• State Resolved
  • Locked Locked
  • Replies 1 reply
  • Subscribers 82 subscribers
  • Views 278 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Original question:

TMS320F28P550SJ: How to create Motor Fault Detection using Embedded AI models

TMS320F28P550SJ: Do I need to provide a faulty motor for "The Motor fault detection Edge AI model" ?

K.Z
Expert 2880 points
Part Number: TMS320F28P550SJ


Tool/software:

Hi All,

I am creating a Motor fault detection model using Efge AI.

If a customer wants to detect failure patterns as shown in the figure below,


Do I have to provide a motor for each failure pattern?
I am making a demo, and I have a normal motor, but I do not have a failed motor.
I think the development cost is very high, but is there any way to develop a motor without having a faulty motor?
For example, if we utilize the model of the failed motor prepared by TI, what would the accuracy be like?
Please let me know your opinion.

Best Regards,

Ito

  • +1
    TI__Expert 5650 points

    Hi Ito-san,

    Yes, with AI based method for fault classification, the AI model must be trained with the data under the number of classes (the number of bearing conditions in this case) you would like the model ("machine") to tell you, which means you will need the data of the faulty bearing as well.

    Since AI model relies heavily on the data, for different systems, the model may need to be trained with the data of the system to achieve the same accuracy performance. In this way, the model's parameters are updated to work for the customized system. 

    Thanks,

    Jiaxin