Research On Carbon Nanotubes Database Creation And Performance Prediction

Carbon nanotubes have a unique structure, excellent performance and vast potential application prospects. In this paper, a carbon nanotubes database was developed under Visual C++6.0and SQL Server2008, based on part of the experimental data related with electrical,mechanical,thermal and optical properties of carbon nanotube. An interaction interface is achieved in the system, which includes the following functions:addition, modification and deletion. Combined with sample data in database and the basics of materials science, and according to BP neural network algorithm in the steepest descent method for processing the sample data, the network adjustable parameters about the electrical, mechanical, thermal and optical performance of carbon nanotube was obtained by training network. Finally, the obtained training parameters and structures of the network are used to predict the mechanical properties (elongation and modulus of elasticity) of carbon nanotubes by inputting parameters (volume fraction, length, cross-sectional area and tensile strength).