The use of NIR technology in sugar beet processing for real-time monitoring of key quality parameters

The automation of production processes has become an omnipresent topic. Thereby, the acquisition of data in real-time with the help of so-called smart sensors plays a crucial role. To continuously detect the traditional parameters, polarimetric and HPLC sucrose, as well as dry substance content at various points in the sugar production process in real-time, the Pfeifer & Langen IP GmbH uses near-infrared spectroscopy (NIRS). To use and sell the NIR technology as ready-to-use solutions for several liquid and solid raw, intermediate, and final products in the sugar process, the IP GmbH has developed technical installation solutions and specific calibration models over several years. The underlying mathematical models are built upon a comprehensive database as well as continuous optimisation. The varying sample matrices, as well as ambient conditions, have made the model development quite complex. As a first step, this paper focuses on NIRS sensor development for several liquid intermediate products of the beet sugar process. An excellent correlation (R2 = 0.993 and R2 = 0.982 for polarimetric and HPLC sucrose content, respectively) between the NIRS data and the reference data demonstrated the applicability of this technique to promptly reduce manual work in the laboratory. Besides, the NIRS real-time data is intended to be used for an automated model-based control of the sugar production process. Apart from the targeted basic function of these sensors, the determination of key quality parameters in real-time, additional findings were generated by evaluating the data more precisely. A detailed comparison between polarimetric and HPLC sucrose measurements enabled the identification of differences attributable to specific non-sucrose components, particularly microbial polysaccharides.