Determination of Critical Conditions for Puncturing Almonds Using Coupled Response Surface Methodology and Genetic Algorithm

Mahmood Mahmoodi-Eshkaftaki, Rahim Ebrahimi* and Mehdi Torki-Harchegani

Department of Agricultural Machinery Engineering, Faculty of Agriculture, University of Shahrekord, P.O. Box 115, Shahrekord, Iran

Article history:
Received January 29, 2013
Accepted September 3, 2013

Key words:
almond, genetic algorithm, mechanical properties, modelling, response surface methodology, Weibull distribution

Summary:
In this study, the effect of seed moisture content, probe diameter and loading velocity (puncture conditions) on some mechanical properties of almond kernel and peeled almond kernel is considered to model a relationship between the puncture conditions and rupture energy. Furthermore, distribution of the mechanical properties is determined. The main objective is to determine the critical values of mechanical properties significant for peeling machines. The response surface methodology was used to find the relationship between the input parameters and the output responses, and the fitness function was applied to measure the optimal values using the genetic algorithm. Two-parameter Weibull function was used to describe the distribution of mechanical properties. Based on the Weibull parameter values, i.e. shape parameter (ß) and scale parameter (η) calculated for each property, the mechanical distribution variations were completely described and it was confirmed that the mechanical properties are rule governed, which makes the Weibull function suitable for estimating their distributions. The energy model estimated using response surface methodology shows that the mechanical properties relate exponentially to the moisture, and polynomially to the loading velocity and probe diameter, which enabled successful estimation of the rupture energy (R²=0.94). The genetic algorithm calculated the critical values of seed moisture, probe diameter, and loading velocity to be 18.11 % on dry mass basis, 0.79 mm, and 0.15 mm/min, respectively, and optimum rupture energy of 1.97·10^–3 J. These conditions were used for comparison with new samples, where the rupture energy was experimentally measured to be 2.68 and 2.21·10^–3 J for kernel and peeled kernel, respectively, which was nearly in agreement with our model results.

*Corresponding author:          This email address is being protected from spambots. You need JavaScript enabled to view it.
                                               ++98 913 381 5761
                                               ++98 381 442 4428