Abstract: (19280 Views)
Coating of a surface by droplet spreading plays an important role in many novas industrial processes, such as plasma spray coating, ink jet printing, nano safeguard coatings and nano self-assembling. Data analysis of nano and micro droplet spreading can be widely used to predict and optimize coating processes. In this article, we want to select the most appropriate statistical distribution for spread data of aluminum oxide splats reinforced with carbon nanotubes. For this purpose a large class of probability models including generalized exponential (GE), Burr X (BX), Weibull (W), Burr III (BIII) distributions are fitted to data. The performance of the distributions are estimated using several statistical criteria, namely , Akaike Information Criterion (AIC), Baysian Information Criterion (BIC), LogLikelihood (LL) and Kolmogorove-Smirnove distance. Also, the fitted plots of probability distribution function and quantile-quantile (q-q) plots are used to verify the results of different criteria. An important implication of the present study is that the GE distribution function, in contrast to other distributions, may describe more appropriately in these datasets.