基于神经网络的森林蓄积量估测  

Prediction of forest stock volume based on neural network model

作　　者：涂云燕[1] 彭道黎[1] 

机构地区：[1]北京林业大学,北京100083

出　　处：《中南林业科技大学学报》2012年第3期49-52,共4页Journal of Central South University of Forestry & Technology

基　　金：国家"十一五"科技支撑计划课题(2006BAD23B05);国家级推广项目(201145)

摘　　要：在前人研究中还没有把基于BP与RBF神经网络的森林蓄积量预测模型的应用效果进行评价。拟在实际应用中对两种方法进行综合分析与评价,找到一种预测精度更高、适用性更强的方法。采用相关分析法选定郁闭度、阴坡、阳坡、TM1、TM2、TM3、TM5、TM7、NDVI、TM(4-3)、TM4/3为输入变量,以密云县森林蓄积量为输出变量,建立蓄积量估测的RBF与BP神经网络模型。并从神经网络的训练步长、训练时间、预测精度、模型适用性对二者进行了综合分析,RBF神经网络无论是在训练步长、训练时间、预测精度、模型适用性上都优于BP神经网络模型。BP and RBF neural network to predict forest stock volume of have been studied,but the study in evaluating the both networks＇ application effects didn＇t conduct.In order to find a higher forecast precision,more strong applicative method,in the practical application,the comprehensive analysis and evaluation on the two methods were carried out.By the correlation analysis,and selecting crown density,shady-slope and sunny-slope.TM1 TM2,TM3,TM5,TM7,NDVI,TM,（4-3）,TM4/3 as input variables,the volume of forest of Miyun county as output variables,RBF and BP neural network model for the volume of forest were established.And the neural network training step length,training time,prediction accuracy and the applicability model of the two methods were comprehensively analyzed.The results show that in the aspacts of the training step length,training time,prediction accuracy and the applicability,the model of RBF neural network model is superior to BP neural network model.

关 键 词：BP神经网络 RBF神经网络 综合分析与评价 北京密云县 森林蓄积量预测 

分 类 号：S757[农业科学—森林经理学]