基于BP神经网络的森林碳汇量估测方法研究

Researches on the Estimation Method of Forest Carbon Sequestration Based on BP Neural Networks

文介绍了BP神经网络理论,以森林面积、森林覆盖率、森林蓄积量、受害森林面积、基本完成投资额、病虫害面积等6个影响森林碳汇量为主要因素,构建了基于BP神经网络的森林碳汇量估测模型;并应用我国第六次森林资源清查数据对BP神经网络模型进行训练和仿真。其分析和仿真结果显示:利用BP神经网络模型来模拟和仿真森林碳汇量精度较高,误差较小,具有有较高可靠性,从而为森林资源管理模拟仿真提供一种新方法。

In this paper,a description is given of the theory of BP neural networks and the forest carbon sequestration model is established on the basis of BP neural networks by use of six main factors such as the forest area, forest coverage rate, stock volume of forest, damaged forest area, completed investment, plant diseases and insect pest area. The BP neural networks were trained and simulated on the sixth forest resource inventory data. The results showed that this model had less error and higher accuracy and relia- bility,thus providing a new method of simulation for forest resource management.