Motor Simulation Apparatus for Bearing Abnormality Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing simulation technologies for factory equipment maintenance in the industrial machinery field lack effective methods for simulating abnormality detection in motor systems using machine learning, particularly in accurately modeling motor dynamics and bearing abnormalities.
Innovation Solution
A simulation apparatus that includes a motor physical model structured with a motion equation section and a wiring circuit section, combined with an abnormal state model that calculates abnormal parameters such as friction torque due to lubrication deficiency or bearing damage, allowing for realistic simulation of motor system abnormalities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional simulation methods are used for factory equipment maintenance, then general maintenance can be performed, but accurate simulation of motor system abnormalities using machine learning is not achieved
Solution Approach 1:
The motor system is segmented into distinct physical models including stator winding models, magnetic circuit models, and bearing models. Each segment is modeled separately with specific equations (e.g., voltage equations for windings, flux equations for magnetic circuits) and then integrated into a comprehensive simulation framework, enabling accurate abnormality detection while managing complexity through modular decomposition.
Solution Approach 2:
A machine learning model serves as an intermediary between the physical motor system and the simulation apparatus. The ML model processes simulation data to detect abnormalities in motor operation, bearing lubrication status, and potential failures. This intermediary enables accurate abnormality detection by translating complex physical phenomena into interpretable diagnostic information without requiring direct complex physical modeling of all failure modes.
2Measurement precision
If detailed physical models are created for accurate simulation, then simulation accuracy improves, but computational complexity increases
Solution Approach 1:
The simulation model uses variable parameters to represent different operating conditions and abnormal states. Key parameters include current, voltage, speed, torque, and temperature that can be adjusted to simulate various motor conditions. The model dynamically changes these parameters based on operating state, enabling accurate representation of different scenarios without requiring separate complex models for each condition. This parameter-based approach allows a single unified model to achieve high accuracy across multiple operating points.
3Reliability
If comprehensive motor models including all components are simulated, then complete system behavior is captured, but simulation time and computational resources increase
Solution Approach 1:
The simulation model incorporates pre-defined abnormal state models for various failure conditions including bearing lubrication deficiency, winding faults, and magnetic circuit issues. These abnormal states are pre-programmed with their characteristic signatures and detection methods. When simulating, the system can directly activate these pre-defined abnormal states rather than having to develop and compute complex failure scenarios in real-time, significantly reducing simulation time while maintaining completeness of system behavior representation.
Data Source
AI summary
A simulation apparatus comprises a model storage unit storing a motor physical model modeled by a wiring circuit section and a rotation motion equation section, and an abnormal state model obtained by modeling a motor abnormal state; and a model arithmetic unit configured to perform arithmetic processing using the motor physical model. The abnormal state model calculates an abnormal parameter indicating a deviation amount from a normal state, and the abnormal parameter is input to the motor physical model.