Erscheint 6 Ausgaben pro Jahr
ISSN Druckformat: 2150-766X
ISSN Online: 2150-7678
Indexed in
MULTILINEAR REGRESSION ANALYSES AND ARTIFICIAL NEURAL NETWORK IN PREDICTION OF HEAT OF DETONATION FOR HIGH-ENERGETIC MATERIALS
ABSTRAKT
In this work, two simple approaches have been introduced to predict heat of detonation of highenergetic materials. Experimental heat of detonation of 74 energetic compounds were collected from articles and this data set was separated randomly into two groups, i.e., training and prediction sets, which were used for generation and evaluation of suitable models. Multiple linear regression (MLR) analysis was employed to build a linear model, while nonlinear models were developed by means of an artificial neural network (ANN). The obtained models with four descriptors involved show good predictive power for the test set: a squared correlation coefficient (R2) of 0.798 and a standard error (SE) of estimation of 606.48 (J/g) were achieved by the MLR model; whereas by the ANN model, R2 and SE were 0.98 and 189.4 (J/g), respectively. On the basis of the large R2 value and small SE values, one can deduce that the predicted results are in good agreement with the measured values.