Anomaly detection in an electric motor
Current sensing to detect abnormal behaviors in motors.
Motors are used for various applications and are becoming increasingly performant. They can be monitored in a very precise way thanks to the data they provide during operation. This data can also be used to enhance the application using Predictive Maintenance techniques.
Predictive maintenance consists in optimizing maintenance strategy by automatically detecting aging or predicting anomalies. Machine learning help making the data generated by the system into meaningful data for Humans. We have added AI solution directly next to the Motor Control algorithm to run both anomaly detection & classification and motor control on the same microcontroller, reducing cost of system and optimizing resources. This approach an easily be adapted to many motors and for various applications.
Predictive maintenance consists in optimizing maintenance strategy by automatically detecting aging or predicting anomalies. Machine learning help making the data generated by the system into meaningful data for Humans. We have added AI solution directly next to the Motor Control algorithm to run both anomaly detection & classification and motor control on the same microcontroller, reducing cost of system and optimizing resources. This approach an easily be adapted to many motors and for various applications.
Approach
By measuring the current consumption instead of the vibration emitted by the motor, we only need the X-NUCLEO-IHM16M1 board, no additional sensor
Each phase of the GBM2804H-100T three-Phase Motor contains the same information. For smaller signals we measured only one of them
We created a dataset of 500 signals for both normal and abnormal behaviors. We changed the speed of the motor to simulate abnormal behaviors
We created an Anomaly Detection dynamic model in NanoEdge AI studio
We trained it directly on the edge on the P-NUCLEO-IHM03 kit (NUCLEO-G431RB + X-NUCLEO-IHM16M1 + GBM2804H-100T motor) and tested it live
Each phase of the GBM2804H-100T three-Phase Motor contains the same information. For smaller signals we measured only one of them
We created a dataset of 500 signals for both normal and abnormal behaviors. We changed the speed of the motor to simulate abnormal behaviors
We created an Anomaly Detection dynamic model in NanoEdge AI studio
We trained it directly on the edge on the P-NUCLEO-IHM03 kit (NUCLEO-G431RB + X-NUCLEO-IHM16M1 + GBM2804H-100T motor) and tested it live
Sensor
Data
Regular and Abnormal signals
- Regular signals: Normal functioning
- Abnormal signals: Different speeds
Signals length 512 (1 axis)
Data rate 24 kHz
- Regular signals: Normal functioning
- Abnormal signals: Different speeds
Signals length 512 (1 axis)
Data rate 24 kHz
Results
Anomaly detection:
100 % accuracy, 0.6 KB RAM, 2.8 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, 0.6 KB RAM, 2.8 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.
