Estimation of the apparent thermal diffusivity coefficient using an inverse technique

Abstract

In this work, an inverse heat conduction problem is solved and optimization methods are used to estimate the variable apparent thermal diffusivity coefficient of bananas during the drying process. Thermal diffusivity is the transport property that controls conduction heat transfer in a transient regime. During banana drying, the thermal diffusivity modifies with the drying time, i.e. with the temperature and moisture content. The estimation is based on transient temperature measurements taken by a thermocouple on the inner part of the banana on which the heat flow occurs. The inverse problem is solved as an optimization problem where the objective function is minimised by two optimization methods: quasi-Newton (Broyden–Fletcher–Goldfarb–Shanno algorithm) and differential evolution. Numerical experiments are used to test the proposed mathematical model of the measurement of thermal diffusivity coefficient. Specifically, statistical analysis showed no significant differences between reported and estimated curves using the differential evolution method, unfortunately the quasi-Newton method performs poorly.