An improved particle swarm optimization algorithm for site index curve model

The precision of ideal site index curve model mainly depends on the solved parameters of the fitting equation. To archive high precision of model, a particle swarm optimization algorithm with iterative improvement strategy was proposed to solve parameters of the model. The improved algorithm makes each particle update it´s current velocity and position dimension by dimension. The result shows the parameters solved using particle swarm optimization with or without iterative improvement strategy make the model to be small overall error, high precision, ideal of fitting effect, scientific, and reasonable. It also indicates that particle swarm optimization with iterative improvement strategy is better than particle swarm optimization without iterative improvement strategy on the performance with the same conditions. The study provides a new way for solving parameter of growth model in forest management, and for the related research. It also enriches not only optimization technology about stand management, but also the application domain of particle swarm optimization algorithm. It can be predicted that particle swarm optimization algorithm will be the broad application prospect in the forestry production and scientific research.