ONLINE PREDICTION OF COMPOUND PROPERTIES THROUGH THE APPLICATION OF STATISTICAL PROCESS MODELS FOR THE INTERNAL MIXER

It is shown that statistical process models for the mixing process in the internal mixer, which are based on characteristical variables, are essentially suitable for on-line calculation of compound properties o f relevance for quality. In order to identify the influences of indivi dual process parameters and any disturbance variables that may prevail on specific quality characteristics, ii is, however; first process an alyses. By using characteristic variables calculated from recorded pro cess curves, it is possible to record process. The current stage of th e research into the modelling of the mixing process in an internal mix er with the aid ed with examples for a natural rubber and an SBR compo und.