Characterization of Carbon nanotube reinforced Silica refractory nanocomposite using Artificial Intelligence Modelling: PART B

摘要

The present study has dwelled on the implementation and evaluation of an artificial intelligence model for the determination of predicted foundry physical properties;linear expansion,bulk density,apparent porosity,thermal shock resistance cycles and cold crushing strength of carbon nanotube(CNT)reinforced silica refractory nanocomposite.A multi input and multi output Artificial Neural Network(ANN)models were developed using the Levenberg Marquardt Back Propagation algorithm(LMBPA)in the neural network toolbox of MATLAB R2015 a to train/predict the foundry physical properties of the CNT-silica refractory nanocomposite bricks obtained experimentally from the previous study.The predicted models were compared with the experimental test results in order to evaluate the power and the accuracy of the artificial intelligence model for the characterization of the entire series of CNT-silica refractory nanocomposite bricks.The developed(LMBPA ANN)model satisfactorily predicts the foundry physical properties of CNT reinforced silica nanocomposite with a coefficient of determination(R~2)in the range 0.75 ≥R~2 ≤ 1.