The main objective of this project is to develop an application where physical models of fibre lines are run in parallel to the real processes. This application can be useful as a diagnostic tool to detect faults and improve process operations. By running the simulation models continuously, feeding input data from the process data base at a pre-determined time interval, the simulation will show results from the “normal operations”. These simulation results are compared to the measured data and hence faults can be detected e.g. hang-ups in the digester and channelling. NIR-spectra lab measurements of the wood chips fed into the digester can as well be considered. This feature give us the possibility of correlating the quality of the obtained pulp to the raw material used, which can be of help when tuning the process parameters.
In this paper results from the use of the application described above in several mills in South Africa and results from some preliminary tests in a mill in Sweden are presented. Moreover, the system structure for the interaction between the physical models designed in Modellica language and the real process data is described.