No abstract

The objective of this paper is to report a study on the predictability of the steady-state and transient thermal properties of fabrics using a feed-forward, back-propagation artificial neural network system. A comparison was made with two different network architectures, one with two sequential networks working in tandem fed with a common input and another with a single network that gave two outputs. A three-layered network was used in both the cases. The networks were then subjected to a set of untrained inputs and the output thermal properties, namely thermal resistance and Qmax, were compared with the values obtained experimentally. The architecture with two networks working in tandem with a common set of inputs gave better results than the architecture with one set of inputs used to give two outputs.

No abstract

No abstract

In this paper studies conducted on Polymethyl methacrylate (PMMA) under combined bending and shear loading are described. A strong dependency of fracture surface features on the mixed mode stress state is observed. Close to pure mode I, the fracture surface is 'mirror-like' in appearance. With increasing mode II component the fracture surface becomes 'misty' and parabolic markings appear on the fracture surface. These observations indicate that the level of stress ahead of the crack tip increases with increasing mode II component. The mixed mode specimens are also observed to fracture at much higher stresses than the pure mode I specimen, contrary to the predictions of the fracture criteria based on linear elastic fracture mechanics (LEFM). The fracture surface features and the higher stresses at fracture in the mixed mode specimens are explained in terms of the increase in stiffness (which has been related to an increase in the effective stress intensity factor per unit opening displacement) with the introduction of a mode II component and the geometry of the 3-dimensional crack tip.