Determination of diesel quality parameters using support vector regression and near infrared spectroscopy for an in-line blending optimizer system

Recent Advances in the Characterization of Gaseous and Liquid Fuels by Vibrational Spectroscopy

“…The best performance was found for the combination of Raman spectroscopy with artificial neural network data analysis. In order to develop an online tool for monitoring and optimizing the blending of diesel fuel, Alves et al [103] applied support vector regression (SVR) to NIR spectra. They determined the flash point and the cetane number, and the results were in good agreement with their conventional reference method.…”

Combination of data rectification techniques and soft sensor model for robust prediction of sulfur content in HDS process

“…They indicated that the SVR is an attractive alternative to artificial neural networks for the development of soft-sensors in bioprocesses. Julio Alves et al [9] demonstrated the application of SVR to determine the quality parameters of diesel for an in-line blending optimizer system in a petroleum refinery. They determined flash point and cetane number using SVR and the results were compared with those obtained by using the partial least squares (PLS) algorithm.…”

Classification of diesel pool refinery streams through near infrared spectroscopy and support vector machines using C-SVC and ν-SVC

“…The final stage of diesel oil production is the mixing of these various streams in adequate proportions at an in-line static mixer to obtain the final diesel blend in accordance with specifications using the streams most suitable available and considering aspects related to economics, logistics and marketing issues in order to avoid product quality giveaway [2]. The contributions of diesel pool streams to obtain the final blend with the required specifications is usually controlled by a real time optimization (RTO) system [2,5,6] that receives information from an on-line analyzer about the quality of samples obtained downstream of the in-line static mixer [7] and using a calibrated model can modify the contributions of the different streams according to indicated need considering a criterion related to diesel quality.…”

“…For determinations of quality parameters of the different streams and the final blend product in petroleum refineries or petrochemical plants [1][2][3][4]8] the process analytical chemistry -PAC [9] can be applied using some analytical method based on techniques such as chromatography [8,10], mass spectrometry [11] and near infrared (NIR) spectroscopy [4,7,10,[12][13][14] usually combined with some chemometric method. As a simpler, faster and cheaper alternative a classification model can be used to identifying the different streams and the final blend product (which must match the blend specification) by means of on-line near infrared spectroscopy analysis.…”