摘要
为了以尽量少的胸径和树高数据来预测材积,利用灰色理论建立以日本落叶松平均材积和D^2H这2个因素的分数阶GM(1,2)模型,该模型通过寻找在0阶与1阶之间的最优阶数r=0.26,从而使得模型的精度得到大幅度提高,模型的平均相对误差为1.905%;模型精度为98.005%;与传统的GM(1,2)模型相比,分数阶GM(1,2)模型更为优秀。分数阶GM(1,2)模型为树木材积的预测提供了一种新方法。
In order to predict the product with minimal chest diameter and tree height data, a fractional order GM(1, 2)model based on the two factors of Japanese larch average product and D^2 H is established by using grey theory. The model of this paper greatly improves the accuracy by finding the optimal order r=0.26 between 0 and 1 orders, the average relative error of the model is 1.905%, and the model accuracy is 98.005%. Compared with the traditional GM(1, 2) model, the fractional GM(1, 2) order model is superior. The fractional order GM(1, 2) model provides a new method for the prediction of tree product.
作者
阙素琴
林艳芳
滕忠铭
QUE Su-qin;LIN Yan-fang;TENG Zhong-ming(College of computer and information science, Fujian Agricultural and Forestry University, Fuzhou 350002, China)
出处
《三明学院学报》
2019年第2期20-25,共6页
Journal of Sanming University
基金
福建农林大学杰出青年基金项目(xjq201727)
