Diagnostic and protection device for robot reducer and method thereof
A machine learning-based diagnostic device for robot reducers addresses malfunctions by detecting drive motor current and speed, diagnosing faults, and predicting lifespan, ensuring continuous production by protecting the drive motor.
Patent Information
- Application Number
- JP2024063129
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2023-07-21
- Filing Date
- 2024-04-10
- Publication Date
- 2025-11-19
- Estimated Expiration
- 2044-04-10
AI Technical Summary
Industrial robot reducers malfunction due to foreign matter mixing with lubricating oil, causing overheating and potential damage to the drive motor encoder, leading to reduced lubricating grease lifespan and requiring shutdown of all robots on the production line.
A diagnostic device and method that uses machine learning to detect drive motor current and speed, construct a learning model, diagnose faults, and predict the reducer's lifespan, providing diagnostic results to the motor driver for protective action.
The solution effectively detects abnormalities and predicts the lifespan of the robot reducer, preventing malfunctions and protecting the drive motor, thereby maintaining production continuity.
Smart Images

Figure 0007773582000001 
Figure 0007773582000002
Abstract
Citation Information
Patent Citations
Industrial robot, failure detecting method thereof, and recording medium recording failure detecting program for industrial robot
JP1999129186A
Machine learning device for learning failure prediction of main axis or motor driving main axis and machine learning method, and failure prediction device having machine learning device and failure prediction system
JP2017188030A
Anomaly detection system, molding machine system, anomaly detection device, anomaly detection method, and computer program
JP2022113522A
Load monitoring device and load monitoring method
JP2022141526A
Abnormality diagnosis device and abnormality diagnosis method
WO2020031225A1