Identification of dynamic process systems with surrogate data methods

Identifying the underlying dynamics of chemical process systems from experi mental data is complicated, owing to a mixture of influences that cause err atic fluctuations in the time series. These influences call be notoriously difficult to disentangle. The development of process models is usually, sub ject to considerable human judgment and call therefore be very, unreliable. This is especially the case when the model priors arc unknown and the mode l is validated empirically such as with cross-validation or holdout methods . A case study shows that more reliable identification of systems is possib le by using surrogate methods to classify the data, as well as to validate models derived from these data.