Big data's big daddy is industrial turbine powered

Everybody’s favourite jet engine, turbine and medical scanner specialist company GE has been on the road recently to explain how its individual approach to big data analytics is applied to the heavy industrial zone to predict maintenance needs.

Computer Weekly recently reported on GE’s use of operational data from sensors on its machinery and engines for pattern analysis. GE is using the analysis to provide services tied to its products, designed to minimise downtime caused by parts failures. Real-time analytics also enables machines to adapt continually and improve efficiency.

As much as we now read about the increasing use of sensors (from those in electronically enabled fridges, to CCTV cameras and right through to industrial turbines) GE Research’s vice-president for software Bill Ruh also reminds us that sensors themselves are getting smarter and there is a point of interest here for software application developers.

Sensors are becoming smarter in some part due to the fact that the software controllers embedded and/or installed alongside the sensors are getting smarter. These are the software drivers capable of making decisions based upon data collected from, say, a group of sensors installed on a turbine.

NOTE: These turbine sensors would typically (and this is a layman’s definition) be in place to monitor heat, rotation speed, vibration, system health etc. For a more detailed explanation, go to your favourite industrial turbine reference manual. Essentially, GE Research’s Ruh explains that they are looking for data patterns to identify changes in the performance of the machinery itself.

Ruh says that these sensors could give pilots more information about the turbines driving a plane engine, which ultimately could make the plane safer to fly. But – importantly, we still need humans in place to interpret these signals, so the automated pilot is some way off yet.

“You still a human to set the cruise control on a car,” said Ruh. So when it comes to machine-to-machine computing, there will still be much need for the human-to-machine element.

Ruh qualifies these points by further explaining that the machine itself needs to become more aware of the human; so if engineer “Bob” tries to tinker with a turbine sensor controller and he is not an approved level of service technician, then the system doesn’t allow him access.

As these Cyber Physical Systems (CPS) now bring us a new empowered form of robotics, a new breed of developers will equally need to skill up for these field of engineering and big data at the industrial turbine powered level is born.