Machine Part Life Estimation Using Physical Quantity Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The scheduled inspection time for ball screws in injection molding machines is not accurately reflective of the actual degree of damage, leading to inadequate maintenance, as the extent of damage varies between machines even at the same scheduled inspection time.
Innovation Solution
A life estimation method that acquires and stores physical quantity data from industrial machine parts, calculates a function to estimate parameter values related to part life, and determines failure time or probability using this data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If scheduled inspection time is determined based on usage conditions and usage time, then inspection planning becomes standardized, but the actual degree of damage varies between machines making inspections inadequate
Solution Approach 1:
The patent changes from fixed scheduled inspection parameters to dynamic parameters based on accumulated physical quantity data. The inspection timing is adjusted according to actual machine state parameters such as vibration, temperature, and load, allowing each machine to be inspected at the appropriate time based on its actual condition rather than a universal schedule.
Solution Approach 2:
The system implements feedback by continuously monitoring physical quantity data from sensors and using this information to update the life estimation function. The accumulated data feeds back into the prediction model, enabling the system to learn from actual machine behavior and improve inspection accuracy over time.
2Measurement precision
If physical quantity data is accumulated and life estimation calculation is performed, then failure timing prediction accuracy improves, but data processing complexity increases
Solution Approach 1:
The patent creates a universal life estimation function that can be applied to multiple machines and different types of physical quantity data. The same calculation framework processes various sensor inputs (vibration, temperature, pressure) uniformly, reducing the need for machine-specific complex processing while maintaining high prediction accuracy.
Data Source
AI summary
Physical quantity data indicating a state of a predetermined part constituting an industrial machine is acquired. The acquired physical quantity data and time data indicating an acquisition time of the physical quantity data are stored in association with each other. A function for estimating a change in a parameter value correlated with a life of the predetermined part over time is calculated on the basis of the acquired physical quantity data and the time data. A failure time or a failure probability of the predetermined part is calculated using the calculated function.