BookmarkSubscribeRSS Feed

Want to detect degradation in your IoT enabled devices? Learn how with SAS Analytics for IoT

Started ‎03-27-2020 by
Modified ‎03-27-2020 by
Views 3,489

Early failure or degradation analysis is crucial for managing operational cost and downtime of high-value assets. With SAS Analytics for IoT you can monitor operations in real-time using streaming data from IoT enabled devices.


In the use case, we used simulated hourly energy data (in kilowatts) produced by four wind turbines. We applied Subspace tracking (SST) algorithm to detect degradation in real-time using streaming data. The algorithm is packaged in SAS Event Stream Processing Studio (included in the SAS Analytics for IoT). SST detects anomalies and system degradation in systems that generate high-frequency, high-dimensional data.


Key take-away from the use case:

  • Become proficient in designing a streaming model for real-time failure detection
  • Effectively use the SST algorithm to detect degradation in real-time
  • Understand how to apply best practices for the SST algorithm

For this exercise Turbine 4 has been simulated to decrease energy production. We are applying the SST algorithm to get timely alerts on its operations.


This approach converts a set of correlated variables to a set of linearly uncorrelated variables known as principal components. Because the first few principal components usually capture most of the variability in the data, they can be tracked over time to assess whether any changes have taken place in the subspace that is spanned by the data. We can use SST to detect degradation by tracking the absolute angle of the first principal component vector.


Figure 1 shows the absolute values of first principal component that is obtained after applying the SST method over the sliding windows of data.




The first principal component displays the same pattern of changes for the first three turbines. However, the first principal component for Turbine 4 starts to behave differently from the other three turbines around window id=700, indicating an abnormality. This plot can help determine which turbines are deteriorating. SST can give you a relatively clear indication of which part of the system is getting out of control.


Figure 2 displays the angle change and absolute angle of the first principal component between consecutive windows.



The angle change and absolute angle plots clearly shows the subspace change or gradual increase over time starting around window id=700. With this information, decisions can be put into place to trigger maintenance activities when the absolute angle or angle change value is above an acceptable level.


The SST algorithm can monitor the system in real-time and the faulty wind turbines can be immediately identified. This method can be used in other systems like HVAC, solar farm, etc. that generate high-frequency, high-dimensional data to detect anomalies and degradation in real-time.


Download the Files to Try it Yourself!

If you would like to build it yourself, please check out our GitHub page where you can find step-by-step instructions, the CSV data file and XML model file.     

Take Me To GitHub


Additional Resources

Version history
Last update:
‎03-27-2020 05:36 PM
Updated by:



Registration is open! SAS is returning to Vegas for an AI and analytics experience like no other! Whether you're an executive, manager, end user or SAS partner, SAS Innovate is designed for everyone on your team. Register for just $495 by 12/31/2023.

If you are interested in speaking, there is still time to submit a session idea. More details are posted on the website. 

Register now!

Free course: Data Literacy Essentials

Data Literacy is for all, even absolute beginners. Jump on board with this free e-learning  and boost your career prospects.

Get Started