Determination of the heat transfer coefficient and thermal conductivity for coconut kernels using an inverse method with a developed hemispherical shell model

Simulation studies for improved efficiency of thermal food processes (dryers) require knowledge of the heat transfer coefficient and thermal conductivity of the food. A novel analytical solution to the heat transfer problem in a hemispherical shell has been developed using the shifted Legendre functions. This was fitted to experimental data to determine the heat transfer coefficient and thermal conductivity of coconut by an inverse method. To minimize errors the natural shape of the food was taken into account. Neither the values nor the solution have been reported to-date. The values of heat transfer coefficients (h) and thermal conductivities (k) were found to be 17 <= h(2) <= 36, 4.0 <= h(2) <= 11 and 7.0 <= h(1) <= 15, 1.2 <= h(1) <= 3.0 W/m(2) degrees C for the outer and inner surfaces, and 0.125 <= k <= 0.127 and 0.030 <= k* <= 0.020 W/m degrees C, respectively. The maximum temperature occurred at the outer equatorial surface and the minimum at the middle near to the pole. In the future these thermal parameters can be used to develop coconut dryers with improved energy efficiency and the method can be used to determine parameters for foods of similar shape. (C) 2011 Elsevier Ltd. All rights reserved.