Journal of Mathematical Sciences: Advances and ApplicationsAuthor's: Wenxia Dong, Ting Wang, Ying Chen Tiandi Pan and Rudong Chen
Pages: [51] - [68]
Received Date: May 6, 2018; Revised July 16, 2018
Submitted by:
DOI: http://dx.doi.org/10.18642/jmsaa_7100121964
Since the diameter of the nanofiber affects the filtration efficiency and resistance of the nonwoven fabric, we use the BP neural network to study the Mathematical Model of the electrospinning nanofiber diameter about the distance (cm), the injection rate (mL/h), and the voltage (kV). First, based on the theory of statistical orthogonality, we make 30 sets of experiments and obtain 30 sets of nanofiber membranes. After, electron microscope images of the nanofiber membrane are got by the TM-1000 electron microscopy. Then, we use C++ to process the fiber images to get the diameter distribution of the nanofibers. The function relationship of the nanofiber diameter under three variables is obtained by MATLAB software to apply BP neural network knowledge. After obtaining the functional relationship, we random make three groups of experiments and compare the results with the experimental results and calculation results finding that the relative error is small, verify the function correct. The BP model predicted the minimum diameter (1,001nm) of electrospinning nanofiber membrane at the conditions of 19cm of spinning distance, 26kV of the applied voltage, and 0.8mL/h of volume flow rate.
BP neural network, electrospinning nanofiber, MATLAB, C++.
