Predicting Contact Angle of Electrospun PAN Nanofiber Mat Using Artificial Neural Network and Response Surface Methodology

In this work, the simultaneous effects of four electrospinning parameters, including solution concentration (wt%), applied voltage (kV), tip to collector distance (cm), and volume flow rate (mL/h), on contact angle (CA) of polyacrylonitrile nanofiber mat are studied. To optimize and predict the CA of electrospun fiber mat, response surface methodology (RSM) and artificial neural network (ANN) are employed and a quantitative relationship between processing variables and CA of the electrospun fibers is established. It is found that the solution concentration is the most important factor impacting the CA of electrospun fiber mat. The obtained results demonstrated that both the proposed models are highly effective in estimating the CA of electrospun fiber mat. However, more accurate results are obtained by the ANN model as compared to the RSM model. In the ANN model, the determination coefficient (R-2) and relative error between actual and predicted response are obtained as 0.965 and 5.94%, respectively. (c) 2013 Wiley Periodicals, Inc. Adv Polym Technol 2013, 32, 21365; View this article online at wileyonlinelibrary.com. DOI 10.1002/adv.21365