Pilot projects are valuable for testing smart metering and smart grid technologies and their deployment. But experience is that in large scale rollouts with end points numbering in the hundreds of thousands and into the millions – such as have been experienced in the larger utilities in North America and are now starting to occur in Europe to meet the EU 2020 smart meter mandate – a new set of challenges occur.
“We observe a number of common challenges around data integration and analysis, both during the deployment and in operation,” Kai Hui, chief technology officer, Bit Stew Systems, told Engerati in an interview. “Often there are multiple communication networks and data streams which need to be upgraded and integrated. During a deployment of thousands of end points in a month, data from disparate sources needs to be brought together to ensure the installs are correct and the devices are communicating. And there is the need to extract useful information from the data, amounting to potentially millions of data points in a day.”
It all comes down to having operational insight, says Mr Hui, who was previously at BC Hydro, where he was involved in the company’s 2 million smart meter rollout. “For some Big Data is seen as the solution, but Big Data is actually the problem. Typically IT apps fail at about the 1 million device level and so suitable apps need to be purpose built.”
Meeting the data challenge
Meeting this data challenge will form the focus of an upcoming Engerati webinar, What’s Next After the Smart Grid?, which will be presented by Mr Hui along with Andrew Miller, Sales Engineer at Bit Stew Systems.
In broad terms, a solution should address three areas, people, process and technology, says Mr Hui. “A new matrix organizational structure is needed to support operations, which can’t happen if people work in silos. Processes need to be defined to handle the scale, and the technology applications also need to be able to scale.”
Bit Stew’s focus is specifically on the third of these areas, with its MIx Director solution, built on the MIx Core platform and with accompanying apps, which is aimed to extract value from data to provide utilities with insight into their operations, assets and customers.
“We use a semantic model to integrate data from multiple sources and with real-time analytics with purpose built apps with machine learning capability, we can scale to provide actionable insights,” says Mr Miller. He adds that although not an original goal of the model, increasingly it is being requested by utilities to help drive the process changes and data governance.
Data analytics in action
Bit Stew’s solution has been implemented at numerous utilities with a million-plus smart meters, both electric and gas. In the webinar, two of these will be presented as real-world examples.
In the case of a North American utility with 2 million smart meters MIx Director was used to help manage the deployment by bringing together all the necessary data streams and for example one use case was to assess where to install devices to optimize the network. Another strong use case was the subsequent ability to rapidly triage data to identify problems on the network.
Similarly in the case of a California utility, the solution has been deployed as a smart meter operation event exception management application. “Although smart meters had been installed for a number of years, the company found it couldn’t manage the volume of events coming in,” says Mr Hui.
He adds that the solution implementation time is very rapid and for example in the case of the North American utility it was fully integrated within 3 months at the start of the smart meter deployment. In another utility in Oregon with 800,000 smart meters the event management use case was implemented in 1.5 weeks.
“This quick integration is part of the value and it’s basically time to value.”
To learn about best practices and a recommended data architecture approach to gaining operational insight from data register for the webinar.