Stopping the Data Chaos: The Million-Dollar Challenge SkySpecs is Solving

In this episode, Allen and Joel speak with Tom Brady, CTO of
SkySpecs, to discuss the challenges and solutions in managing
multiple data sources in wind farm operations. Brady explores how
SkySpecs is revolutionizing wind turbine maintenance through
integrated data analytics, advanced drone technology, and AI-driven
decision-making tools, while sharing a glimpse into exciting
R&D developments that promise to transform the industry's
approach to predictive maintenance. Sign up now for Uptime Tech
News, our weekly email update on all things wind technology. This
episode is sponsored by Weather Guard Lightning Tech. Learn
more about Weather Guard's StrikeTape Wind Turbine LPS
retrofit. Follow the show
on Facebook, YouTube, Twitter, Linkedin and visit
Weather Guard on the web. And subscribe to Rosemary Barnes'
YouTube channel here. Have a question we can answer on the
show? Email us! Pardalote Consulting -
https://www.pardaloteconsulting.comWeather Guard Lightning Tech -
www.weatherguardwind.comIntelstor - https://www.intelstor.comWind
Energy O&M Australia Conference - https://www.windaustralia.com
Welcome to Uptime Spotlight. Shining light on wind energy's
brightest innovators. This is the progress powering tomorrow. Allen
Hall: Welcome to the Uptime Wind Energy Podcast. I'm your host,
Allen Hall, along with my co host, Joel Saxum. We're going to
discuss a challenge that's becoming increasingly critical in our
data driven world, managing and integrating data from multiple
sources in wind farm operations. In the age of smart turbines and
lot sensors, wind farm operators are awash in data. But here's the
catch. According to a recent industry survey, a staggering 54
percent of operators find managing multiple data sources to be
difficult or very difficult. It's like trying to conduct an
orchestra where every instrument is playing from a different sheet
of music. In today's episode, we'll be diving into this data
dilemma. We'll explore why integrating data from various sources is
so challenging, how it impacts decision making and operational
efficiency, And, most importantly, what innovative solutions are
emerging to tackle this issue. But that's not all. We are also
pulling back the curtain on some exciting R& D projects at
SkySpecs that promise to revolutionize how we handle data in the
wind energy sectors. From advanced analytics to machine learning,
we'll get a glimpse of the future of wind farm data management. Our
guest is Tom Brady, the CTO at SkySpecs. And Tom leads the
technology vision and development at SkySpecs, overseeing the
creation of innovative solutions for the wind energy industry. His
expertise in managing complex data systems and R& D initiatives
is crucial to addressing the challenges of multi source data
management in wind farms. Tom, welcome to the program. Tom Brady:
Glad to be here. Thank you for having me. Allen Hall: So we have a
lot to discuss actually. So we just got the grand tour of the Sky
Specs R& D facility. Both facilities. This is true. We were in
the offices also this morning and, we meet with operators all the
time. And one of the things they'll tell us is, or especially if he
asked them for data, they go, yeah, we have it, but I don't know
how to access it. Or I'm not sure what system it's in. Let me go
figure it out. Let me call somebody see if they can figure out
where the data is. Or I have too many logins. Oh, absolutely.
That's a common one. Yeah. So what is happening right now? Is it
just because we're just getting so much information? We just lost
track of it. We don't have any place to put it. We're using Google
Drive still. Some of the operators are using Google Drive, which is
insane to me. Is that where we are in the wind industry? Tom Brady:
I would say all of the above. Winding back a little bit to when we
got started in the industry back in 2016, we launched our
autonomous drone inspection or blade inspection product. And prior
to us being on the scenes we did our typical market discovery,
customer discovery, learning a little bit about the market that we
were about to try to enter and something that we saw was common
across the board was exactly as you say, I've got folder folders in
Google drive folders in box. And, maybe I've got I've got something
like maybe I'm organizing by site, turbine, inspection date, blade,
and then, radial distance. That might be how I'm organizing my
inspection data. We, a couple of times, tried to ask our customers,
Okay, can you show me the data from last year versus this year? Has
anything gotten any worse? And, I'm like, It wasn't even a question
that anyone was thinking about, at least at the time. There's also
this backdrop of increasing blade issues and blades becoming an
even more important operations and maintenance concern. But I would
say it wasn't even a thing back then to try to answer those kinds
of questions. So I think that is one part of it. There weren't,
without now we're seeing more of these modern data systems that can
actually relate. And answer those questions in different ways that
customers want to ask them. So that's a big part of it. And I think
the other big part of it, which we probably don't talk enough about
is that it's really hard to compare apples to apples when I'm
talking about, maybe it's a drive train related issue or a blade
related issue or performance related issue. So all of these systems
exist in these different silos. And you talked about the, I have
too many logins. Like I'm thinking about who is that asset manager?
