A simplified flowchart of the industrial SRMG polymerization process

All the data samples have been collected from daily process records and the corresponding laboratory analysis. Melt Index (MI) is sampled and analyzed in the lab once a day, always in the morning. Owing to lack of online analyzers for MI, the operating variables have to be manipulated until the assay results are available. Therefore, efficient soft sensors for MI in SRMG (Sequential-Reactor-Multi-Grade) processes are necessary. As for this industrial SRMG process, there are 12 input variables in the 1st reactor, 8 input variables in the 2nd reactor, and 9 variables in the 3rd reactor. MI1, MI2 and MI3 are for three sequential reactors, respectively. Besides, there are 3 main product grades for this production line.

In this data set (download), the first 1-249 samples are used for training and the rest (250-331) for testing in our published paper. Note that the data set can be utilized as a public one only for academic research. If you use this data set, please cite the following article. Y. Liu, Z.L. Gao, and J. Chen, Development of Soft-Sensors for Online Quality Prediction of Sequential-Reactor-Multi-Grade Industrial Processes, Chemical Engineering Science, 2013. Any questions? Please send an email to jason@xxxwavenet.cycu.edu.tw (please remove the xxx for a correct e-mail address)