Fan Abnormality Detection Solution Based On Edge Learning

Fan Abnormality Detection Solution Based On Edge Learning.

Before an anomaly or failure occurs, the machine will produce slightly abnormal vibration patterns. Install sensors on the machine to monitor its working condition. And use machine learning algorithms to learn its normal operating status directly from the machine. By analyzing the changes in vibration, it can be detected when an anomaly occurs. Similarly, this method is also suitable for other industrial equipment.

Solution example

We put the accelerometer/circuit board on the fan and glued it with blue glue.

300 normal signals: 3 speeds (low, medium, high), 100 signals at each speed.

300 abnormal signals: blocking the airflow, moving the fan, or knocking the fan at each speed.

Create an "Anomaly Detection" project on NanoEdge AI Studio and test it in real time on NUCLEO-L432KC.

Product Advantages

Let's make something beautiful together

+852-306 976 86 See More Projects