基于二维多规则云模型定性推理的场地分类方法
Site Classification Method Based on Qualitative Reasoning of 2-D & Multi-rules Cloud Model
-
摘要: 针对场地土及其分类的模糊性和随机性特征,建立了基于二维多规则云模型定性推理的场地分类方法,实现了场地分类边界不确定性划分.设计了场地分类的二维多规则定性推理生成器,将其应用于场地类别和特征周期的识别过程中,实现了场地分类和特征周期的连续化,克服了在分类边界处易出现类别跳跃等不合理现象,基本实现了现行建(构)筑物抗震规范在场地分类方面的统一.最后利用该方法对74个实例进行分析,场地分类的正确率达到97.3%,表明了该方法是可行的、有效的.Abstract: According to the site classifications and their uncertain and stochastic characteristics,the site classification method based on qualitative reasoning of 2-D & multi-rules cloud model was constructed and the uncertain division of the site classification was realized.Then,the generator on 2-D & multi-rules for uncertainty reasoning was designed to identify site classifications and site characteristic period,so the continuity of the site classification and site characteristic period were actualized.This method got rid of the illogical discontinuity and appreciable distinction displayed on the code chart between the site edges,basically realized the unification of the site classification between code of seismic design of buildings and special structures.Finally,74 examples were analyzed by the method.It was shown that the sites were classified with the correct rate of 97.3%,and the method proposed in the paper proved to be effective and feasible.