Rotor Bearing Defect Detector Ref. Application

od: Intel

AI-enabled approach to detect and predict industrial defects

The Rotor Bearing Defect Detector (RBDD) is a reference implementation that provides the user an AI-enabled approach to predict performance issues and detect bearing failure by analyzing vibration data. This industrial reference implementation uses the inference engine included in the Intel Distribution of OpenVINO™ toolkit and offers an example of how to leverage long short-term memory (LSTM) networks for preventive maintenance scenarios. 


  • Pre-built containers
  • Emulate, configure and, customize OPC client
  • Visualization of Inference results right on the edge through Grafana Dashboard
  • Edge2Cloud through Azure IoT Central, TSI, and more
  • Supported hardware: Intel® CPUs*
  • Validated hardware: Advantech* EPC-C301 Developer Kit, JWIPC* iFactry Developer Kit

* Please refer to the recommended hardware list to select the right hardware.

Continue to enhance the application just for the edge or extend it to the cloud through Microsoft Azure* IoT Hub or by creating a Custom App through Azure* IoT Central. Run a sample script to fetch telemetry data from Influx DB and send data to the Azure IoT Central dashboard, run further Analytics, or combine the results from multiple sources through Time Series Insights.

The RBDD application is built using the Edge Insights for Industrial (EII) reference stack and it utilizes the same Message Bus to communicate between modules. Users can extend Time-Series RBDD solution to include Video Ingestion and Analytics by using additional containers available under EII. Refer to the documentation on the additional components available through Industrial package on Edge Software Hub where you can also find a list of all Reference Implementation, Use Cases, and Packages.

By downloading and using this container and the included software, you agree to the terms and conditions presented at the time of the download.

Súhrnný prehľad