STOCHASTIC PERTURBATION ANALYSIS OF THERMAL FOOD PROCESSES WITH RANDOM FIELD PARAMETERS

Abstract

A first-order perturbation algorithm has been used to evaluate the effect of parameter uncertainties of the random variable and random field type on the temperature inside a can during a typical thermal sterilization process. The algorithm is based on the finite element formulation of the heat conduction equation and is considerably faster than a Monte C lo algorithm for a comparable accuracy. The perturbation algorithm is, however, only applicable when the coefficient of variation of the random parameters is smaller than 20%. In the case of random field parameters the finite elements should be smaller than half the scale of fluctuation. It was shown that, in the case of random field parameters, the magnitude of the temperature fluctuations in the can increases with increasing scale of fluctuation. If the scale of fluctuation becomes very large, the random field degenerates to a random variable and the variance of the temperature at an arbitrary position and time is maximal. For a typical sterilization process it appears that the thermophysical properties are the most important sources of variability.