Community Tip - You can subscribe to a forum, label or individual post and receive email notifications when someone posts a new topic or reply. Learn more! X
I’m working on predictive maintenance of combustion engines, for this I have 2 years of sensor data of one engine, with close to ten reported failures during that time window.
Failures are of different kinds, say:
type 1: 3 failures
type 2: 2 failure2
type 3: 1 failure
...
type n: 1 failure
So far I have these ideas:
1) Use a classification model to predict probability of failure within a time window.
Problem: Very few data points to model, with less than 10 occurrences of the same failure I don’t see feasible to overcome the risk of overfit or obtain any reliable metric, even using over-sampling
2) Principal Component analysis: I have found some white papers where PCA is used as a clustering technique to identify anomalies which may include failures as well. What I like of this approach is that will not require to have labeled data, on the other hand it seems to rely heavily on industry expertise to define categories within the groups of data. Another drawback of this approach is that thingworx does not seem to have this technique available, but have to be used with an external tool as mathcad.
3) Anomaly detection algorithm available on thingworx: It will be easier to implement than PCA, but it seems less powerful than the two previous options.
All three options suffer from the same problem, the amount of data available if very limited to generate reliable results. One alternative I have seen used for these cases is to generate synthetic data based on first principles models to simulate the different operation conditions which will include failures as well. But this may require more time than I have for the task.
Based on that information, is there a better way to approach this problem? if not, which alternatives of the mentioned above is a better fit for the thingworx platform?
Solved! Go to Solution.
I'm unfortunately unable to provide any additional advice as the concepts and use case you provided is best suited to use of ThingWorx Analytics. ThingWorx Analytics is part of the ThingWorx Platform IoT ecosystem, but requires a separate installation.
I would recommend that you investigate the use of ThingWorx Analytics which will give you the capability to do everything you described, including Anomaly Detection with minimal customization on your end.
You can review the Features and Capabilities of the Analytics Suite here: ThingWorx Analytics product offerings and functionality
Regards,
Neel
Thank you for posting to the PTC Community.
One item I want to ask before we continue, are you just using ThingWorx Platform for your solution? Or have you implemented ThingWorx Analytics, which is a distinct and separate application that uses ThingWorx Platform as a front end?
Also, have you had the opportunity to review this post about using Predictive Maintenance with ThingWorx Analytics?
Predicting Time To Failure with ThingWorx Analytics
Regards,
Neel
Thanks for the quick response.
Ideally we will kept all the development in the Thingworx ecosystem, It so we are looking into the best fit for our problem with the current data restrictions.
I'm unfortunately unable to provide any additional advice as the concepts and use case you provided is best suited to use of ThingWorx Analytics. ThingWorx Analytics is part of the ThingWorx Platform IoT ecosystem, but requires a separate installation.
I would recommend that you investigate the use of ThingWorx Analytics which will give you the capability to do everything you described, including Anomaly Detection with minimal customization on your end.
You can review the Features and Capabilities of the Analytics Suite here: ThingWorx Analytics product offerings and functionality
Regards,
Neel
Thanks, in just one more question.
Does thingWorx have support for PCA or it has to be made outside the platform, E.G. using Mathcad?
Principal Component analysis is not natively part of ThingWorx Platform, and the closest item that can achieve this is using MathCad.
As MathCad is a different product, I recommend that you reach out to their Technical Support Team for any additional support regarding that product:
Something of note, you can upload models from PTC MathCad into ThingWorx Analytics for additional analysis.
Regards,
Neel
Sorry for this second reply, but something to keep in mind, the Anomaly Detection you referenced within ThingWorx Platform requires the deployment of a ThingWorx Analytics Server in order to use it.
Details around this is in the article I previously linked regarding the Product Offerings and Capabilities of ThingWorx Analytics.
Regards,
Neel
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