Neural Network Based Analysis of Thermal Properties Rubber Composite Material - Pneumatic Tire.

The evolution of pneumatic tires has been alongside the evolution of the automobiles. The demand of the modern automotive industry has been driving the tire industry to come with high performance tire. The tire construction and geometry are very complex in nature especially tire design and stress analysis are very difficult. The study of tire performance and deformation are very challenging owing to the non-linearity associated with geometry as well as composition of material. The tire material is a cord-rubber composite, its properties anisotropic in nature. Failure analysis of cord-reinforced rubber composite tires may be useful to predict the lifetime of a tire. In this background, the present attempt is to analyze the tire using artificial neural network. The shear modulus and the temperature are measured against various frequencies. The above properties are analysed using artificial neural network. The study has been undertaken using MATLAB software. The results were compared with those of dynamic moduli master curves obtained through frequency-temperature reduction of data measured by a commercial dynamic mechanical thermal analyser (DMTA), by scanning temperature at various frequencies in the range 0.3-30 Hz. The results obtained by DMTA are trained in the Neural Network. Very good agreement of the data obtained by the two different approaches was found. [ABSTRACT FROM AUTHOR]