Predicting recycled water quality using digital twin

AWS Partner Arq Group (now part of NCS Australia) has worked with Melbourne Water Corporation (MWC) to create a digital twin solution in three months, helping the organisation predict water quality issues and giving customers the opportunity to be proactive about water shortages.

Searching for a way to predict water production impacts

Melbourne Water Corporation manages and protects the water supply for more than five million people. Owned by the Victorian Government, MWC works with three water retailers to ensure citizens have reliable, safe drinking water. As Australia’s largest producer of Class A recycled water — the highest quality of recycled water — the organisation safely treats sewage and uses the waste to power treatment plants, fertilise fields and produce high-quality recycled water for a range of non-drinking uses, helping to reduce the use of water in reservoirs.

Increasingly, MWC wanted to be able to predict the impact of weather and other outside factors on recycled water production. For example, turbidity — the measure of water clarity — can negatively impact production.

“As a wholesale water provider, we provide services to hundreds of thousands of households and farmers, and they depend on our recycled water,” said Blair Smith, innovation lead at MWC.

“However, if the water is too cloudy, we can’t take it through our tertiary treatment plant and we have to advise our customers that we can’t produce water, which means they need to seek alternative water sources.”

To avoid this outcome, MWC sought to perform predictive analyses on water data captured from more than 700 sensors and meters in water treatment plants.

“We wanted to analyse factors impacting production to see if we could provide early warnings to our customers that we can’t produce while indicating when we might be able to produce water again,” Smith said.

Creating a digital twin to analyse recycled water production processes

MWC reached out to Arq Group (NCS Australia), an AWS Partner, to help solve the problem. The organisation asked for a digital twin, a virtual model that reflects the it’s recycling water production.

“Because of our existing relationship with Melbourne Water, we were able to extend our footprint to deliver a digital twin for Class A recycled water,” said Steve Avery, head of the Smart Cities and Digital Twin Program at Arq Group (NCS Australia).

“We wanted to help them get operational data to inform engineers and gain new insights into how they run water treatment plants.”

Arq Group (NCS Australia) used a service that simplifies the development of digital twins of real-world systems, to build a digital twin to model Melbourne Water’s end-to-end process of recycled water production. The process took three months.

The solution collects and analyse real-time IoT sensor data, laboratory data and weather data — such as wind speed, rain levels and temperature — and uses machine learning models to predict what might happen to water quality. Now, MWC has a holistic view of its recycled water production operations and can use dashboards to quickly view any water quality issues at a glance.

Increasing prediction accuracy

With its digital twin solution, MWC increased its ability to predict water quality issues from day one. The organisation can also predict water quality for the next two days with 75% accuracy.

“Previously, we had no predictive capabilities at all,” Smith said.

“Now, with the Arq Group (NCS Australia) solution, we have an accurate three-day outlook on turbidity and other conditions that impact our recycled water production. We can detect water quality issues faster now, so we can see whether or not we can produce water. We can then address water quality issues quickly while also predicting when water conditions will support the resumption of production.”

Because Melbourne Water is the country’s largest recycled water producer, it is critical for the organisation to better understand water quality for the production of recycled water.

“Melbourne Water can now see, in real time, production issues that need prompt attention,” Avery said.

“This gives them fast turnaround and a better ability to predict what direction the quality is going in, which is key for the production of recycled water for their customers.”

Helping customers proactively plan for water shortages

With an improved ability to predict water quality, MWC can now provide early warning to water utilities by communicating when the organisation can or cannot produce Class A recycled water.

“This allows our customers to plan more effectively when it comes to accessing recycled water for a range of purposes. They can be proactive instead of reactive, which is better for their own end customers,” Smith said.

The two organisations are currently working together to create another digital twin project based on real-time video analytics for drones.

“Projects like these are complex and require detailed data science, which is what our data science and engineering teams can provide,” Avery said.

Image credit: iStock.com/Maxxa_Satori