Predictive Model Equations for Palm Kernel (Elaeis guneensis J.) and Sesame (Sesamum indicum L.) Oil Colour
A 3-factor experimental design was used to determine the influence of moisture content, roasting duration and temperature on palm kernel and sesame oil colours. Four levels each of these parameters were used. The data obtained were used to develop prediction models for palm kernel and sesame oil colours. Coefficient of determination R2 , probability of prediction F, and analysis of variance technique were employed to authenticate the adequacy of the models. Colour intensity increased with increase in moisture content, roasting duration and temperature of both oilseeds. Rated by lovibond unit, palm kernel oil colour varied from 6.4 to 8.8 yellow and 2.7 to 3.8 red. Sesame oil colour varied from 5.8 to 8.3 yellow and 2.3 to 3.4 red. Therefore the three parameters investigated all had significant effects on palm kernel and sesame oil colour. Coefficients of determination R2 at 95 % confidence level for palm kernel and sesame oil colours were 0.94 and 0.93 respectively. Probability of prediction F, for palm kernel oil colour was 0.92 and 0.77 was recorded for sesame oil. Estimated error of ± 0.18 and ± 0.2 are envisaged while applying the models for predicting palm kernel and sesame oil colours respectively.
Keywords: Palm kernel, Sesame, Palm kernel, Oil Colour, Process Parameters, Model.
Journal of Applied Science, Engineering and Technology Vol. 6 (1) 2006 pp. 34-38