水环境模拟中多孔介质场空间变异性研究方法述评

阐述了研究多孔介质空间变异性的经典统计法、地质统计学方法以及新兴的分形方法.经典统计方法由于不考虑研究对象的空间相关性,已经不能对这种变异性做出客观的分析;地质统计学方法很好地克服了经典统计学的缺点,是目前应用最多的一种方法;分形的研究对象为无规则和具有自相似性的体系,这一特性决定了它研究这种变异性有着独特的优点.最后对多孔介质空间变异性研究的趋势进行了展望.

Several methods for research of spatial variability of porous media field in water environmental simulation

This paper has been devoted to an analysis of the theories of classical statistics and geostatistics to find, the spatial variability of the characteristics of porous media and contaminants transfer in them, which proves highly useful in many areas, such as in groundwater contaminants control, exploitation of soil resources, and the improvement of wastewater irrigation technology. However, traditional statistical methods ignore the spatial correlation of parameters and take it as a pure random variable. They even could not give a proper analysis of the spatial variability. Of course, it is undeniable that a lot of classical statistical methods, for instance, the method of multivariate statistical analysis, can still be successfully applied in this field. In the research practice, when geostatistics takes the randomicity and spatial correlation of the objects into account, it is possible to overcome the main limitations of classical statistics, which makes it possible to be commonly used till today. Since the fractal statistics takes as its research objects the irregularity and self-similarity, though brought about later than the other two, it has quickly become fashionable for its conformity with the complexity and chaos of nature, and has even turned to be the unique method for the system research. Therefore, the present paper has introduced the ways on how to use the geostatistics and fractal statistics in spatial variability. Lastly of all,based on the knowledge mentioned above, the perspective of the research of spatial variability of porous media has also been discussed.