Abstract: To adjust the reaction conditions of LTAG process, it is necessary to monitor the hydrocarbon compositions of feedstock LCO and products of hydrotreating unit. In this work, MIR calibration model was established by using the mid-infrared spectra data of a large number of LCO and hydrotreated diesel samples and partial least squares method to analyze rapidly and accurately the hydrocarbon compositions of LCO feedstock and products of hydrotreating unit in LTAG process. The MIR method has good consistency with mass spectrometry results. Compared with near infrared fast analysis method, MIR method performs higher accuracy. The method is simple and repeatable, and meets the needs of field fast analysis.

Key words: LTAG, LCO, mid infrared spectrum, partial least squares