Schneider Electric SE is using edge computing to predict costly mechanical problems with rod pumps, which extract oil in remote locations where wireless connectivity isnt widely available. Photo: Schneider Electric SE
Global industrial company Schneider Electric SE is connecting oil pumps to the internet, with the intention of gathering data that can be used for advanced services such as predictive maintenance. Realizing such ambitions, however, requires a new kind of network with vastly more computing power placed at the edge, far away from the corporate data center or the cloud.
The company is using edge computing to predict costly mechanical problems with equipment that extracts oil in remote locations, where wireless connectivity isn’t widely available, said Cyril Perducat, executive vice president of digital services and Internet of Things at Schneider Electric.
The company is packaging the predictive maintenance solution as a new service offering and selling it to oil field customers.
“We’re able to go from describing a problem to recommending a path to prevent the problem even before it happens,” he said.
In edge computing, data is processed and analyzed on or near the device where it’s generated instead of first being sent to a corporate cloud or data center. This way, devices ranging from factory machines to elevators and vehicles can compute and analyze data in real-time without always relying on connectivity to a corporate cloud.
Schneider Electric is one of several enterprises experimenting with edge computing. By 2021, 40% of enterprises will have an edge computing strategy in place, up from about 1% in 2017, according to Gartner Research Inc.
Schneider Electric in 2017 began experimenting with Microsoft Corp.’s Azure IoT Edge, which connects devices in the field to gateways and is an extension of its public cloud. A gateway, used in many edge computing scenarios, is a piece of hardware located physically near an object, which aggregates information from sensors, analyzes it with software, and pushes insights and data to a corporate cloud, when necessary.
Figuring out a way to detect abnormalities with oil well rod pump systems days or weeks ahead of time is important, Mr. Perducat said, because revenue is lost for each day a pump is down and needs to be serviced.
Before Schneider Electric began its edge computing experiment this summer, a remote terminal unit was responsible for gathering and analyzing data from the pump in real-time and sending a signal to an operator when an abnormality was detected, Mr. Perducat said.
Workers would then have to physically travel to the rod pump to copy data such as temperature, pressure and sediment onto a hard drive and bring it back for analysis to discover what went wrong.
Now, a gateway device located physically near the rod pump makes it possible to aggregate data, analyze it in real time and predict rod pumps that might fail in the coming days or weeks. Schneider Electric is currently in the process of packaging the predictive maintenance solution as a new service offering and selling it to oil field customers.
“We’ve moved from reactive to proactive,” Mr. Perducat said.
The edge computing model could help detect a critical pump failure that might have otherwise led to the equipment being out of commission for weeks and repair costs of up to $1 million, he said. It’ll also help avoid unnecessary trips to the wells, which could in turn improve worker safety, he said.