13 October 2008 An optimization algorithm of neural network for wood physical property modeling
Author Affiliations +
Abstract
As a heterogeneous nature material, there are unknown nonlinear relationships existing in the wood physical property modeling. To solve the complex nonlinear relationship of modeling parameters, an optimization algorithm of neural network based on Gauss-Chebyshev for wood physical property modeling is presented in this paper. The density of wood ring and moisture content are considered as the model inputs, while wood vertical elastic modulus as the output. By comparison the performance between Gauss neural network and Gauss-Chebyshev neural network, the latter is convergence fast with high generalization ability and approximation accuracy of the model.
© (2008) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Mingbao Li, Mingbao Li, Jiawei Zhang, Jiawei Zhang, Runlong Guo, Runlong Guo, Hongyu Su, Hongyu Su, "An optimization algorithm of neural network for wood physical property modeling", Proc. SPIE 7127, Seventh International Symposium on Instrumentation and Control Technology: Sensors and Instruments, Computer Simulation, and Artificial Intelligence, 71272F (13 October 2008); doi: 10.1117/12.806743; https://doi.org/10.1117/12.806743
PROCEEDINGS
6 PAGES


SHARE
Back to Top