AI on the edge

Finger on the pulse of your machine

Holistic monitoring through AI in field-level devices

 

In the course of digitalization and optimization of industrial processes, information on the current status of machines and systems, for example, is becoming increasingly important. In addition to data from regular operation, information on maintenance requirements and intervals or deviations is now important for the planning and implementation of such production processes. But how can machine and plant operators be informed about changes in a situational manner?

You can find our English-language video here.


The solution presented by SYS TEC electronic and PowerBrain.Shop uses structure-borne sound analysis for machines and systems to show how data can be generated and processed correctly, meaningfully and currently. Based on a practical example in the video of the same name, vibration patterns of recorded data during rotation can be used to map downtimes, regular operation as well as deviations or other anomalies, among other things.

 

The recorded data is converted from the time to the frequency domain in the sysWORXX SSM-100 analysis module using Fast Fourier Transformation, or FFT for short. The frequency spectrum is transmitted together with the continuous RMS value to the controller, the sysWORXX CTR-700, and the linking of the data with the PowerBrain.Shop AI is implemented using Node-RED. The data of the FFT analysis itself is also sent via Node-RED to the AI PowerBrains, the integration of which is done by so-called Docker containers. The controller thus serves as an execution platform and connects sensors and actuators with local artificial intelligence. In this way, autonomous artificial intelligence can be conveniently ported to field devices using AI PowerBrain technology.

The acquired and labeled data is communicated to the PowerBrains via an MQTT instance. In the AI itself, the information undergoes a check and is subsequently classified accordingly. A Node-RED interface designed by SYS TEC electronic enables an uncomplicated integration and operation of the system, e.g. the training start of the PowerBrains and the monitoring of the progress. This is visually displayed in the monitor window.

Conclusion: FFT data are complex sensor data that can be meaningfully processed with the SYS TEC electronic/Powerbrain.Shop AI solution. This path enables direct on-site monitoring and control with low latency as well as an optionally available network and cloud connection for remote monitoring or control. Machine and plant operators benefit from AI as well as machine data monitoring and processing. Depending on the scenario, various deductions can be made regarding maintenance requirements, wear or loosening of mechanical connections. Along with this, anomalies or damages can be detected at an early stage and countermeasures can be initiated.

The advantages of edge AI technology over cloud applications are obvious: The system acts autonomously and is attack-proof, can also be operated without Internet connectivity and thus offers the highest cyber security. Since data sovereignty is self-evident, it remains on the machine of the operator/company and this can lead to cost savings, e.g. by eliminating costs for data aggregation from the field level to the cloud as well as transaction costs within the cloud.

Our solution is implemented using our analytics module, our controller and the low-code Node-RED interface implemented there, and a PowerBrainShop AI.