A fleet of trains from the Waratah region of Sydney, Australia, was equipped with intelligent software created by Downer EDI and Microsoft Azure teams to track vehicle maintenance and other data-driven decisions.
The government of New South Wales commissioned 24 Waratah series 2 trains under the Sydney Growth Trains Project in 2016 – another 17 trains were added earlier this year and another 24 will be delivered this month. The trains are maintained by Downer EDI – which maintains a 30-year contract with the state government to manage and maintain the existing fleet of 78 trains.
Downer has recently equipped the Waratah train fleet with more than 300 sensors and about 90 cameras, backed up by software that consumes sensor data from the train fleet. This allows for predictive maintenance and data-driven decisions, said Mike Ayling, general manager of digital technology and innovation at Downer.
According to Ayling, its engineers will be able to analyze trends in very granular data such as train temperature, atypical values in voltages and currents, and opening and closing times. This means that any small change in the data can give Downer an early warning of what is happening and what may need attention.
Machine learning and intelligent data analysis can also allow engineers to schedule preventive maintenance before a fault occurs; and assist in the need for spare parts to be ordered ahead of time from foreign suppliers.
Downer rolling stock services is one of the first adopters of the Azure-based solution used as infrastructure for its TrainDNA product.
Tim Young, chief executive officer, Rolling Stock, Transportation and Infrastructure Services, Downer compared the solution to a “steroid data analysis platform.”
“With massive volumes of data, this will allow us to establish relative real-time trends, allow us to predict failures in advance, and calculate the remaining useful life of an asset more efficiently,” he said. “The benefit to our customers is that all of this occurs while the train is in operation without disrupting it, while at the same time increasing worker safety through the potential for removal of high-risk inspections.”
The solution interface is an Angular Web Application built on top ASP.net services, with the solution hosted on the Azure Service Fabric, ensuring scale and resiliency.
The Azure IOT Hub feeds flow analysis into an Azure Data Lake storage and an Azure SQL database, and access is managed by Azure’s Active Directory with Power BI providing analysis and reporting.
According to Ayling, for an asset maintainer, “automation and digitization” takes the company rather than an “inspector and custodian”.