Fan anomaly detection based on vibrations
Learn to detect abnormal behavior at the edge on a vibrating machine.
All the equipment interacts with their environment by emitting various signals. These signals are source of relevant information reflecting equipment functioning. Being able to understand these signals allows significant optimization capabilities.
For example, before an anomaly or failure occurs, your machine generates slightly abnormal vibration pattern. By placing a sensor on the machine, we can monitor its activity. Thanks to Machine Learning, we learn directly from the machine what is its normal functioning. By analyzing the evolution of vibrations, we can detect the appearance of an anomaly. We have implemented this approach on a fan motor for demonstration purposes, but this approach can easily be adapted to many industrial machines.
For example, before an anomaly or failure occurs, your machine generates slightly abnormal vibration pattern. By placing a sensor on the machine, we can monitor its activity. Thanks to Machine Learning, we learn directly from the machine what is its normal functioning. By analyzing the evolution of vibrations, we can detect the appearance of an anomaly. We have implemented this approach on a fan motor for demonstration purposes, but this approach can easily be adapted to many industrial machines.
Approach
We put the accelerometer / board on a fan. We stick it using blu-tack
300 regular signals: 3 speeds (low, medium, high), 100 signals per speed
300 abnormal signals: for each speed, block the air flow, move the fan, tap on it
We created an Anomaly detection project on NanoEdge AI Studio and tested it live on a NUCLEO-L432KC
300 regular signals: 3 speeds (low, medium, high), 100 signals per speed
300 abnormal signals: for each speed, block the air flow, move the fan, tap on it
We created an Anomaly detection project on NanoEdge AI Studio and tested it live on a NUCLEO-L432KC
Sensor
Accelerometer (3-axis): LIS3DH
Data
Regular and Abnormal signals
- Regular signals: 100 signals per speed (low, medium, high)
- Abnormal signals: for each speed, block the air flow, move the fan, tap on it, etc.
Signal length 1536 (512 per axis, 3 axes)
Data rate 1.6 kHz, Range: 2g
- Regular signals: 100 signals per speed (low, medium, high)
- Abnormal signals: for each speed, block the air flow, move the fan, tap on it, etc.
Signal length 1536 (512 per axis, 3 axes)
Data rate 1.6 kHz, Range: 2g
Results
Anomaly detection classes:
100 % accuracy, 7.8 KB RAM, 6.1 KB Flash
Blue points correspond to normal signals, red points to abnormal ones.
The signal numbers are on the abscissa and the confidence of the prediction is on the ordinate
100 % accuracy, 7.8 KB RAM, 6.1 KB Flash
The signal numbers are on the abscissa and the confidence of the prediction is on the ordinate
Resources
Model created with NanoEdge AI Studio
A free AutoML software for adding AI to embedded projects, guiding users step by step to easily find the optimal AI model for their requirements.
Compatible with STM32
The STM32 family of 32-bit microcontrollers based on the Arm Cortex®-M processor is designed to offer new degrees of freedom to MCU users. It offers products combining very high performance, real-time capabilities, digital signal processing, low-power / low-voltage operation, and connectivity, while maintaining full integration and ease of development.
