Skip to main content
LR_97965861-Visitor
1-Visitor
Jun 17, 2021 3:51 PM UTC
Solved

Time dependency of the anomaly detection module

I’m interested in predict anomalies in a process that works 24/7, but for training, I only have data of two days for each week, as depicts in the image below: 

LR_9796586_0-1623945012823.png

Considering that I have several months of data and the timestamps for each sample, is this data enough to properly train an anomaly detection model? 

nsampatAnswer
17-Peridot
Jun 17, 2021 3:59 PM UTC

@LR_9796586 ,

 

Thank you for posting your question to the PTC Community.

 

The typical suggestion for training is to have 5 full cycles of the data to obtain a good training set.

 

Details on this can be found in this Knowledge Article: https://www.ptc.com/en/support/article/CS248761

 

Should be note, Anomaly Detection has a max period of 23 hours and 59 Minutes for its monitoring, and is not capable of multi day monitoring at this time.

 

Regards,

 

Neel

 

View in thread

4 replies

nsampat17-PeridotAnswer
17-Peridot
Jun 17, 2021 3:59 PM UTC

@LR_9796586 ,

 

Thank you for posting your question to the PTC Community.

 

The typical suggestion for training is to have 5 full cycles of the data to obtain a good training set.

 

Details on this can be found in this Knowledge Article: https://www.ptc.com/en/support/article/CS248761

 

Should be note, Anomaly Detection has a max period of 23 hours and 59 Minutes for its monitoring, and is not capable of multi day monitoring at this time.

 

Regards,

 

Neel

 

Translate
LR_97965861-VisitorAuthor
1-Visitor
Jun 17, 2021 6:57 PM UTC

@nsampat  Regarding the max period of 24 hours does that mean that the sum of all 5 full cycles must be less than 24 hours, or that each individual cycle must be of 23h59m at much?

Translate
nsampat17-Peridot
17-Peridot
Jun 21, 2021 1:44 PM UTC

@LR_9796586 ,

 

The 5 cycles must be within the 23 hours and 59 minutes, otherwise its considered a new training window.

 

Ideally you would want your cycles to be periodic and cyclical, Anomaly Detection does not handle acyclical data very well.

 

Regards,

 

Neel

Translate
slangleyCommunity Manager
Support
Jul 12, 2021 8:43 PM UTC

Hi @LR_9796586

 

If you feel your question has been answered, please mark the appropriate response as the Accepted Solution for the benefit of others with the same question.

 

Regards.

 

--Sharon

Translate
Terms & ConditionsAccessibility statement