FT-IR VGO Composition Fit Quality Assessment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for estimating the composition of vacuum gas oil (VGO) compositions using FT-IR spectra lack the ability to determine the fit quality of the predicted models, making it difficult to assess the accuracy of the compositional classes and spectral regions, which can lead to unreliable predictions.
Innovation Solution
The method involves representing the FT-IR spectrum of a VGO composition as a weighted combination of FT-IR spectra from a database, using partial least squares analysis to determine partial representations for specific compositional classes, calculating leverage and residual values, and identifying poor fit quality based on these values to ensure accurate representation and confidence in the model.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If FT-IR spectrum is represented as a weighted combination of database spectra to estimate VGO composition, then compositional analysis speed is improved, but fit quality determination capability deteriorates
Solution Approach 1:
The patent segments the FT-IR spectrum into multiple spectral regions, each associated with specific compositional classes (e.g., paraffins, aromatics, naphthenes). By performing partial least squares analysis on individual spectral regions rather than the entire spectrum, the method maintains analytical speed while enabling targeted fit quality assessment for each compositional class, thereby resolving the contradiction between productivity and measurement precision.
2Loss of time
If multivariate analytical techniques are used to model compositional properties, then analysis time is reduced, but reliability of composition modeling deteriorates
Solution Approach 1:
The patent introduces a feedback mechanism by calculating leverage and residual values for each spectral region and compositional class. These values provide quantitative feedback on the quality of the multivariate model fit, allowing operators to assess reliability and determine when additional analytical techniques are needed. This feedback system maintains the speed advantage of multivariate techniques while ensuring modeling reliability through systematic quality assessment.
3Loss of information
If comprehensive spectral analysis is performed on all spectral regions, then complete compositional information is obtained, but complexity of the analysis process increases
Solution Approach 1:
The patent divides the complex FT-IR spectrum into multiple manageable spectral regions, each associated with specific compositional classes. This segmentation allows the analysis process to focus on relevant spectral regions for each compositional class, maintaining information completeness while reducing overall analysis complexity through structured, region-specific processing.
Solution Approach 2:
The patent applies local quality analysis by performing partial least squares analysis on individual spectral regions rather than treating the entire spectrum uniformly. Each spectral region is analyzed with appropriate weighting and assessment criteria tailored to its associated compositional classes, improving analytical efficiency while maintaining comprehensive compositional information through the aggregation of regional analyses.
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
This approach provides a reliable indication of fit quality for the model of composition, allowing operators to assess the accuracy of predictions and refine the model using known bulk properties, thereby enhancing confidence in the compositional analysis.
Implementation Method 1
representing a FT-IR spectrum of a first vacuum gas oil (VGO) composition
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Systems and methods for implicit chemical resolution of vacuum gas oils and fit quality determination are disclosed. The systems and methods include utilizing an FT-IR spectrum of an unknown VGO composition, and a database of FT-IR spectra of known VGO compositions, to determine a model of composition for the unknown VGO composition. Additionally, the fit quality for the model of composition is determined by performing a partial least squares analysis on specific spectral regions of interest in the FT-IR spectrum of the unknown VGO composition.