Servomotor Insulation Resistance Estimation Using Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional motor insulation deterioration diagnosis methods lack accuracy in estimating the insulation resistance of servomotors.
Innovation Solution
A machine learning device and method that generates a learning model using training data including initial and final insulation resistance, and time-series data of servomotor conditions, enabling the estimation of future insulation resistance through supervised learning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional motor insulation deterioration diagnosis methods are used, then the diagnosis can be performed, but the accuracy of insulation resistance estimation is insufficient
Solution Approach 1:
The patent transforms the insulation resistance estimation problem from a direct measurement task to a predictive task by changing the parameters used for estimation. Instead of relying on single-point measurements, the system uses multiple parameters including initial insulation resistance, operating hours, temperature history, humidity, and load conditions to predict future insulation resistance values. This parameter transformation enables more accurate and reliable estimation of motor insulation deterioration.
2Loss of information
If simple insulation resistance measurement is performed, then the measurement process is simple, but the predictive capability for future insulation resistance is insufficient
Solution Approach 1:
The patent applies preliminary action by collecting and storing various operational parameters (temperature, humidity, load conditions, operating hours) during the motor's operation phase. This preliminary data collection enables the system to predict future insulation resistance values before actual deterioration occurs, providing advance warning for maintenance planning without requiring complex real-time monitoring during critical periods.
Solution Approach 2:
The patent replaces direct mechanical/electrical measurement systems with an information processing and prediction system. Instead of continuously measuring insulation resistance through complex electrical tests, the system substitutes this with a prediction model that processes operational data (temperature, humidity, load patterns) to estimate future insulation resistance, thereby reducing measurement complexity while improving predictive capability.
3Measurement precision
If insulation resistance is measured at multiple time points, then more data is available, but the time and resources required increase
Solution Approach 1:
The patent creates a virtual copy of the insulation resistance measurement process through predictive modeling. Instead of performing actual repeated measurements over time, the system copies the essential relationships between operational conditions and insulation deterioration by training a prediction model on initial measurement data and operational parameters. This model then generates predicted insulation resistance values at future time points without requiring physical re-measurement, saving time and resources while maintaining data accuracy.
Data Source
AI summary
A machine learning device includes: a training data acquisition unit configured to acquire multiple pieces of training data each including insulation resistances of a servomotor at the beginning and the end of a certain period and time-series data indicating conditions of the servomotor in the certain period; and a learning model generating unit configured to perform a supervised learning using the training data to thereby generate a learning model.