the intelligence of the system as a whole; any failings in the implementation of APC
have been due to poor data quality from the instruments. This approach put more
intelligence into the control system to identify when an instrument becomes
unreliable and, for the system as a whole, to replace the unreliable data with an
inferred value based upon the readings being received from other instruments within
the system.
For example, the control model “knows” what each DO sensor should measure at
any given time, given the influent flow, blower load, valve positions, manifold
pressures and treated water quality. If any probes report values that are significantly
different from those that are expected, an alarm is raised, and the inferred value is
used to exercise control of the process. Optimized control can be maintained even
when real-time measurements become unreliable.
The multivariate process approach has advantages of being a system based upon
the control element and is much more widespread within the plant, taking into
account the whole treatment facility rather than just the activated sludge plant on
its own. Case studies of this approach in three UK water and sewage plants realized
savings between 20 and 35% of the aeration costs whilst also reducing the risk of
compliance failure as the treatment plant operates more efficiently under automated
control [30].
All of these systems are designed on the basis of the International Water
Association Activated Sludge Models (ASM), and it is these models that are
providing the fundamental basis for the control systems of the secondary treatment
plant stages. The challenge to the industry is to use these models at their core and
stretch them further into not only the treatment works level but across the whole
wastewater system. Although there are individual models for aspects of control
within the wastewater system, it is likely that the real complication in controlling
the entire system will be melding together the different models that are available.
These include hydraulic collection network models, multivariate process control
models for individual aspects of the treatment system and SCADA and control
systems. This does not take into account sporadic inputs into the wastewater system
such as from customers and of course weather.
4 A Smarter Wastewater Industry
Ubiquitous sensing will create many opportunities and threats for urban water
management and calls for a digital transformation [31]. Increasing amounts of data
is only of real business value if this valuable resource is ultimately used to inform
and support decision-making, i.e. data to information to insight to action. Smart
water network technologies have the potential to deliver an improved service to
customers and cost-effective performance improvements for the water industry. On
the wastewater side of the water industry, the “smart water” approach that we have
seen applied to applications in potable water (such as water leakage) is much more
difficult. The first barrier is the value of wastewater, as it is very much seen as
136
O. Grievson
have been due to poor data quality from the instruments. This approach put more
intelligence into the control system to identify when an instrument becomes
unreliable and, for the system as a whole, to replace the unreliable data with an
inferred value based upon the readings being received from other instruments within
the system.
For example, the control model “knows” what each DO sensor should measure at
any given time, given the influent flow, blower load, valve positions, manifold
pressures and treated water quality. If any probes report values that are significantly
different from those that are expected, an alarm is raised, and the inferred value is
used to exercise control of the process. Optimized control can be maintained even
when real-time measurements become unreliable.
The multivariate process approach has advantages of being a system based upon
the control element and is much more widespread within the plant, taking into
account the whole treatment facility rather than just the activated sludge plant on
its own. Case studies of this approach in three UK water and sewage plants realized
savings between 20 and 35% of the aeration costs whilst also reducing the risk of
compliance failure as the treatment plant operates more efficiently under automated
control [30].
All of these systems are designed on the basis of the International Water
Association Activated Sludge Models (ASM), and it is these models that are
providing the fundamental basis for the control systems of the secondary treatment
plant stages. The challenge to the industry is to use these models at their core and
stretch them further into not only the treatment works level but across the whole
wastewater system. Although there are individual models for aspects of control
within the wastewater system, it is likely that the real complication in controlling
the entire system will be melding together the different models that are available.
These include hydraulic collection network models, multivariate process control
models for individual aspects of the treatment system and SCADA and control
systems. This does not take into account sporadic inputs into the wastewater system
such as from customers and of course weather.
4 A Smarter Wastewater Industry
Ubiquitous sensing will create many opportunities and threats for urban water
management and calls for a digital transformation [31]. Increasing amounts of data
is only of real business value if this valuable resource is ultimately used to inform
and support decision-making, i.e. data to information to insight to action. Smart
water network technologies have the potential to deliver an improved service to
customers and cost-effective performance improvements for the water industry. On
the wastewater side of the water industry, the “smart water” approach that we have
seen applied to applications in potable water (such as water leakage) is much more
difficult. The first barrier is the value of wastewater, as it is very much seen as
136
O. Grievson