The global site of the UK's leading magazine for automation, motion engineering and power transmission
18 March, 2024

LinkedIn
Twitter
Twitter link

Software integrates machine learning and drive programming

09 May, 2019

Beckhoff has integrated a real-time machine learning (ML) capability into its TwinCat 3 automation software. It says that the ML technology is powerful enough to handle advanced tasks such as motion control, and predicts that it will open up new possibilities and optimisation potential in areas such as predictive maintenance, anomaly detection, collaborative robotics, automated quality control, and machine optimisation.

Beckhoff unveiled the ML capability at the Hannover Fair, where it was also demonstrating a plug-and-play integration technology, developed in conjunction with Danfoss Drives, that allows Danfoss’ AC drives to be programmed and commissioned directly from TwinCat. Having a single tool that can handle all aspects of programming and commissioning of these drives is expected to save time and money for systems integrators and OEMs.

Machine learning allows control algorithms to be developed by analysing process data. The models to be learned are trained in an ML framework, such as Matlab or TensorFlow, and then imported into TwinCat runtime using the Open Neural Network Exchange Format (ONNX) – a standard data exchange format used to describe trained models.

TwinCat runtime incorporates new functions to support the ML capabilities: an ML inference engine for algorithms such as support vector machine (SVM) and principal component analysis (PCA); and a neural network inference engine for deep learning and neural networks.

Beckhoff demonstrated the ability of its TwinCat software to program Danfoss drives at the recent Hannover Fair

Trained ML models can be executed in real-time with a TwinCAT TcCom object. With smaller networks, response times of less than 100µs are possible, corresponding to a TwinCat cycle time of 50µs. Models can be called via PLC, C/C++ TcCom interfaces, or a cyclical task.

TwinCat 3’s multi-core capabilities are available for machine learning applications. This means, for instance, that different tasks can access an inference engine without restricting each other. The ML technology can also access the large amounts of data available in TwinCat, allowing complex data-merging operations.




Magazine
  • To view a digital copy of the latest issue of Drives & Controls, click here.

    To visit the digital library of past issues, click here

    To subscribe to the magazine, click here

     

Poll

"Do you think that robots create or destroy jobs?"

Newsletter
Newsletter

Events

Most Read Articles