Infrared Spectroscopy Octane Cetane Prediction Model

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for determining performance properties of petroleum distillate fractions, such as Research Octane Number (RON) and cetane number, from optical spectral data are imprecise due to the complex relationship between performance properties and optical spectra, and are not transferrable across different refineries with varying chemical slates.

Innovation Solution

Representing the optical spectrum of an unknown composition as a weighted combination of known spectra from a database to generate a model of composition, which includes the compositional profile and relative abundance of components, and using this model to calculate performance properties like RON and cetane number, while determining the fit quality of the model to ensure confidence in the results.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If bulk property measurement methods are used to determine performance properties, then measurement reliability is improved, but resource consumption increases

Engineering Contradiction:

Improvemeasurement reliabilityVSAvoidresource consumption

Core Design Contradiction:

ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent replaces traditional bulk property measurement methods (mechanical/chemical testing) with optical spectroscopy (infrared or Raman spectroscopy). The system uses spectral data combined with machine learning models to predict performance properties like octane number and cetane number, eliminating the need for resource-intensive bulk testing while maintaining measurement reliability.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent creates a virtual model of the petroleum distillate fraction using spectral fingerprints and machine learning algorithms. This digital copy allows prediction of performance properties without physically testing the actual sample, reducing resource consumption while preserving measurement accuracy.

Inventive Principle:

Principle #26Copying

2Use of energy by moving object

If spectral information methods are used to determine performance properties, then resource consumption is reduced, but measurement precision deteriorates

Engineering Contradiction:

Improveresource consumptionVSAvoidmeasurement precision

Core Design Contradiction:

Use of energy by moving objectVSMeasurement precision

Solution Approach 1:

The patent performs preliminary actions by building and training machine learning models using extensive spectral data from known samples before analyzing unknown samples. This pre-processing and model training phase enables the system to achieve high measurement precision when predicting performance properties of new samples, overcoming the inherent limitations of direct spectral analysis.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent transforms the raw spectral data into meaningful performance property predictions by changing the analysis parameters through machine learning algorithms. The system learns complex non-linear relationships between spectral features and performance properties, converting imprecise spectral information into accurate predictions of octane number, cetane number, and other performance metrics.

Inventive Principle:

Principle #35Parameter changes

3Adaptability or versatility

If existing spectral methods are applied across different refineries, then versatility is improved, but measurement precision deteriorates due to varying chemical slates

Engineering Contradiction:

Improvecross-refinery applicabilityVSAvoidprediction accuracy

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent implements dynamic adaptability by enabling the machine learning model to be retrained or recalibrated for each specific refinery's chemical slate. The system can adapt to varying feedstocks, processing conditions, and product compositions at different refineries, maintaining high measurement precision across diverse operational contexts through localized model training.

Inventive Principle:

Principle #15Dynamics

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 allows for accurate and reliable determination of performance properties of unknown petroleum distillate fractions, providing enhanced confidence in the calculated values and indicating when alternative assays may be necessary, thus overcoming the limitations of existing methods.

Implementation Method 1

obtaining an infrared spectrum of the unknown composition

Methodology Applied

Scientific EffectInfrared spectroscopy: Absorption Spectroscopy

Data Source

PatentEP3861321B1Method of determining octane number of naphtha and of determining cetane number of diesel fuel or jet fuel using infrared spectroscopy

Publication Date: 2023.04.26 EXXONMOBIL TECHNOLOGY & ENGINEERING CO

Figure 1
Figure 2

AI summary

Systems and methods for determining performance properties of an unknown composition are disclosed. The performance properties can include a Research Octane Number (RON), a Motor Octane Number (MON), and/or a cetane number. The systems and methods include utilizing an optical spectrum of an unknown composition to determine a model of composition, where the model of composition includes a molecular identity and a relative abundance for components therein. The model of composition is then utilized to calculate one or more performance properties, 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 optical spectra of the unknown composition.