基于BP神经网络的油松人工林生长模型  

Growth Model of Pinus tabulaeformis Plantation Based on BP Neural Network

作　　者：徐步强[1] 张秋良[1] 弥宏卓[1] 常亮[2] 春兰[1] 吴彤 

机构地区：[1]内蒙古农业大学,呼和浩特010018 [2]内蒙古建筑职业技术学院 [3]乌兰察布市林业局

出　　处：《东北林业大学学报》2011年第12期33-35,共3页Journal of Northeast Forestry University

基　　金：林业公益性行业专项(200804027-03)

摘　　要：以内蒙古大青山油松人工纯林作为研究对象,利用Matlab神经网络工具箱中的logsig、tansig、trainlm等函数,采用二类调查数据中的平均年龄、郁闭度、地位级作为输入层,平均高为输出层,建立油松人工林树高生长BP人工神经网络模型。结果表明,模型的拟合精度为99.98%,检验精度为98.62%,说明该模型对当地的油松人工林生长具有良好的预测能力。A BP artificial neural network model for the growth of tree height of Pinus tabulaeformis artificial forests in Mount Daqing, Inner Mongolia was established, with the help of logsig, tansig, traiulm and other functions in Maflab neural network toolbox, using average forest age, canopy density and site class of the inventory data for management as the input layer, and average tree height as the output layer. Restdt shows that the fitting precision of the model is 99.98%, and the testing accuracy is 98.62%, which means that the model has good forecasting ability for the growth of local artificial forests of P. tabulaeformis.

关 键 词：人工神经网络 树高生长模型 油松 

分 类 号：S757.3[农业科学—森林经理学]