HomeVideos

Demo: A real-time 3D digital twin built on Microsoft Fabric

Now Playing

Demo: A real-time 3D digital twin built on Microsoft Fabric

Transcript

106 segments

0:00

Here we can see our

0:02

digital twin app running inside a Fabric

0:06

Raven application. And what we can see

0:09

here in the screen, so

0:11

at least on the middle here, we have the

0:14

3D visual representation of our wind

0:17

farm.

0:18

And here we can see we can zoom in and

0:21

out, pan and zoom here, of course.

0:24

Then we can see

0:26

the real terrain that's made from the

0:30

National Survey of Finland lidar data.

0:33

Now here

0:34

in real representation.

0:37

But it's on top of that, we also have

0:39

the live telemetry

0:41

from the wind park itself. So we have

0:44

here

0:45

some

0:46

output, what's our production gap, and

0:48

what kind of alerts we have there.

0:50

Then we have also the weather

0:52

information that's matching

0:55

the actual weather

0:56

in our wind park location.

0:59

And here we can see

1:01

that when we click on a turbine,

1:03

we can see here that this turbine number

1:05

seven,

1:06

it has some gearbox degradation. So

1:09

let's open up an engineering view of

1:11

this turbine. So now we can see the

1:13

turbine in more close-up view.

1:16

And then we can also explode this.

1:18

So now

1:20

as this turbine, it's a real 3D

1:23

digital

1:24

um

1:25

presentation, we can see the turbine.

1:28

And here we had

1:29

noticed that it had a gearbox issue. So

1:32

we can click here that see that, "Hey,

1:34

we have now some

1:36

vibration things going on."

1:39

So here we can see explode the turbine.

1:42

And now

1:43

this was the turbine uh rendered in as a

1:47

as a real 3D model in Blender.

1:51

And of course, we can then talk to our

1:53

turbine.

1:54

So let's see. Let's start to explain the

1:57

biggest production gap right now with

1:59

the chat.

2:01

So,

2:02

now it's uh

2:04

doing its job inside Microsoft Foundry,

2:07

calling the model modules, and now also

2:11

>> [clears throat]

2:11

>> we are getting the response. So, there

2:14

is the production gap in the turbine

2:15

number seven.

2:17

And there's a gearbox degradation.

2:20

And then we got some information that

2:22

what kind of things are there happening

2:25

with this turbine.

2:26

But now,

2:27

on top of this,

2:29

we can also open up a chat with the

2:31

turbine.

2:33

So, now

2:35

we can see here that the turbine the

2:37

chat is open. So, let's ask a couple of

2:39

questions.

2:43

>> Hi there. Welcome to the wind farm demo.

2:47

How can I help with the farm today?

2:50

>> Tell me about turbine seven.

2:53

>> Please hold as I check.

2:56

WT-07

2:58

shows a critical risk with a lost output

3:01

of .20 MW and an impact of about 14 euro

3:06

per hour.

3:08

The key issue is indicated gearbox

3:10

degradation. Next action, prioritize an

3:13

immediate maintenance inspection and

3:15

mitigation planning for that turbine.

3:18

>> Tell me also about other active alerts.

3:23

>> Please hold as I check.

3:25

There are four active alerts with one

3:27

critical on WT-07

3:30

and three warnings on WT-03,

3:33

WT-04,

3:35

and WT-11.

3:37

The most important KPI is WT-07

3:41

at 28% below expected output. Risk level

3:45

is critical for WT-07.

3:48

Next action, address WT-07

3:52

first, then plan corrective checks for

3:55

the warning turbines.

3:57

>> Now here we saw that that

4:00

we were able to have a good conversation

4:02

with our wind farm and that's was

4:05

very much due to the fact that the tool

4:07

calls they were really really fast and

4:10

then we got a really nice answers while

4:13

talking to the wind turbine.

4:16

So here we have our wind turbine demo.

Interactive Summary

This video presents a digital twin application for a wind farm, hosted within Microsoft Fabric and Raven. It demonstrates how users can visualize wind farm telemetry, interact with 3D models of individual turbines to diagnose issues like gearbox degradation, and utilize an integrated AI chat interface to query real-time performance data and receive maintenance recommendations.

Suggested questions

3 ready-made prompts