Determination of API Gravity and Total and Basic Nitrogen Content by Mid- and Near-Infrared Spectroscopy in Crude Oil with Multivariate Regression and Variable Selection Tools

Rainha, Karla Pereira; Rocha, Júlia Tristão do Carmo; Rodrigues, Rayza R.T.; Lovatti, Betina Pires de Oliveira; Castro, Eustáquio Vinícius Ribeiro de; Filgueiras, Paulo R.

doi:10.1080/00032719.2019.1628246

Cited by 16 publications

(14 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Most selected NIR variables (Figure 6) are in the chemical bonding, harmonic transitions (overtones), and combination region (4000–8500 cm –1 ). This region contained vibrations of N–H, C–H, S–H, and O–H covalent bonds that included a hydrogen atom 4,16,42,43 . About 82%, 54%, and 67% of all variables were selected in this region for TAN, BNC, and SC, respectively.…”

Section: Resultsmentioning

confidence: 99%

“…Several papers applied linear calibration methods to all problems, assuming a linear relationship between the property of interest and the instrumental data. However, in many cases, nonlinearity can exist between the instrumental data and the property of interest 3,4,8,9,30–32,36–38 . Thus, we tested the spectroscopic data used in this study for their relationship with the property of interest, using both Durbin–Watson (dw) and Wald–Wolfowitz runs test.…”

Section: Resultsmentioning

confidence: 99%

“…The 1500-1700 cm -1 , 1240-1030 cm -1 , and 820-650 cm -1 regions indicate the N-H and O-H group concentrations, C-N stretching, and cis-trans N-H angular deformation presence, respectively. 4,16,42,43 The 1655-1989 cm -1 region containing symmetric C-H deformation and stretching of C-N and C-O was also selected. 4,16,42,43 These other bands were also chosen in the variable selection of BNC ( Figure 7B).…”

Section: Nir (B) and Mir (C)mentioning

confidence: 99%

“…4,16,42,43 The 1655-1989 cm -1 region containing symmetric C-H deformation and stretching of C-N and C-O was also selected. 4,16,42,43 These other bands were also chosen in the variable selection of BNC ( Figure 7B). The 600-700 cm -1 area shows C-S bond axial deformation vibrations, 16 and it was also chosen for the selection of SC variables ( Figure 7C).…”

Section: Nir (B) and Mir (C)mentioning

confidence: 99%

“…In this case, the MPA selection model produced more satisfactory results than the other models: UOP characterization factor (RMSEP = 0.06), nitrogen content (RMSEP[%] = 0.048 wt%), kinematic viscosity at 40°C (RMSEP[%] = 21.9%), pour point (14.58°C and 19.2°C), and the sulfur content (SC) (RMSEP% = 0.09 wt%). Rainha et al 4 determined the essential variables for API, total nitrogen and basic nitrogen content, using near‐ and mid‐infrared spectroscopy (MIR and NIR) coupled with variable selection (iPLS, siPLS, and CARS‐PLS) and PLS regression. They reported the best performance parameters for API and total nitrogen content of the models using NIR coupled to CARS‐PLS (RMSEP equal to 0.9 and 0.0275 wt% for API and nitrogen content, respectively).…”

Section: Introductionmentioning

confidence: 99%

See 4 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

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: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Section: Nir (B) and Mir (C)mentioning

confidence: 99%

Section: Nir (B) and Mir (C)mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 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

View full text Add to dashboard Cite

show abstract

Variable selection by permutation applied in support vector regression models

Cunha

Paulo

Folli

et al. 2022

Journal of Chemometrics

View full text Add to dashboard Cite

Support vector regression (SVR) can be considered as a black box machine‐learning method. Thus, identifying the cause/effect relationship and the synergism between the most important variables is a difficult task. This study demonstrates the potential of two variable selection methods by permutation—subwindow permutation analysis (SPA) and noise incorporated subwindow permutation analysis (NISPA)—to overcome this limitation. The application of these two variable selections in SVR is poorly explored in literature, mainly for regression problems. The algorithms were applied in FTIR (Fourier transform mid infrared spectroscopy) data of crude oil samples to estimate API gravity, kinematic viscosity at 50°C, saturates, aromatics, resins, and asphaltene content. The results were compared to other variable selection methods. SPA and NISPA provided the most accurate models for kinematic viscosity, saturates, and aromatic content. The root‐mean‐squared percentage error of prediction (RMSPEP) of the SPA and NISPA were, respectively, 14.26% and 14.62% for kinematic viscosity, 4.7 wt% and 4.4 wt% for saturates content, and 3.4 wt% and 3.1 wt% for aromatic content. Regarding API prediction, despite obtaining similar accuracy to the other selection methods, SPA produced a more simplified model, using only 3.5% of the 3351 total variables, with RMSEP equal to 1.0 and R2p to 0.981. Therefore, SPA and NISPA, besides obtaining, in general, faster and more accurate and parsimonious models, revealed the most important variables for building the SVR models.

show abstract

Concise review on bacterial degradation of petroleum hydrocarbons: Emphasis on Indian marine environment

Goveas

Nayak

Selvaraj

2022

Bioresource Technology Reports

View full text Add to dashboard Cite

Determination of API Gravity and Total and Basic Nitrogen Content by Mid- and Near-Infrared Spectroscopy in Crude Oil with Multivariate Regression and Variable Selection Tools

Cited by 16 publications

References 49 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

Variable selection by permutation applied in support vector regression models

Concise review on bacterial degradation of petroleum hydrocarbons: Emphasis on Indian marine environment

Contact Info

Product

Resources

About