TwinSpectra-Spectroscopic scans of refuse-derived fuel particles

TwinSpectra is a novel data comprising spectroscopic measurements of refuse-derived fuel particles sampled during the cement manufacturing process. The data consists of two alternative sets of material reflectance spectra acquired using two spectroradiometers based on distinct measurement principles. In total, 120 samples representing six particle groups are contained in the datasets, with each sample measured twice using each device. After data preprocessing and cleansing, two separate files of 718 and 720 measurements were compiled, respectively, for a total of 1438 readouts. The whole process was carried out with respect to the required calibration procedures. Eventually, all the records were assigned to one of six material groups that could be later utilized as classes. This data can serve as a solid foundation for subsequent analyses and the development of machine learning-based solutions. In particular, the datasets and the algorithms built on them should enable the automation of sustainable cement manufacturing or aid in the development of techniques for working with highly dimensional, sequential spectroscopic data.