Check Valve Abrasion Estimation via Optical Sensor and Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating the abrasion amount of a check valve in injection molding machines are inaccurate due to variations in operation conditions, requiring direct measurement that suspends production and affecting productivity.
Innovation Solution
A machine learning-based system that performs supervised learning using feature extraction from physical data during injection operations with check valves of varying abrasion amounts to estimate abrasion accurately without relying on operation conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If direct measurement of check valve dimension is performed by taking out the screw, then measurement precision is improved, but productivity deteriorates due to production suspension
Solution Approach 1:
The patent replaces the mechanical direct measurement method (taking out the screw and physically measuring) with a non-contact optical measurement system. The optical sensor detects the check valve's position and dimension through light interaction, eliminating the need to suspend production while maintaining measurement accuracy.
Solution Approach 2:
The patent introduces an optical sensor as an intermediary between the measurement system and the check valve. Instead of direct physical contact measurement, the optical sensor acts as a mediator to detect dimensional changes through optical fields, enabling indirect measurement during operation.
2Productivity
If indirect detection of backflow phenomenon is performed using load torque or advancing speed, then productivity is maintained, but measurement precision deteriorates due to influence of operation conditions
Solution Approach 1:
The patent extracts the check valve's position information as a separate measurable parameter from the complex backflow phenomenon. By directly measuring the check valve's position using optical sensors, the system isolates the key indicator of abrasion (dimensional change) from other operation condition variables, enabling accurate detection independent of load torque or advancing speed variations.
Data Source
AI summary
An abrasion amount estimation device stores a learning result obtained through supervised learning performed based on a feature amount, which is extracted from a physical amount which is acquired in injection performed by an injection molding machine, and information related to an abrasion amount of a check valve which has been attached to the injection molding machine in the injection. The abrasion amount estimation device estimates an abrasion amount of a check valve which has been attached to the injection molding machine in the injection based on the learning result which is stored and the feature amount which is extracted.


