A switched Kalman filter dedicated to assisted pressure food thawing

Foodstuffs' thawing under the effect of high pressure is an innovative technique, which considerably increases the dynamics of the process, while limiting subsequent food damage by controlling the melting kinetics. The principle consists of lowering the solid/liquid phase change temperature of water, the main component of foods, with pressure, which allows a considerably larger entering heat flow. To reach this objective, the knowledge of the melting kinetics is needed, either measured or estimated. Because of constraints prohibiting the use of invasive sensors, we propose to estimate the internal temperatures of the foodstuff from measurements on the surface. The first stage to achieve this aim is to model the heat transfer with water phase shift under the couple temperature/pressure. One particularly efficient approach consists in representing these phenomena by means of switched linear hybrid systems, where temperature is a continuous variable and pressure is an event variable. The second stage proposed in this work is internal temperature estimation, using an adaptation of the well-known Kalman filter, applied on the switched linear hybrid system. This technique is carried out during the experimental thawing of a food gelatin, in a pressure vessel developed by Alstom Company.