I want ask about the wavelet scattering feature extraction with machine learning -- if it is correct to use it for fault detection in induction machines?

  • $\begingroup$ Please clarify your specific problem or provide additional details to highlight exactly what you need. As it's currently written, it's hard to tell exactly what you're asking. $\endgroup$
    – ZaellixA
    Feb 19 at 12:53
  • $\begingroup$ I have healthy and faulty (short circuit) data, and I want to detect faults using wavelet scattering and machine learning. How can wavelet scattering be applied to signal data in MATLAB? $\endgroup$ Feb 20 at 9:01
  • $\begingroup$ Asli, first of all, the clarifications are part of the question and belong to the question body and not in the comments so I suggest you add them there. Furthermore, I believe you should describe your problem in a bit more detail. What does healthy and what does faulty mean in the context of your problem? What kind of features are you looking for, what machine learning methods are you referring to (supervised, unsupervised, clustering, classification, etc.). What kind of fault detection are you looking into? Answers to such questions could help ppl understand the problem and provide help. $\endgroup$
    – ZaellixA
    Feb 20 at 14:07
  • $\begingroup$ the stator current of three phase induction machine should be detected as a healthy class and short circut in phase A to ground as a fault class. i want to extract features through using wavelet scattering and then feed the features to supervised SVM. $\endgroup$ Feb 20 at 18:55
  • $\begingroup$ As I said, you should include all that information in the body of your question and not down here in the comments. These are parts of your problem and not random comments, so please go ahead and edit your question with that information. This may help people provide possible solutions to your problem $\endgroup$
    – ZaellixA
    Feb 20 at 19:35


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.