抗反射和疏水性粗糙表面的拓扑表征:随机过程理论和分形模型适用吗?
Topological characterization of antireflective and hydrophobic rough surfaces: are random process theory and fractal modeling applicable?
摘要:
随机过程理论(RPT)已被广泛应用于预测粗糙表面的粗糙度和曲率的联合概率分布函数(PDF)。RPT的预测对数值生成的随机分形表面和真正的粗糙表面的实际统计数据的检查已经只进行了部分。目前的实验和数值研究提供了一个深刻的关键比较在这个问题上,提供了一些洞察力的能力和局限性,在应用RPT和分形建模的抗反射和疏水粗糙表面,两种重要类型的纹理表面。利用共焦轮廓仪进行了不同透镜的多分辨率实验运动,并开发了一个用于粗糙表面统计描述的综合软件。研究发现,RPT和分形模型不能完全描述所分析的纹理表面的拓扑结构。出现了以下复杂情况:(i)幂律功率谱密度(PSD)函数中存在截断或双分形;(ii)与分形建模相比,通过改变分辨率,PSD的偏移更明显;(iii)RPT在描述粗糙度高度和纹理表面曲率的联合PDF时不准确;(iv)在特殊表面处理的情况下,纹理化表面的联合PDF的分辨率不变性的缺乏,不被分形建模考虑。
英文摘要:
The random process theory (RPT) has been widely applied to predict the joint probability distribution functions (PDFs) of asperity heights and curvatures of rough surfaces. A check of the predictions of RPT against the actual statistics of numerically generated random fractal surfaces and of real rough surfaces has been only partially undertaken. The present experimental and numerical study provides a deep critical comparison on this matter, providing some insight into the capabilities and limitations in applying RPT and fractal modeling to antireflective and hydrophobic rough surfaces, two important types of textured surfaces. A multi-resolution experimental campaign by using a confocal profilometer with different lenses is carried out and a comprehensive software for the statistical description of rough surfaces is developed. It is found that the topology of the analyzed textured surfaces cannot be fully described according to RPT and fractal modeling. The following complexities emerge: (i) the presence of cut-offs or bi-fractality in the power-law power-spectral density (PSD) functions; (ii) a more pronounced shift of the PSD by changing resolution as compared to what expected from fractal modeling; (iii) inaccuracy of the RPT in describing the joint PDFs of asperity heights and curvatures of textured surfaces; (iv) lack of resolution-invariance of joint PDFs of textured surfaces in case of special surface treatments, not accounted by fractal modeling.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。