Sayyed Saeed Mir ghasemi - Department of Mechanical Engineering, MUT, Iran;
Mohammad Abbasi - Department of Mechanical Engineering, MUT, Iran;
Mehdi Tajdari - Department of Mechanical Engineering, Azad University of Arak

The objective of this paper is to use artificial neural networks (ANNs) for heat transfer analysis in gun barrels. This method is based on an inverse algorithm which applies to estimate the unknown time-dependent heat flux at the inner surface of gun barrel. A data set evaluated experimentally is prepared for processing with the use of neural networks. While knowing the temperature history at the measuring position, no prior information is needed on the functional form of the unknown heat flux. The temperature data calculated from the direct problem are used to train the networks.This paper employs the Back Propagation algorithm, the most common learning method for ANNs, to train and test the network. The neural network structure used in this work is a 2-layer perceptron. Results show that an excellent estimation on the timedependent heat flux can be obtained for the case considered in this study.

Artificial neural networks, Inverse heat conduction, Heat flux estimation, Gun barrel