This standard aims to address the accurate identification of post-consumer recycled polyethylene terephthalate (PET) plastics. It specifies an innovative testing method: analyzing the differences in volatile components in materials using headspace gas chromatography-mass spectrometry (HS-GC-MS) and constructing a predictive model using the Random Forest machine learning algorithm to effectively distinguish between "post-consumer mechanically recycled PET plastics" and "virgin PET plastics." This method is primarily used for quality identification of recycled materials in the form of flakes or pellets produced from recycled PET packaging bottles through physical processing. Unlike traditional material identification methods (such as infrared spectroscopy), this standard focuses on more complex "source identification," verifying the recycled identity of materials through chemical fingerprint characteristics. It is not applicable to recycled plastics obtained through chemical recycling processes. Its development provides a key technical basis for regulating the recycled plastics market and combating the practice of passing off virgin materials as recycled ones.