Determination of Saturates, Aromatics, and Polars in Crude Oil by <sup>13</sup>C NMR and Support Vector Regression with Variable Selection by Genetic Algorithm

Filgueiras, Paulo R.; Portela, Natália A.; Silva, Samantha R.C.; Castro, Eustáquio Vinícius Ribeiro de; Oliveira, Lize M. S. L.; Dias, Júlio Cesar; Neto, Álvaro Cunha; Romão, Wanderson; Poppi, Ronei J.

doi:10.1021/acs.energyfuels.5b02377

Cited by 45 publications

(29 citation statements)

References 33 publications

(60 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The methodology applied for NMR has some references, such as Lovatti et al 21 Vieira et al 3 Duarte et al 22 and Filgueiras et al 8,23 1 H NMR spectra were acquired at a temperature of 25°C and a Varian VNMRS‐400 instrument with a magnetic field of 9.4T and a 5 mm 1 H/ 19 F/ 13 C broadband probe. For sample preparation, a crude oil mass, corresponding to 5% of the final solution volume in deuterated dichloromethane (CD 2 Cl 2 ), was dissolved in a 600‐μL total volume.…”

Section: Methodsmentioning

confidence: 99%

“…A variable number limit was defined as ranging from 2 to 50 variables, and each variable group was randomly exchanged to calculate the RMSECV (Equation ). Next, the variable set with the lowest RMSECV value was adopted for SVR modeling 8

italicRMSECV = \sqrt{\frac{\sum_{i = 1}^{n} {((), y_{i} - {\overset{false^}{y}}_{i})}^{2}}{n}},

where y i ,

{\overset{false^}{y}}_{i}

, and n are the physicochemical property reference value, physicochemical estimated value by the regression model, and the number of samples, respectively.…”

Section: Methodsmentioning

confidence: 99%

“…The subset with the lowest RMSECV is chosen to create the final model. For this current work, the RMSECV calculation was adapted for the SVR machine learning model 8 . After these procedures, the final SVR model was generated for each property.…”

Section: Methodsmentioning

confidence: 99%

See 2 more Smart Citations

Variable selection in support vector regression using angular search algorithm and variance inflation factor

Folli

Nascimento

Paulo

et al. 2020

Journal of Chemometrics

Self Cite

View full text Add to dashboard Cite

Here, we combine angular search algorithm and variance inflation factor (ASA-VIF) with support vector regression (SVR) (ASA-VIF-SVR) to estimate total acid number (TAN), basic nitrogen content (BNC), and sulfur content (SC) in Brazilian crude oils. To prevent the interference of outliers, we further developed a strategy for outlier identification and applied it to nonlinear models based on RMSE (root mean square error). ASA-VIF-SVR was applied to near-and mid-infrared spectroscopy (NIR and MIR) and hydrogen nuclear magnetic resonance (1 H NMR) spectroscopy data available in a range of 93-194 samples. The models were evaluated for accuracy (root mean square error of calibration [RMSEC] and root mean square error of prediction [RMSEP]) and linearity (coefficient of determination, R 2). The removal of outliers increased accuracy and linearity of our models. The ASA-VIF model for TAN, BNC, and SC selected 0.37%, 0.93%, and 0.30% of variables from full NIR spectra; 0.21%, 0.27%, and 0.21% from full MIR; and 0.20%, 0.42%, and 0.15% from full 1 H NMR. In most cases, the best results were obtained with variable selection compared with the full dataset. Also, 1 H NMR generated more accurate and linear models with RMSEP and R 2 p of 0.0071 wt% and 0.86 for BNC and 0.0623 wt% and 0.79 for SC. TAN showed a better MIR result with RMSEP of 0.1426 mg KOH g-1 and R 2 p of 0.47. The most important region for 1 H NMR and MIR was the one with the largest quantity of unpaired electrons (aromatic region).

show abstract

Section: Methodsmentioning

confidence: 99%

italicRMSECV = \sqrt{\frac{\sum_{i = 1}^{n} {((), y_{i} - {\overset{false^}{y}}_{i})}^{2}}{n}},

where y i ,

{\overset{false^}{y}}_{i}

, and n are the physicochemical property reference value, physicochemical estimated value by the regression model, and the number of samples, respectively.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Variable selection in support vector regression using angular search algorithm and variance inflation factor

Folli

Nascimento

Paulo

et al. 2020

Journal of Chemometrics

Self Cite

View full text Add to dashboard Cite

show abstract

“…Data acquired using C-13 nuclear magnetic resonance (NMR) spectroscopy were evaluated using SVR with variable selection by genetic algorithm (GA) to determine the contents of saturated, aromatic, and polar compounds in crude oil. The use of small amounts of samples was highlighted by the authors 39 . Sulfur content in petroleum derivatives was determined using MIR spectroscopy and PLS in association with variable selection methods, among them iPLS, siPLS, uninformative variable elimination (UVE), and GA 40 .…”

