How is Compass different to an automatic coagulation control algorithm based on raw water organics (UV254 or colour) and turbidity?
In drinking water treatment applications, Compass utilises the full power of UV-Vis spectroscopy and can characterise natural organic material (NOM) in real time, similar to a SUVA measurement. This enables the Compass system to truly optimise coagulant dosing depending upon the nature of the organics in the raw water (molecular size and hydrophobicity). Compass is managed through a separate controller which is connected to s::can hardware and was developed by Lutra in the early 2000’s and PMA represent both s::can and Lutra in the UK.
There are two versions of Compass available for drinking water treatment.
- Compass is the standard feed forward approach using NOM characterisation and turbidity on the raw water sample to predict an ideal coagulant dose.
- Compass EOR (enhanced organics removal) utilises the same approach on the raw water, but utilises post treatment data from a second UV sensor for a feedback trim to maximise organics removal where disinfection by product (DBP) formation potential is an issue.
For more information on NOM characterisation and how Compass uses this, see here.
There are versions of Compass available for wastewater applications, see here for more details of these.
2. Who are Lutra?
For more information on Lutra, check out their website.
Real time NOM characterisation is unique to Compass systems and uses multiple UV wavelengths to better understand hydrophobicity and molecular size of organics in a similar way to a SUVA measurement. There has been a lot of research published on the benefits of multiple UV measurements in better understanding the character of organics in raw water. It is important to note that although the online NOM parameter correlates with a SUVA measurement, they are not exactly the same.
No. pH is not an input variable although it is of course recognised that pH control is important in coagulation performance. The preference should be for the pH to be controlled to an optimum setpoint in an independent control loop. In this way, pH is not driving the dose requirement.
Having said that, there are some Compass sites where flocculation pH control is not practiced, and we can accommodate this within our site-specific calibration procedures. There is one aspect of pH influence on such sites which we do not factor into our algorithms and that is when the raw water pH is high (usually due to seasonal algae influence). In such situations, the coagulant is sometimes used as a weak and expensive acid. This can be managed by modifying the dose control via PLC whereby the Compass predicted dose is trimmed to maintain a pH below an upper limit.
5. How much variation in raw water quality do you need to see in order to optimise an algorithm and/or understand the nature of organics in a raw water source? Do you need seasonal variation for example?
The NOM characterisation parameters integral to Compass do not need to be calibrated per se and therefore we do not need to see a years’ worth of data for example. What we like to see is how the plant responds to transient conditions. There is typically a three-month evaluation period as this is normally sufficient to observe several “events”. If the source is stable a longer period may be required. Conversely, if the source is variable, it can be done faster. Our fastest evaluation period was 4 weeks from install to full plant control. The longest was over 12 months – simply because the source didn’t change. In this instance, Compass could control under the prevalent conditions but the performance couldn’t be validated under variable conditions.
Essentially, this can be inferred. Lutra have done a lot of work with Sydney Water in this area looking at how to manage non-coagulable DOC.
Lutra have also done extensive work on THMFP and HAAFP measuring the spectral changes following chlorine addition to raw water, partially treated and fully treated samples from a wide range of sources.
Our online THMFP and HAAFP parameters are used online at a number of plants for full online control strategies.
7. With Compass EOR, you can output online DBP formation potential parameters, such as THM formation potential or HAA5 formation potential, how are these parameters developed on a particular site?
These parameters use genetic algorithms and spectral data and are validated on site by laboratory tests. The Compass NOM characterisation parameter is used too. The online THMFP parameter is based on the standard method – so chlorine to excess and 7 days at a controlled temperature. It can calibrated on a simulated distribution system test.
We tend to validate on source water and filtered water prior to chlorination. The Compass parameters are transferable from points in the process and across different sites. A system using a UV254 measurement only is not.
Compass online DBPFP parameters are used for source management and continuous plant optimisation, whereby removal of TMFP can be optimised. This is exactly what can be done with the Compass EOR package.
It should also be noted that our THMFP parameter is a “bulk water” parameter. Compass cannot predict “wall effect” due to biofilms.
This is part of the Compass intellectual property. What can be said is that Compass can dive into the spectral data a little deeper than others. Our validation test for THMFP and HAAFP show very good correlations across the treatment process and different sources.
Algae have an affinity to coagulant and thus will also consume coagulant which otherwise would have been consumed by the organics. As then not all dissolved organics might be coagulated and removed as predicted by the feed-forward loop, the UV spectral sensor in a Compass EOR system picks up excess organics and adjusts the coagulant dose. However, algae blooms are mainly a polymer dose issue, and a case-by-case adjustment would be needed to manage loads. Here a feed forward component by additional data from algae and cyanobacteria sensors as part of the Compass system in the raw water stream can add on-line process control capability. Further, algae blooms have a strong effect on pH and as coagulation is pH dependent, pH control in the process is important. Lutra is also currently developing a Compass module for optimization of e.g. PAC dosing to remove potential cyanotoxins from treated water. In summary, Compass EOR takes out some of the process control uncertainty added by infrequent algal blooms.
Compass will just output an ideal coagulant dose based on the raw water quality at that time (trimmed using a UV sensor on the primary treatment outlet if using Compass EOR). It is up to site how to use that output. Some sites just use it as a reference point and maintain manual control. Other sites allow Compass to work automatically within a band and then defer to manual if Compass starts predicting an ideal dose outside of that band. Some sites use Compass in fully automatic mode and the PLC controls the dose based on the output from Compass.
To discover more about Compass coagulation control systems, check out the PMA website here.
Knowledgeable, experienced staff.
Need some advice or support? We’re here to help.
If you need a call back about how Process Measurement and Analysis can help you or have any questions at all please feel free to contact us at [email protected]. A member of our experienced team will contact you directly.