Who is that person that cares about, All of the issues, not just
the blades, not just the drivetrains, not just the performance, but
all of it holistically and I personally can't imagine what that
must be like for them because there is no, there's not a there's
not a common way of speaking about maintenance issues or asset
health risks that is common among all of those different data
sources or main components. So I think that, that's a big one for
me and that's one that. Drives a lot of our strategy, especially as
it relates to our acquisition strategy. You may have seen in the
last three years, we've acquired, a CMS company. We've acquired a
finance class management company and a performance company. And
that's what we have our eye on is how can we actually tell a
cohesive story and drive that insight to action across all of these
different main component areas? Joel Saxum: At the simplest level
let's dive down and do just a blade problem because I've seen this
before, right? At the simplest level and internal damage. And an
external damage many times are related. Absolutely. But unless you
have a decent data set, that is a, catalog data set, you can't do
that. You can't look at them because what you're saying about the,
how you guys got into, the blade inspections with autonomous
drones, that was really the first, really good data structured
inspections where you could call up on, Hey, I would like to look
at this radius on the leading edge of these blades. Okay, so you
guys started that as a cascade and everybody else has followed suit
and now it's become normal in the industry and that's fantastic.
But now even two years ago, internal inspections, people were not
looking as it was more of there's technicians in there, they have a
camera and they're pulling a tape measure and they think they're
about four, 14 meters in the blade and they got a picture. Am I
looking at the trailing edge or what bond line is that? But now you
see that how those things could be related. So when we talk about
large data sets, Now you're talking, you're getting even more in
depth saying there may be a way where we can tie CMS data to blade
damage data or something of that sort. Absolutely. But you need to
be speaking the same language. This is a thing I always go back to
language wise is what people don't understand, they say the Wi Fi.
Actually, Wi Fi is a common language. And that's why everything can
speak in it. And that's something that we see every day, but not a
lot of people realize if you could put that together for data,
which is what you guys are starting to do to amalgamate all these
different data sources and data types into one system, you can
really get a holistic view of what could be happening to your
blades or anything else for that matter. Tom Brady: Yeah. I think
there's a lot of work to be done in that, at the intersection of
data, exactly. As you say I'm particularly interested in some of
the work that's ongoing both. Projects that we're working on, but
also out in the industry on things like blade sensing. So what can
we do to actually drive the increased frequency or the the
capability of continuously monitoring blades, I think is really
exciting. Overlaying blade issues with SCADA issues, or excuse me,
overlaying blade issues with SCADA data. Yeah. I'm very curious to
say, or to see what kind of information we can derive about, Why
blade damages propagate? Can we say something about how blades are
loaded and therefore how they propagate? I think we're starting to
see glimpses of that in some of our research and development, but
there's a lot of work to do. I think that's a really exciting area
for future devs. Allen Hall: So what are those things that you're
finding right now? Just by, for just a top level look at connecting
data sources together, are you realizing, yeah, there is a
connection between this gearbox and that blade having a problem?
Are those sort of Concepts and ideas and problems starting to rise
to the top a little bit more. It's an integrated system more than
just individual components. Tom Brady: I would say right now we're
at more of the stage of having the data backing to support the
things that some experts probably already had intuitions about.