Production Line Analysis Using Virtual Physical Quantity Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In production lines, it is challenging to obtain sufficient data for accurate statistical analysis due to limitations in installing measurement devices and measuring difficult physical quantities like density, stress, and concentration, which hinders the prediction of defects and operational conditions.
Innovation Solution
An analysis system that combines operational data with numerical analysis using physical models to estimate unmeasurable physical quantities, enabling more accurate correlation analysis between operational data and product defects, by integrating an operational data acquisition unit, production state analyzer, and correlation analyzer.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If measurement devices are installed to obtain operational data, then data availability for statistical analysis is improved, but installation complexity and cost increase due to harsh environments
Solution Approach 1:
The patent introduces numerical analysis as an intermediary method to obtain operational data indirectly. Instead of directly measuring physical quantities in harsh environments with sensors, the system uses numerical models to calculate and estimate these quantities based on readily available operational parameters, thus avoiding the complexity of installing measurement devices in difficult-to-access locations
Solution Approach 2:
The patent replaces physical measurement devices with computational methods. Rather than using mechanical or electronic sensors to directly detect physical quantities like temperature, stress, or flow rate in harsh environments, the system substitutes these physical measurement approaches with numerical analysis and simulation techniques that compute these quantities from operational data
2Measurement precision
If direct measurement of physical quantities is performed, then measurement accuracy is improved, but measurement capability is limited by environmental constraints
Solution Approach 1:
The patent uses numerical analysis as an intermediary to bridge the gap between unavailable direct measurements and needed physical quantity data. The numerical models act as mediators that translate readily available operational parameters into estimates of physical quantities that cannot be directly measured due to environmental constraints
Solution Approach 2:
The patent creates virtual copies of physical measurement data through numerical simulation. Instead of obtaining actual sensor readings from harsh environments, the system generates simulated data that replicates what measurement devices would detect, allowing analysis without physical presence in difficult-to-access locations
3Adaptability or versatility
If numerical analysis is used to obtain physical quantities, then adaptability to harsh environments is improved, but computational cost increases
Solution Approach 1:
The patent performs numerical analysis in advance to pre-compute physical quantities before they are needed for defect analysis. By calculating and storing these quantities beforehand based on operational data, the system avoids the need for real-time computational resources during actual defect investigation, thus reducing immediate computational cost while maintaining adaptability
Data Source
AI summary
An analysis system 10 includes: an operational data acquisition unit 11 that acquires operational data including an operational state of a production line 20; a product information acquisition unit 13 that acquires a state of a product manufactured in the production line and outputs the state of the product as product information; a production state analyzer 12 that obtains a predetermined physical quantity of the product on a basis of the operational data acquired by the operational data acquisition unit 11 and outputs the predetermined physical quantity as information on the physical quantity; and a correlation analyzer 14 that performs analysis of a correlation between the information on the physical quantity and the product information.