Neural Network Modeling of Parallel-Plain Fin Heat Sink
Wafa Batayneh, Hala Khalaf, Bahgat G. Sammakia

Abstract
The paper aims at optimizing the heat sink dimensions by maximizing the heat dissipation and minimizing thermal resistance and pressure drop. In this paper, a Neural network model is built for a parallel-plain fin heat sink. The model is developed using an experimental data from the literature. In addition, a quadratic model equation of the affecting parameters is constructed and analyzed using Response Surface Methodology for determining the important factors affecting the performance of the heat sink, and the quadratic effect of every factor by using design of experiment, analysis of variance and regression analysis. The results of the neural network model are compared with the experiment and it is shown that the error does not exceed 13.54%. This value is considered small and acceptable for such system.

Full Text: PDF