Section: Fuelmentioning

confidence: 99%

Chemometrics in analytical chemistry – an overview of applications from 2014 to 2018

et al. 2019

View full text Add to dashboard Cite

A compilation of papers published between 2014 and 2018 was evaluated. Many papers related to multivariate calibration and classification have been reported, as well as, design of experiments applications and artificial intelligence methods. Some applications were highlighted, as medical and pharmaceutical, food analysis, fuels, biological and forensic for the chemometric techniques on this review. Most studies are related to developing methods for practical solutions in industry or routine analysis. A promising scenario is shown considering the number of published papers: a total of 832 for this period using the keywords, multivariate classification, multivariate calibration, analysis, chemometrics, prediction, analytical chemistry, artificial neural networks (ANN), design of experiments (DoE) and factorial design. An useful overview for Analytical Chemistry researchers´ combined with Chemometrics is presented in this review.

show abstract

“…Saturated, aromatic and polar (resins and asphaltenes) compounds were obtained by preparative column chromatography on silica gel. 18 The crude oil (200 mg) was dissolved in n-hexane and then adsorbed onto the surface of activated silica gel (230 to 400 mesh; SiliCycle® Inc., Quebec, Canada). 13 Particle sizes of 230 to 400 mesh (10 g), 100-200 mesh (5 g) and 70-30 mesh (5 g) were activated at 120 °C for 12 hours and poured into a glass column (250 x 18 mm).…”

Section: Sap Fractionsmentioning

confidence: 99%

Use of the Comprehensive Two-Dimensional Gas Chromatography (GC×GC-qMS) to Characterize The Classes of Saturated Compounds in Brazilian crude Oils

Silva

Borel²,

Puziol³

et al. 2018

Rev. Virtual Quim

View full text Add to dashboard Cite

http://rvq.sbq.org.br Uso da Cromatografia Gasosa Bidimensional Abrangente (GC×GC-qMS) para Caracterizar as Classes de Compostos Saturados em Petróleos Brasileiros Resumo: A produção de petróleo brasileiro na camada do pré-sal consiste em óleos com alto teor de compostos saturados. A identificação de compostos parafínicos e naftênicos nesses petróleos é fundamental para entender seu comportamento reológico. No presente estudo, cromatografia gasosa bidimensional abrangente foi aplicada para avaliar a capacidade de separação destes componentes quando comparada a cromatografia gasosa monodimensional. Consequentemente, foi possível estimar de forma relativa as várias classes existentes na fração saturada do petróleo. Os resultados indicam que os compostos naftênicos (53-77%) são mais abundantes que as n-parafinas e as iso-parafinas (22-46%) nas frações apolares de petróleo em amostras de média a baixa densidade, de acordo com o Instituto Americano do Petróleo (API). Óleos leves apresentaram alta quantidade de iso-parafinas (37%) e a segunda classe mais abundante foi os compostos monocíclicos (30%). Petróleos com alto ponto de fluidez apresentaram maior quantidade de n-parafinas pesada que petróleos com baixo ponto de fluidez, sugerindo uma correlação entre o ponto de fluidez e as n-parafinas de alta massa molar.

show abstract

Determination of Saturates, Aromatics, and Polars in Crude Oil by ¹³C NMR and Support Vector Regression with Variable Selection by Genetic Algorithm

Cited by 45 publications

References 33 publications

Variable selection in support vector regression using angular search algorithm and variance inflation factor

Variable selection in support vector regression using angular search algorithm and variance inflation factor

Chemometrics in analytical chemistry – an overview of applications from 2014 to 2018

Use of the Comprehensive Two-Dimensional Gas Chromatography (GC×GC-qMS) to Characterize The Classes of Saturated Compounds in Brazilian crude Oils

Contact Info

Product

Resources

About

Determination of Saturates, Aromatics, and Polars in Crude Oil by 13C NMR and Support Vector Regression with Variable Selection by Genetic Algorithm

Cited by 45 publications

References 33 publications

Variable selection in support vector regression using angular search algorithm and variance inflation factor

Variable selection in support vector regression using angular search algorithm and variance inflation factor

Chemometrics in analytical chemistry – an overview of applications from 2014 to 2018

Use of the Comprehensive Two-Dimensional Gas Chromatography (GC×GC-qMS) to Characterize The Classes of Saturated Compounds in Brazilian crude Oils

Contact Info

Product

Resources

About

Determination of Saturates, Aromatics, and Polars in Crude Oil by ¹³C NMR and Support Vector Regression with Variable Selection by Genetic Algorithm