System Integration & Control Systems Design

Machine learning for all areas of automation

June 2019 System Integration & Control Systems Design

Beckhoff now offers a machine learning (ML) solution that is seamlessly integrated into TwinCAT 3 software. Building on established standards, TwinCAT 3 Machine Learning brings to ML applications the advantages of system openness from PC-based control. In addition, the TwinCAT solution supports machine learning in real-time, allowing it to handle even demanding tasks like motion control. These capabilities provide machine builders and manufacturers with an optimum foundation to enhance machine performance, e.g. through prescriptive maintenance, process self-optimisation and autonomous detection of process anomalies.

The fundamental concept of machine learning is not to follow the classic engineering route of designing solutions for specific tasks and then turning these into algorithms, but to learn the desired algorithms from process data instead. With this alternative approach, powerful ML models can be trained and then used to deliver superior solutions. In automation technology, this opens up new possibilities and optimisation potential in many areas, including predictive maintenance and process control, anomaly detection, collaborative robotics, automated quality control and machine optimisation.

The models to be learned are trained in an ML framework, such as MATLAB or TensorFlow, and then imported into the TwinCAT runtime via the Open Neural Network Exchange Format (ONNX), a standardised data exchange format used to describe trained models. The TwinCAT runtime incorporates the following new functions for this purpose:

• TwinCAT 3 Machine Learning Inference Engine for classic ML algorithms, such as support vector machine (SVM) and principal component analysis (PCA).

• TwinCAT 3 Neural Network Inference Engine for deep learning and neural networks, such as multilayer perceptrons (MLPs) and convolutional neural networks (CNNs).

Model results are directly executable in real-time

Inference i.e. the execution of a trained ML model, can be performed directly in real-time with a TwinCAT TcCOM object. With smaller networks, system response times of less than 100 s corresponding to a TwinCAT cycle time of 50 s are supported. Models can be called via PLC, C/C++ TcCOM interfaces or a cyclical task.

Through seamless integration with the control technology, the multi-core support provided by TwinCAT 3 is also available for machine learning applications. This means, for instance, that different task contexts can access a particular TwinCAT 3 Inference Engine without restricting each other. All the fieldbus interfaces and data available in TwinCAT can be fully accessed as well. This allows ML solutions to use immense amounts of data, for example, for complex sensor data fusion (data merging), and it also means that real-time interfaces to actuators are available to enable, among other things, optimal control.

For more information contact Michelle Murphy, Beckhoff Automation, +27 11 795 2898,,


Share this article:
Share via emailShare via LinkedInPrint this page

Further reading:

SCiBOTRON’s quality culture the key to success
May 2020 , System Integration & Control Systems Design
The company was founded on lean principles and quickly grew into a QSE (Qualifying Small Enterprise). Ricardo Paddy, managing director and founding member, attributes one of the reasons for the company’s ...

Portion cutter for fish and poultry
November 2020, Beckhoff Automation , System Integration & Control Systems Design
Extremely fast and highly precise meat portion cutter leverages PC control.

Omniflex enables mine water quality
October 2020, Omniflex Remote Monitoring Specialists , System Integration & Control Systems Design
Water licences carry a ‘Burden of Proof’ that the licence holder will not harm the environment or the water system through its activities. This covers the use of water from the system and the responsible ...

Safe and efficient water and wastewater treatment
October 2020, Endress+Hauser South Africa , System Integration & Control Systems Design
Depending on the application and the industry, water needs to have different properties to ensure the quality of finished products and the overall safety of production.

PC-based control monitors road tunnel with flexible scalability
October 2020, Beckhoff Automation , System Integration & Control Systems Design
With more than 200 industrial PCs in operation, the Kaisermühlen Tunnel in Vienna shows that scalable PC-based control technology from Beckhoff is an optimum automation platform in this environment.

PC-based control reduces complexity in folio applications
October 2020, Beckhoff Automation , System Integration & Control Systems Design
BW Papersystems relied on PC-based control for the development of the eCon folio-size sheeter, which greatly reduced complexity and costs compared to the legacy system it replaced.

Improved production capacity and efficiency
September 2020, Omron Electronics , System Integration & Control Systems Design
BOLÇİ Bolu Chocolate (Bolci) was formed in 1992. It now has over 300 employees and produces over 800 different products. The company produces over 200 tons of chocolate items each month in a factory that ...

Modular towel production lines
September 2020, Beckhoff Automation , Motion Control & Drives
Rapid development and configuration of manufacturing systems for terrycloth towels with PC-based control.

Open Beckhoff technology sets adoption record
September 2020, Beckhoff Automation , News
Now, with 3000 officially registered device manufacturers, EtherCAT has the broadest base of adoption in the industrial Ethernet market.

EtherCAT I/O for conveyor control
September 2020, Beckhoff Automation , System Integration & Control Systems Design
Integrated compact motor controller reduces cabling effort and increases diagnostic capabilities for roller conveyor systems.