关键词: 时滞细胞神经网络, 线性矩阵不等式, 全局稳定, entropy generation, velocity profile, duct, arbitrary cross section, laminar internal flow

Abstract: The objective of this work is to estimate the accuracy of a predicted velocity profile which can be gained from experimental results, in comparison with the exact ones by the methodology of entropy generation. The analysis is concerned with the entropy generation rate in hydrodynamic, steady, laminar, and incompressible flow for Newtonian fluids in the insulated channels of arbitrary cross section. The entropy generation can be calculated from two local and overall techniques. Adaptation of the results of these techniques depends on the used velocity profile. Results express that in experimental works, whatever the values of local and overall entropy generation rates are close to each other, the results are more accuracy. In order to extent the subject, different geometries have been investigated. Also, the influence of geometry on the entropy generation rate is studied, and the distribution of volumetric local entropy generation rate for the selected geometries is drawn.

Key words: delayed cellular neural networks (DCNNs),linear matrix inequality (LMI), global stability, entropy generation, velocity profile, duct, arbitrary cross section, laminar internal flow