Melbourne Water is testing a custom-developed artificial intelligence (AI) programme that coordinates pump configuration with the amount of treated water required at any given time – without any human intervention. The settings are then applied in real time.
“The Python programme is able to utilize our historical data to determine the most energy efficient combinations of pumps and the associated speeds to run them at in order to achieve the necessary flow rate,” said Melbourne Water Automation Team leader Russell Riding. The platform is powerful enough to consider a wide range of factors unique to the city’s water supply system, including reservoir level, available pumps and past performance.
Cybersecurity was an important consideration when trialling the system, added Riding. “The AI is stored on a computer which is not connected to the broader Melbourne Water network, or the internet. This is best practice to ensure that cyber security risks are minimized.”
The project is expected to reduce Melbourne Water’s pump station energy costs at its Winneke treatment plant, which handles some 350 million litres of water every day, by around 20 per cent per year. Python is now being tested on another pump station with different requirements to find out if it can replicate the same kind of results that have been achieved at Winneke.
