Industry partnership: Sydney Water

Sydney Water maintains 22,600km of pipes at a cost of $40 million per year. When these pipes burst, they cause major disruption – shutting down streets, cutting off supply to homes and businesses, racking up a significant emergency repair bill. With aging infrastructure and a growing population, this challenge is becoming increasingly critical.

But most of these assets are difficult to access. That’s why Sydney Water approached UTS. The water utility wanted to explore smarter ways to manage and maintain its assets, including the anticipation of failures in their critical water mains. 

This long-running partnership has since produced a range of forward-thinking solutions to the challenges facing Sydney Water – and the water industry as a whole. 

According to Daniel Cooper, Sydney Water’s former Head of Innovation and Product Development, the impact of this work cannot be overstated.

“It has not only improved the efficiency and effectiveness of our operations but has also played a pivotal role in shaping Australia’s water security, quality and sustainability.” 

Getting specialised robots to do the hard work 

When a water pipe bursts in a Sydney street at 3am, Professor Sarath Kodagoda’s team gets a call. 

Deploying a purpose-built robot into the pipe, the UTS Robotics Institute enables Sydney Water to assess the damage and understand the condition of the remaining pipe wall. 

These remote-controlled robots are the result of years of development and trials on Sydney Water’s network. 

According to Professor Kodagoda: “These robots provide crucial data for making informed timely decisions that save money on renewal costs, help protect the environment and improve human health and safety. This technology has the potential to completely revolutionise how  water and wastewater infrastructure is inspected, in Australia and overseas.”

Collaborating across iterative projects, UTS and Sydney Water have refined the design of the robots and their deployment mechanisms to make them increasingly adaptable to different network conditions.

Originally developed to inspect large water mains during planned maintenance shutdowns, the condition assessment robots can now be deployed into much smaller pipes and into live, high-pressure environments. 

Another type of robot are deployed into the sewer system conducting safe and remote inspections in confined spaces. The technology also allows them to monitor subsurface conditions, amount of corrosion, and three-dimensional reconstruction of the pipes , giving them an advantage on CCTV cameras 

“Once operationalised, we estimate the technology will save Sydney Water $2 million per year in maintenance costs and, with timely renewals, more than $5 million per year in structural renewal costs,” says Professor Kodagoda.

Sydney Water has incorporated the use of the robots into its rapid response and planned maintenance, using the robots to assess wat`er pipe condition where a breakage has occurred and inspect pipes to predict future failures.

Smart sensors and AI saving millions of dollars in lost drinking water

UTS is also supporting Sydney Water by using smart sensors and artificial intelligence (AI) to detect leaks and potential breaks. 

Developed through a collaboration between the UTS Data Science Institute (DSI) and UTS Robotics Institute, this world-first technology has been rolled out to hotspots across Greater Sydney to monitor pipelines and cut down water wastage.

“These acoustic sensor devices allow us to listen to the sounds within the pipe,” says Sydney Water’s Head of Customer Hub, Darren Cash. “Some people like to call that the music of the network.”

The buried sensors use acoustic vibrations to pick up variations in critical water mains and wastewater pressure mains, with a connected AI platform distinguishing between leaks and other disturbances. 

With 95% accuracy, Sydney Water can target its resources to repair these pipes before the leaks become breaks.

“It’s changing the game of maintenance from reactive to proactive,” says Distinguished Professor Fang Chen, Executive Director of the UTS Data Science Institute.

During a 30-month trial, the world-first technology saved over 10,000 megalitres of water – the equivalent of 4,000 Olympic swimming pools. In monetary terms, this water is worth $23 million to Sydney Water and its customers in supply and repair costs.

Using predictive analytics to save on money, water and disruption

Another initiative of the partnership involves using a predictive model to anticipate pipe failures.

Based on 20 years of Sydney Water data, UTS DSI developed a failure prediction tool for critical water mains – the first of its kind in the world.

Combined with information from sensors developed by UTS robotics researchers, the tool is being used to target ‘intelligent’ condition assessments, repairs and renewals. 

“With our prediction tool we can identify and visualise high-risk pipes on a map-based platform,” says DSI’s Associate Professor Yang Wang.

Helping Sydney Water pinpoint its efforts, the tool can predict with 80 per cent accuracy within 200m of where a water main is likely to fail. 

Sydney Water’s Asset Infrastructure Research Coordinator, Dammika Vitanage, highlights the significance of this achievement, the result of “Sydney Water and UTS collaborating over a long period”.

“They have delivered world-first predictability of water pipe failures.” 

Reflecting on the culture of the collaboration, he notes: “I don’t think there is any team that is more cordial and value-adding, that I’ve worked with in any organisation.” 

Looking at UTS’s support for Sydney Water holistically, the failure prediction tool can be used to identify candidate locations for placing acoustic sensors or for deploying condition assessment robots.