基于知识粒度的林分择伐空间结构优化研究

Research on spatial structure optimization of stand selection cutting based on knowledge granularity

以湖南省大围山自然保护区2个天然阔叶混交林样地为应用实例,选取自由度、混交度、大小比数、健康指数、空间密度指数和目的树种特性指数等6个空间结构影响因子来分析林分采伐木的确定,建立林分择伐空间结构优化模型。应用粗糙集理论中的知识粒度赋权法挖掘林分空间结构影响因子之间的相关性及其重要度,得到各影响因子的权重,从而确定间伐指数。应用知识粒度的赋权法对其中一块样地各影响因子的原始数据进行分析,得到各影响因子的权重。在确定间伐强度的前提下,对另一样地进行林分择伐空间结构优化,确定采伐木。结果表明:知识粒度赋权法不依赖专家经验,直接从原始数据中挖掘信息来确定权重,可使间伐指数的计算更客观实际。

Taken two sample plots of natural broad - leaf mixed forest in Hunan Dawei Mountain as example, structure decision function according to system theory, from the aspects of freedom, mixed degree, the size of the score, the healthy index, spatial density index and objective tree spatial structure characteristic index of six factors to analyze the de- termination of forest cutting wood, the stand selective spatial structure optimization model was established. Application of knowledge granularity in rough set theory method to dig the relationship between the impact factors of the forest stand spa- tial structure and its importance, the weight of each influence factor was obtained to determine the thinning index. Using the method of knowledge granularity of empowerment in one piece sample, the raw data of each influence factor was ana- lyzed to get the weight of each influence factor. And then determined the thinning intensity conditions, the spatial struc- ture optimization of forest stand selective in the other sample was conducted to determine the cutting wood. Results showed that the method of knowledge granularity of empowerment is not depend on expert experience, directly from the original data to determine the weight, so that the thinning index calculation is more objective reality.