5 Comments

Summary:

GE is pushing new technology that uses hundreds of sensors and advanced analytics to make its fleet of gas turbines run more efficiently. GE is a big fan of big data, investing more than $100 million in tech companies this year alone.

A heavy duty gas turbine.

GE announced new technology for its utility customers running gas turbines on Thursday, but the bigger news might have been the technology that’s powering them. The new products are part of GE’s broader Predictivity lineup that analyzes data from the company’s large installed turbine base to help the machines and their operators work better.

Here’s how the company quantifies the promise of smarter turbines:

“In September 2013, GE eclipsed 100 million hours of operational data documented on its globally monitored gas turbine fleet of more than 1,600 units, the world’s largest. The insights derived from analysis of this operational “big data” can be applied to help customers expand their earning power while reducing operational costs and risk. As these “intelligent” machines communicate their operating statistics through an average of 100 physical sensors and 300 virtual sensors on each gas turbine, the GE team can help customers translate that information into actionable decisions. …

“… Unlocking the full capacity of a 500-MW power plant could be worth more than $500,000 annually in increased revenue, while a public utility that could reduce its heat rate efficiency curve by 1 percent could save up to $1.25 million dollars annually in fuel costs.”

GE, of course, is no stranger to the idea of the industrial internet. In April the company invested $105 million — a 10 percent stake — in EMC/VMware spinoff Pivotal, which is promising a collection of infrastructure technologies that will enable better collection and analysis of the untold amounts of data streaming off machines across the industrial landscape. In July, GE Ventures took part in a $30.6 million investment round for Ayasdi, a machine learning startup that’s helping GE identify signals that can predict failure in jet engines.

  1. This is a great to see the hype replaced by the real stories of what Big Data can do, especially alongside the Internet of Things. We’re going to see the constant monitoring through a myriad of sensors become the norm on things that are less costly than turbines.

    “Things” rarely fail in a moment…they give signals that can be monitored and problems proactively addressed. We just never had that data before. Fascinating times.

    Share
    1. Gigs of data can be downloaded from various sensors, from the compressor to the thermocouple..but all the data in the world will not stop a catastrophic event from occurring. That could be an object (bird) getting past the air filters or a bucket breaking off and destroying the compressor section. So actually “Things” can fail in a moment and can’t be put into a pretty graph.

      Share
      1. Ma'en Al-Tamemi Sunday, October 6, 2013

        So lets drop this tech altogether just because there are still certain errors which we can’t anticipate?!

        One could easily argue that when the sensors are doing their job and the numbers are analyzed, performance is maximized and technical problems are minimized; THEN you will have more time to watch out for birds and buckets.

        I see this as a great example of technology growing organically from the core of application to aid in the process rather than re-inventing and redesigning it.

        Share
    2. Excellent story and an even better observation, Chris. No argument here. Case studies are starting to trickle in.

      Share
  2. As proverb says prevention is better than cure. Analysis is always helps to operate the units in a better way and improve the efficiency and system availability.

    Share

Comments have been disabled for this post