Title :
A Fusion Algorithm Based Improved Function Link Artificial Neural Network for Lumber Moisture Content (LMC) Measuring
Author :
Li, Ming-bao ; Zhang, Jia-wei ; Zheng, Shi-Qiang
Author_Institution :
Northeast Forestry Univ., Harbin
Abstract :
Aimed to improve the measurement precision of the traditional microwave transmission method for lumber moisture content (LMC), this paper presents a dynamic compensation technique based on function link neural networks (FLNN). The microwave attenuation and phase shift are taken as the inputs of the dynamic compensation model. Consider that the traditional BP algorithm has shortcomings of converging slowly and easily trapping a local minimum value, a combination learning algorithm using particle swarm optimization (PSO) and BP is adopted to train the neural network dynamic compensation model. It will enable the compensation process with an overall accuracy. Experimental results show that the use of the technology on lumber moisture content measurements for calibration is an effective method and has certain project value.
Keywords :
learning (artificial intelligence); neural nets; particle swarm optimisation; production engineering computing; wood processing; combination learning algorithm; dynamic compensation technique; function link neural networks; fusion algorithm; improved function link artificial neural network; local minimum value; lumber moisture content; microwave attenuation; particle swarm optimization; phase shift; Artificial neural networks; Attenuation; Cement industry; Dielectric constant; Microwave measurements; Microwave technology; Microwave theory and techniques; Moisture measurement; Neural networks; Particle swarm optimization; Function link neural network; Lumber moisture content; Particle swarm optimization;
Conference_Titel :
Machine Learning and Cybernetics, 2007 International Conference on
Conference_Location :
Hong Kong
Print_ISBN :
978-1-4244-0973-0
Electronic_ISBN :
978-1-4244-0973-0
DOI :
10.1109/ICMLC.2007.4370628
