NREL releases free, open-source energy analysis tool
- By Patrick Marshall
- Aug 13, 2013
The Energy Department’s National Renewable Energy Laboratory (NREL) has developed Energy DataBus, an open-source application for monitoring, storing and analyzing energy-related data for optimizing energy use and detecting energy "leaks."
The free software is available for download from GitHub and can be used by organizations to do their own energy data research.
What makes the application special is its use of a parallel-processing database that allows it to manage the massive amounts of data generated by energy sensors running 24/7 in buildings.
"We're using a new type of data store called NoSQL," said Dean Hiller, architect on the NREL team that developed the Energy DataBus. Specifically, the Energy DataBus employs Apache's Cassandra database. "That allows us to collect massive amounts of data," Hiller said. "We just keep adding data nodes. Not only that, we can have a higher throughput. See, we can be doing 100,000 events per second or a million events per second. All that is required is that we add more computers."
Another feature that distinguishes the Energy DataBus is that, like Cassandra, it was designed using open-source software.
"There are a lot of advantages with open source," said Keith Searight, project leader for the Energy DataBus. "It's free, it's easy and it has a low barrier to entry. And then you can leverage all the value that you get from other people in your community. In that way you can get a better product for a low price. You can build a community rapidly when there is a low hurdle."
The team also turned to PlayORM, an open-source program for integrating data from different sources. That's important, since the Energy DataBus is receiving data from a wide variety of sources, including power meters, thermometers, carbon-dioxide sensors, air-conditioning equipment and meteorological sensors. "All of those are pulled from around the NREL campus and they go into the database data store," Searight said.
The variety of data offered another challenge to the Energy DataBus developers. "One of the problems with all these sensors is that you have a huge amount of data, and they all work with different times," Searight said. DataBus can align the data points, giving them a common time stamp so analysts can draw graphs or see averages or totals. “That's one of the capabilities that DataBus provides that we didn't have before."
The Energy DataBus was not only developed in-house at NREL's facility in Golden, Colo., it was implemented in-house. "The buildings on the campus are state-of-the-art buildings," Searight said. "They're chock full of smart meters. So NREL has used our office space as a sort of a living laboratory."
The project has been under development for two years. "It started out with computer science and math Ph.D.s doing some brainstorming," he said. "We've had about five developers working on this over the course of the project.
While the Energy DataBus is currently limited to monitoring and analyzing energy data, Searight said the next logical step would be to add command-and-control capabilities, so that when a problem is detected the program can make corrections in devices. "That's a logical place to go with this," he said.
Patrick Marshall is a freelance technology writer for GCN.