Online detection of multiple quality parameters of finished oil products in high pressure multi-product pipeline by NIR combined with a library method

Abstract This study aims to solve simultaneously detecting multiple quality parameters of pipeline-transported refined oil, providing data support for the high-precision and intelligent operation of refined oil transportation pipeline. Based on near-infrared (NIR) spectroscopy and a multivariate analysis method of library, an NIR online detection system was designed to simultaneously online detect the multiple quality parameters of refined oil and the proportion of mixed oil transported in high-pressure pipelines. The multiple key quality parameters of different oil including flash point and solidification point of diesel, final distillation point and water content of gasoline were online detected by this system. The the online NIR detection values output simultaneously every 30 seconds, and the results were uploaded to the station control system, achieving the goal of real-time monitoring of pipeline-transported oil quality. The NIR online prediction standard deviations of flash point, final distillation point, water content, and solidification point compared with the reference values measured by the laboratory national standard method, which were 2.7°C, 1.8°C, 0.009% (v/v), and 3°C, respectively, and the online detection values were consistent with the reference values. The NIR detection values of mixed oil ratios were consistent with the prepared values. The new NIR online detection system developed in this study overcomes the technical challenges of NIR detection of flash point and gasoline final distillation point. It is of great significance for the intelligent production process of refined oil sequential transportation in pipelines.