Grit removal typically does not have any measurement and control system and so
from this aspect can be ignored.
The flow control and storm system within preliminary treatment is an area that
has a vast potential for intelligence in operation. It is going to be an important area of
development in England and Wales between 2020 and 2025 [22] and is all about
controlling the flow to full treatment (FFT). There is a legal duty under a treatment
work’s environmental permit for it to treat a certain flow before excess flows pass to
storm storage tanks. If excess flows are seen at the works for greater than 2 h, they
pass directly into the environment. Across the various treatment works in the
industry, for various reasons, there are problems with this concept. FFT flow control
is normally achieved by using either a static weir and flow control device (such as a
flume or hydrobrake) or a modulating device complete with flow measurement (such
as a modulating penstock). The modulating methodology can, on occasion, have
problems due to poor engineering practices.
As a result of this, there are moves within the English and Welsh wastewater
industry to install flow measurement and sensing on storm splits as well as on the
effluent point of the WwTW storm storage tanks. The retrofitting of flow to full
treatment flow measurement for flow control is disproportionately expensive. However, it does have further benefits for advanced control of the wastewater treatment
system.
Measurement of the FFT at a treatment works allows for the performance of the
wastewater collection system along with the potential for control of any terminal
pumping station. At the most basic, any large changes in the amount of flow over a
defined period of time can indicate whether there is a blockage within the wastewater
system. At its simplest, if a WTW doesn’t see flow over a period of time, then this is
a good indication that a blockage within the network is causing the flow that is
normally seen to escape elsewhere (normally through a CSO). It will vary from
system to system depending upon how many pumped flows and how many gravity
flows feed the treatment works, but a simple algorithm of a defined low flow rate
over a defined time will allow for anomaly detection.
In a more complicated system, there is potential to automate the pumped feeds
into the treatment works by using the FFT flow meter and automated control on the
terminal pumping stations. The latter is achieved using a priority based-system
related to the capacity within each wet well, and possibly artificial intelligence to
manage the contents of the wet wells, to smooth flows entering the treatment works.
This looks at the capacity of the network and the performance of the inlet works to
see what flows (and loads) can be received at the WTW whilst also minimising the
risk to both the environment and the customer. The current growing “big data”
context provides abundant opportunities for the application of artificial intelligence
to such problems as optimisation of sewerage networks’ operation, predicting urban
flooding [23], CSO monitoring and analysis [24, 25] and automatic control of sewer
pumping stations utilising fuzzy logic [26]. Latest approaches have proposed more
distributed real-time control of urban drainage systems to locally manage flooding
and overflow [27, 28]. Artificial neural networks (ANNs) have become an increasingly popular data-driven approach for water industry applications. ANNs have been
128
O. Grievson
from this aspect can be ignored.
The flow control and storm system within preliminary treatment is an area that
has a vast potential for intelligence in operation. It is going to be an important area of
development in England and Wales between 2020 and 2025 [22] and is all about
controlling the flow to full treatment (FFT). There is a legal duty under a treatment
work’s environmental permit for it to treat a certain flow before excess flows pass to
storm storage tanks. If excess flows are seen at the works for greater than 2 h, they
pass directly into the environment. Across the various treatment works in the
industry, for various reasons, there are problems with this concept. FFT flow control
is normally achieved by using either a static weir and flow control device (such as a
flume or hydrobrake) or a modulating device complete with flow measurement (such
as a modulating penstock). The modulating methodology can, on occasion, have
problems due to poor engineering practices.
As a result of this, there are moves within the English and Welsh wastewater
industry to install flow measurement and sensing on storm splits as well as on the
effluent point of the WwTW storm storage tanks. The retrofitting of flow to full
treatment flow measurement for flow control is disproportionately expensive. However, it does have further benefits for advanced control of the wastewater treatment
system.
Measurement of the FFT at a treatment works allows for the performance of the
wastewater collection system along with the potential for control of any terminal
pumping station. At the most basic, any large changes in the amount of flow over a
defined period of time can indicate whether there is a blockage within the wastewater
system. At its simplest, if a WTW doesn’t see flow over a period of time, then this is
a good indication that a blockage within the network is causing the flow that is
normally seen to escape elsewhere (normally through a CSO). It will vary from
system to system depending upon how many pumped flows and how many gravity
flows feed the treatment works, but a simple algorithm of a defined low flow rate
over a defined time will allow for anomaly detection.
In a more complicated system, there is potential to automate the pumped feeds
into the treatment works by using the FFT flow meter and automated control on the
terminal pumping stations. The latter is achieved using a priority based-system
related to the capacity within each wet well, and possibly artificial intelligence to
manage the contents of the wet wells, to smooth flows entering the treatment works.
This looks at the capacity of the network and the performance of the inlet works to
see what flows (and loads) can be received at the WTW whilst also minimising the
risk to both the environment and the customer. The current growing “big data”
context provides abundant opportunities for the application of artificial intelligence
to such problems as optimisation of sewerage networks’ operation, predicting urban
flooding [23], CSO monitoring and analysis [24, 25] and automatic control of sewer
pumping stations utilising fuzzy logic [26]. Latest approaches have proposed more
distributed real-time control of urban drainage systems to locally manage flooding
and overflow [27, 28]. Artificial neural networks (ANNs) have become an increasingly popular data-driven approach for water industry applications. ANNs have been
128
O. Grievson