Title :
Hybrid pso algorithm for estimation modulus of elasticity of wood
Author :
Li, Mingbao ; Zhang, Jiawei
Author_Institution :
Sch. of Civil Eng., Northeast Forestry Univ., Harbin
Abstract :
Particle swarm optimization algorithm based neural network construction has been presented to calibrate the complex nonlinear relationship between modulus of elasticity (MOE) and wood physical property parameters. Consider that the traditional BP algorithm has shortcomings of converging slowly and easily trapping a local minimum value, a hybrid algorithm using particle swarm optimization (PSO) and back propagation (BP) is adopted to train the neural network. Modeling and simulation results show that the optimization technique based on PSO modeling method is feasible and effective, with high generalization ability of the model and forecast accuracy.
Keywords :
backpropagation; elasticity; neural nets; particle swarm optimisation; timber; back propagation; complex nonlinear relationship; estimation modulus; hybrid PSO algorithm; modulus of elasticity; neural network construction; particle swarm optimization algorithm; wood elasticity; Breast; Density measurement; Elasticity; Forestry; Moisture measurement; Neural networks; Particle swarm optimization; Physics; Predictive models; Testing; Modulus of elasticity of wood; neural network; particle swarm optimization;
Conference_Titel :
Computational Intelligence for Measurement Systems and Applications, 2009. CIMSA '09. IEEE International Conference on
Conference_Location :
Hong Kong
Print_ISBN :
978-1-4244-3819-8
Electronic_ISBN :
978-1-4244-3820-4
DOI :
10.1109/CIMSA.2009.5069959