The
content distribution difference of organic impurities entering a
polyester macromolecular structure is an identification basis of protogenetic
polyester and regenerated
polyester generated through a chemical recycling method. A method for identifying the regenerated polyester generated through the chemical recycling method comprises the steps of depolymerizing the regenerated polyester generated through the chemical recycling method and the protogenetic polyester by adopting a methanolysis method, and extracting alcoholysate of trace organic impurities embedded into the polyester macromolecular structure; obtaining detection
signal data by adopting HPLC (
High Performance Liquid Chromatography), and selecting detection
signal data of a para orientation object of
methyl benzoate; merging the detection
signal data into a variable matrix, carrying out data preprocessing, then carrying out dimension reduction
processing by adopting a PCA (Principle
Component Analysis), extracting 2 to 6 principle components as characteristic variables, building a BP (Back-Propagation)
artificial neural network to carry out a training and
simulation verification by taking the obtained characteristic variables as input variables and protogenetic attributes and regenerated attributes of
fiber as target variables, and identifying a to-be-identified sample to be the regenerated polyester generated through the chemical recycling method when a predicted value of the to-be-identified sample is matched with an attribute predicted value of the regenerated polyester generated through the chemical recycling method. The method disclosed by the invention has the advantages that the accurate rate is high, and convenience and quickness are realized.