Beckhoff: TF3800 – TC3 Machine Learning Inference Engine
December 21, 2020
Beckhoff offers a machine learning (ML) solution that is seamlessly integrated into TwinCAT 3. Building on established standards, it brings to ML applications the advantages of system openness familiar from PC-based control. In addition, the TwinCAT solution supports the execution of the machine learning models in real time. Its capabilities provide machine builders with an optimum foundation for enhancing machine performance.
The TF3800 TwinCAT 3 function is a high-performance execution module (inference machine) for trained classic machine-learning algorithms. The algorithms are trained in established frameworks such as SciKit-Learn, libSVM or MATLAB®. The information of the learned model is loaded to the inference machine as a description file.
The execution module (ML runtime) can be called in TwinCAT 3 from the PLC, from C++ and via a cyclic task. The loaded algorithm is executed directly on the machine controller in the TwinCAT real-time cycle.
An example of a classic machine learning algorithm is the Support Vector Machine (SVM). This can be used for the classification of data sets. Example applications are End-of-Line tests, anomaly detection or process monitoring.
