Modeling of burning rate equation of ammonium perchlorate particles over Cu–Cr–O nanocomposites (original) (raw)

Journal of Thermal Analysis and Calorimetry, 2014

Abstract

ABSTRACT The present study is aimed at introducing the application of design of experiment (DOE) with artificial neural networks (ANN) in couple over Cu–Cr–O nanocomposites, which can be used as catalysts in the combustion of ammonium perchlorate (AP), as well as a method of data collection/fitting for the experiments. To this end, a practical scheme has been proposed to select the characterization parameters of Cu–Cr–O nanocomposites as the catalytic combustion of AP propellant. Moreover, a calculation model has been established to identify the primary combustion characteristics based on backpropagation neural networks. The model was subsequently validated and then used to predict the primary combustion characteristics of the aforementioned propellant. In addition, due to the complex nature of the system, neural networks were employed as an efficient and accurate tool to model the behavior of the system. Response surface methodology (RSM) and ANN methods were also constructed based upon the DOE’s points and were then utilized to generate extra simulated data. The data sets, including the original experimental data and the simulation results yielded by the ANN and RSM methods were subsequently used to fit the combustion rate expression for AP. A comparison of the results of kinetic modeling with the simulated data sets from ANN and RSM models was then made, which indicated that both methods could satisfactorily fit the experimental data presented in the literature. The results also revealed that the error of burning rate calculation is less than ±5 % and the variations of the calculation results were consistent with those of the experimental results.

Azam Ghavi hasn't uploaded this paper.

Let Azam know you want this paper to be uploaded.

Ask for this paper to be uploaded.