Abstract
A new measurement system is presented, that delivers online data from a mashing process. By Tucker3 decomposition it is shown that the data contain relevant information about the quality of the employed malt. In contrast to bilinear models the Tucker3 preserves the three-dimensional character of batch data. Individual factors for variable, batch and time are obtained and can be analysed separately. An important pre-processing step is the synchronisation of the data. Dynamic Time Warping (DTW) is a suitable method that brings data sets to common length and aligns the data according to its form.
| Original language | English |
|---|---|
| Pages | 21-25 |
| Number of pages | 5 |
| State | Published - 2006 |
| Externally published | Yes |
| Event | 2006 4th International Conference on Simulation and Modelling in the Food and Bio-Industry, FOODSIM 2006 - Naples, Italy Duration: 15 Jun 2006 → 17 Jun 2006 |
Conference
| Conference | 2006 4th International Conference on Simulation and Modelling in the Food and Bio-Industry, FOODSIM 2006 |
|---|---|
| Country/Territory | Italy |
| City | Naples |
| Period | 15/06/06 → 17/06/06 |
Keywords
- Dynamic time warping
- Malt quality
- Mashing
- Process analysis
- Tucker3