Experimentally supported mathematical modeling of continuous baking processes

The scope of the PhD project was to increase knowledge on the process-to-product interactions in continuous tunnel ovens. The work has focused on five main objectives. These objectives cover development of new experimental equipment for pilot plant baking experiments, mathematical modeling of heat and mass transfer in a butter cookie product, and evaluation of quality assessment methods.

The pilot plant oven is a special batch oven designed to emulate continuous convection tunnel oven baking. The design, construction, and validation of the oven has been part of the project and is described in this thesis. The oven was successfully validated against a 10 m tunnel oven. Besides the ability to emulate the baking conditions in a tunnel oven, the new batch oven is designed and constructed for experimental research work.

In the design options to follow the product continuously (especially weight and temperature) and control the process (air flow, temperature, and humidity) are therefore emphasized. The oven is furthermore designed to work outside the range of standard tunnel ovens, making it interesting for manufacturers of both baking products and baking equipment.

A mathematical model describing the heat and mass transfer in butter cookies during baking was formulated. The model was solved numerically by the use of a finite element method. Model optimization and validation was successfully carried out against experimental data obtained in the new pilot plant oven.

The effect of the baking tray on mass transfer was examined through comparison of different modeling set-ups and experimental data. It was found that while the baking tray is likely to reduce the evaporation from the bottom surface, it is not correct to assume that no evaporation takes place at the covered surface.

Parallel to the construction of the pilot oven an advanced multi-spectral imaging method was investigated as a method for quality assessment of butter cookies. The ability of the method to assess multiple quality aspects from one image was the main focus of the study. The system was able to predict both the surface browning and the water content in butter cookies.