Digital Strategies to Improve the Product Quality and Production Efficiency of Fluorinated Polymers: 3. AI-Assisted Modeling of Industrial Dispersion Polymerization of TFE/PPVE/HFP

Modeling and simulation have provided a platform for virtual industrial production process optimization with reduced cost and time. Artificial intelligence (AI), combined with modeling and simulation, could help accelerate model development and improve model prediction capability. This work first reports kinetic modeling of tetrafluoroethylene (TFE) with perfluoropropyl Vinyl Ether (PPVE), and hexafluoropropylene (HFP) in a dispersion polymerization system at an industrial scale. Furthermore, AI tools with reasoning functions are utilized for model development. The model parameters, such as critical chain lengths, partition coefficients, and Henry coefficients, are estimated by AI, which significantly reduces unknown parameters for model validation with industrial data and improves model prediction capability. The model can provide deep insight into the production process and predict the key product properties such as monomer conversion, composition, and molecular weight.