Particle swarm optimization and ordered predictors selection applied in NMR to predict crude oil properties

Paulo, Ellisson H. de; Folli, Gabriely S.; Nascimento, Márcia H.C.; Moro, Mariana K.; Cunha, Pedro Henrique Pereira da; Castro, Eustáquio Vinícius Ribeiro de; Neto, Álvaro Cunha; Filgueiras, Paulo R.

doi:10.1016/j.fuel.2020.118462

Cited by 8 publications

(3 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The regions selected by SPA were the paraffin region (3000–2800 cm −1 ); the bands in 1454 and 1375 cm −1 , resulting from CH 3 alkane and CH 3 aliphatic; and the region 900–650 cm −1 , resulting from substitution in benzene (Figure 3A), 35 which suggests that these regions are more relevant to predict API gravity in oil. Corroborating our results, Paulo et al 36 also selected the paraffin region using nuclear magnetic resonance ( 1 H and 13 C‐NMR) spectra for predicting API gravity.…”

Section: Resultssupporting

confidence: 66%

Variable selection by permutation applied in support vector regression models

Cunha

Paulo

Folli

et al. 2022

Journal of Chemometrics

Self Cite

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

Section: Resultssupporting

confidence: 66%

Variable selection by permutation applied in support vector regression models

Cunha

Paulo

Folli

et al. 2022

Journal of Chemometrics

Self Cite

View full text Add to dashboard Cite

show abstract

“…That is to say, an initial set of particles position x id and velocity v id is generated randomly. In the optimisation of the problem, each particle with a randomised range from 0 to 1 in the particle swarm represents a possible solution [25]. After iteration and search, the optimal particle becomes the optimal solution.…”

Section: Methodsmentioning

confidence: 99%

Residual life prediction of marine EPR cable under discontinuous operation

Cai

Lin

Liao

2020

IET Generation Trans & Dist

View full text Add to dashboard Cite

It is known that marine cable is responsible for the transmission of electric energy and electrical signals of the ship. In order to predict the residual life of cable under discontinuous operation, a non‐destructive online life prediction model is established based on retention rate of hardness (RRH). This study takes the marine ethylene propylene rubber (EPR) cable as the research object. Using time temperature superposition principle, the optimal time temperature shift factor is calculated using particle swarm optimisation and steepest descent method. After that the main curve was fitted exponentially to acquire the relationship between RRH and ageing time. Compared with the existing model, the results indicate that the proposed model has higher prediction accuracy. On this basis, the residual life prediction model based on RRH of marine EPR under discontinuous operation is presented by adding time parameters. The effectiveness of the method is verified by simulation. Thus, it provides theoretical guidance for solving the problem of marine cable replacement time.

show abstract

“…There is more detailed information of these analyses as we reported in another study. 34 HTGC. The HTGC chromatograms were obtained following the ASTM D7169 (ref.…”

Section: Crude Oil Analysismentioning

confidence: 99%