Linear movements: Compensating for cogging forces with machine learning

Linear movements: Compensating for cogging forces with machine learning




Source: Beckhoff Automation

Beckhoff Automation has developed a solution that can be used to compensate for the undesirable detent forces in linear motors (cogging).

  • The AL8000 linear servomotors from the automation specialist Beckhoff meet the new TwinCAT Cogging Compensation (Engineering TE5920, Runtime TF5920) now even higher demands on accuracy and synchronization.
  • They are therefore also suitable for high precision applications such as milling machines or laser cutting machines.
  • What comes into play here is what is seamlessly integrated into TwinCAT and applied fully automatically machine learning for compensating for the cogging forces.

That cogging or the Cogging forces in linear motors are caused by the magnetic attraction between the iron core in the primary part and the permanent magnets in the secondary part. This physical effect leads to an unwanted and uneven “snapping” of the motor. This means that applications with extremely high demands on accuracy and synchronization can only be implemented to a limited extent. A solution for this is offered by AL8000 linear motors in connection with the TwinCAT Cogging Compensation software. With their help, the locking forces can be reliably compensated. In addition to the magnetic effects, those of the mechanical construction or energy chains can also be taken into account. This significantly expands the possible uses of the AL8000 iron-core linear motors.

Fully automated machine learning for cogging force compensation

Cogging Compensation is based on the fully automated application of machine learning in TwinCAT. The required data is automatically recorded by TwinCAT Cogging Compensation in the respective customer application during a reference run over the entire magnetic track length. With the help of the data recorded, the software trains a neural network, which is finally integrated into the control for current pre-control. With the current pre-control adapted in this way, the following error can be reduced by up to a factor of 7 – without a hardware change on the AL8000. In addition, the synchronization of the machine increases by a factor of up to 5.

Image: With TwinCAT Cogging Compensation, the AL8000 linear motors also benefit from modern machine learning processes. Image source: Beckhoff Automation

You can find more information here:

Learn more about what makes linear motion systems intelligent here.

Also read: “Predictive Maintenance: Monitoring robots using digital gearbox models”

Share the message “Linear movements: Using machine learning to compensate for cogging forces” with your contacts:

#Linear #movements #Compensating #cogging #forces #machine #learning

Leave a Comment

Your email address will not be published.