Topological Spectral Analysis for Product Characterization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing topological spectral analysis methods require a large number of standards for database creation, are limited by the need for strong correlations between spectra and properties, and struggle with nonlinear relationships, leading to reduced precision and stability, especially when dealing with new samples or processes.
Innovation Solution
A method that creates an expanded spectral database using a limited number of standards, eliminates 'polluting' wavelengths, generates synthetic standards (intergerms and extragerms) independently of chemical properties, and allows for characterization with a reduced number of standards, enabling more efficient and accurate analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a large number of standards are used to create the spectral database, then the precision and stability of product characterization are improved, but the complexity and time required for database creation and maintenance increase significantly
Solution Approach 1:
The patent extracts and eliminates 'polluting wavelengths' from the spectral database - these are wavelengths that contribute noise or interference rather than useful information. By removing these detrimental elements, the database becomes simpler while maintaining or improving characterization precision.
Solution Approach 2:
The patent generates synthetic standards (intergerms and extragerms) that replicate the spectral characteristics of actual standards. These synthetic copies expand the effective database size and improve precision without requiring proportional increases in physical standards, thus reducing complexity.
2Reliability
If a large number of standards are used to create the spectral database, then the stability of analysis results is improved, but the time required for database creation and model updates increases
Solution Approach 1:
Synthetic standards (intergerms and extragerms) are generated computationally to expand the database. This copying approach allows the database to achieve stability with fewer physical standards and enables faster updates since synthetic standards can be generated immediately without requiring additional physical samples.
Solution Approach 2:
The patent performs preliminary elimination of polluting wavelengths during database creation. This upfront action prevents these wavelengths from degrading result stability later and reduces the need for subsequent database revisions when new samples are encountered.
3Adaptability or versatility
If the spectral database is expanded with more standards, then the ability to handle new samples and processes is improved, but the difficulty of database maintenance and model revision increases
Solution Approach 1:
When new samples or processes need to be handled, the system generates synthetic standards (extragerms) that extend the database coverage. This copying mechanism provides adaptability to new situations without requiring manual collection and integration of numerous new physical standards, thus maintaining ease of database maintenance.
Solution Approach 2:
The elimination of polluting wavelengths creates a cleaner, more robust database foundation that is less sensitive to variations in new samples. This extraction of harmful elements improves the database's inherent adaptability while reducing the maintenance burden.
4Ease of manufacture
If classical mathematical regression methods are used for spectral analysis, then the implementation is straightforward, but the precision is reduced due to inability to handle nonlinear relationships
Solution Approach 1:
The patent generates synthetic standards that capture nonlinear spectral relationships. These synthetic copies allow the system to model complex nonlinear behavior while maintaining a straightforward database structure, thereby improving precision without sacrificing implementation simplicity.
Solution Approach 2:
The patent transforms the spectral data by eliminating polluting wavelengths and organizing it into a refined database structure. This parameter change in the data representation enables more accurate modeling of nonlinear relationships while keeping the analysis approach straightforward.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables characterization of products with a significantly reduced number of standards, improving precision and stability, and allowing for faster adaptation to new samples and processes without the need for extensive database revisions.
Implementation Method 1
measuring the absorption Dix of said material at more than one wavelength in the region of 600 to 2600 nm
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to a method for characterising a target product, comprising the following steps: forming a bank of spectral data comprising samples, having measured characteristics and spectra; performing a spectral analysis of the target product and comparing the spectrum obtained with the spectral data in the data bank; identifying the "near neighbour" points of the target product; and performing a topological calculation of the characteristic of the target product as a function of the corresponding characteristics of the near neighbour points, based on a weighting linked to the inverse of the distance between the target product and the near neighbour points.