Coupling of PLS method and mechanistic model for monitoring and controlling the industrial production of fluorinated polymers

The performance of a process can be monitored by comparing the experimental results to the predictions from a mechanistic model or by using multivariate statistical projection methods such as principal component analysis (PCA) and projection to latent structures (PLS). Both approaches have advantages and drawbacks. In order to overcome such drawbacks, in this work an attempt to merge mechanistic and statistical models has been made.