Predictive Analytics & Real Time Monitoring (GE Intelligent Platforms)



this is Peggy Smedley the editorial director of connected world and the host of the Peggy Smedley show and today I'm with Chad Stoker who is the manager of remote monitoring and diagnostics for GE intelligent platforms Chad welcome thank you so today tell us a little bit about GE and what you're doing there and how it all relates to the Internet of Things sure yeah absolutely so I manage a remote monitoring Diagnostic Center for intelligent platforms so when GE talks about the Internet of Things our things are high value industrial equipment so our monitoring center monitors more than 6,000 pieces of equipment GE and non GE equipment were om agnostic we collect data in real time from things like combustion turbines boiler feed pumps steam turbines and generators we analyze that data in real time and we use predictive analytics to help our customers gain advantages in predictive maintenance and optimization talk about what that means a predictive analytic analytics and maintenance you know so people understand because sometimes people really don't understand exactly what that means in the big scheme of things I think that you know the change that's coming to the industrial segment relative to the Internet of Things is is changing how maintenance professionals work today maintenance professionals are used to responding to hard alarms they're used to responding to equipment when it's broken but there's opportunities to better there's more money to be made by fixing equipment before it's broken by scheduling outages on weekends as opposed to weekdays they can save on on production costs they can have parts staged they don't have to fly people in on an emergency basis so the ability to know ahead of time whether it's days weeks months can turn into significant dollars for industrial customers can you give me some examples of how customers are being more proactive instead of reactive than if what you're talking about we publish what we call a catch of the week so every week we generate about 60 customer reports and we generate about 40 actual notification per week so about 40 times per week customers are writing predictive maintenance work orders based upon the information our Center sends out in the predictive analytics so they're changing sensors before they cause you know trips or alarms they're doing inspections before they cause trips or alarms you know so on and so forth and one of our best stories recently was a combustion turbine we were monitoring a combustion turbine for an oil and gas facility in the Middle East we picked up slight vibration changes so very small increases in vibration not enough to hit a heart alarm not enough to trip the piece of equipment and we told the customer about those those vibration increases the evidence wasn't compelling enough for the customer to take action for a customer take a piece of equipment off line it's a very very costly proposition so we continue to help that customer monitor we saw increasing evidence of a potential issue developing the customer was able to use the information to bring in a vibration specialist they made the call to do a proactive inspection during an upcoming outage but they weren't planning on doing so they did a proactive inspection they found cracks and turbine blades the estimated turbine blades were three to five days away from liberation would have caused a catastrophic wreck of the turbine plus you know a week or more production loss than that total to around 30 million dollars so just the ability to know something early and to do an inspection early can can lead to significant cost savings we could we talked about as the right information to the right person at the right time when you talk about 30 million dollars I think everybody who's watching right now you just woke up them all up Chad thank you for spending time with us we appreciate it sir no no worries I'm Travis you

Be First to Comment

Leave a Reply

Your email address will not be published. Required fields are marked *