulated to achieve this in SSF bioreactors are, depending on the
design, the temperature, flow rate and humidity of of the inlet air, and variables
associated with the agitation system, such as the frequency and intensity of
agitation. It may also be possible to manipulate the temperature and flowrate of
the surrounding air or cooling water passing through jackets or heat transfer
plates. If water or aqueous nutrient or pH-correcting solutions are added to the
reactor the timing and quantity of these additions can be controlled.
The state variables which can be measured on-line and can therefore readily
be used in such control systems are temperatures, and O 2 and CO 2 concentra120
D.A. Mitchell et al.
[146].
In these mathematical models most attention has been paid to the heat
transfer problem. Similar approaches need to be developed for the water
balance to ensure that substrate beds do not dry out to levels which will
decrease bioreactor performance.
These quantitative methods are potentially powerful tools in guiding the
design and operation of large-scale bioreactors. They can be used in a rapid
evaluation of designs before they are built, identifying those with the greatest
potential to overcome overheating problems, and eliminating ideas which seem
promising but in fact have no or poor potential. Use of these methods is more cost
effective than relying solely on experimental data from large-scale bioreactors
built on the basis of best guesses, although of course the most fruitful approach
will be to run simultaneous modeling and experimental programs. Unfortunately,
many of the mass and heat transfer coefficients are poorly characterized for SSF
bioreactors. Rather, they are often estimated from the non-SSF literature. There is
an urgent for experimental work, not only to determine these parameters for SSF
bioreactors, but also to test these theoretical approaches by applying them in the
development of real large-scale SSF bioreactors.
7
Measurement and Control Within SSF Bioreactors
The variables associated with bioreactor operation, whether SLF or SSF, can be
divided into two groups – state variables and operating variables. Operating
variables are those which can be varied directly through external control. State
variables represent the state of the system or some part of it. They can only be
controlled indirectly, through manipulation of the operating variables. State
variables can be further divided into those which can be directly measured online, for example temperatures, and those which cannot be measured on-line,
such as biomass concentrations. The basic goal of an on-line control system is
to take the values of some state variables measured on-line, and to use these
values to manipulate the values of the operating variables in order to influence
the values of various state variables (which may or may not be directly measurable) in such a way as to optimize growth and product formation.
The key objective of a control system for an SSF bioreactor is typically to
control the bed temperature and water content at values which will lead to
optimal growth and product formation [167]. The operating variables which
can be manipulated to achieve this in SSF bioreactors are, depending on the
design, the temperature, flow rate and humidity of of the inlet air, and variables
associated with the agitation system, such as the frequency and intensity of
agitation. It may also be possible to manipulate the temperature and flowrate of
the surrounding air or cooling water passing through jackets or heat transfer
plates. If water or aqueous nutrient or pH-correcting solutions are added to the
reactor the timing and quantity of these additions can be controlled.
The state variables which can be measured on-line and can therefore readily
be used in such control systems are temperatures, and O 2 and CO 2 concentra120
D.A. Mitchell et al.