使用表面轮廓表征路面纹理

[中文适用范围]： 本文件描述了可用于对路面轮廓进行光谱分析的方法。它规定了一种对二维表面轮廓进行空间频率分析（或纹理波长分析）的方法，该方法将路面纹理振幅描述为路面上直线或曲线轨迹距离的函数。它还详细介绍了一种获取这些光谱的替代（非首选）方法：a) 通过数字滤波获得的恒定百分比带宽（规范方法）；b) 通过离散傅里叶变换 (DFT) 进行恒定窄带宽频率分析，然后将窄带频谱转换为倍频程或三分之一倍频程频谱（参考性）。频率分析的结果将是倍频程或三分之一倍频程带宽的恒定百分比带宽带中的空间频率（或纹理波长）频谱。本文件的目的是标准化路面轮廓的光谱特性。本文件旨在为那些参与路面特性分析但不熟悉随机信号频率分析一般原理的人员提供分析方法和相关要求的详细描述。这些方法和要求通常适用于所有类型的随机信号；但是，本文件对它们在路面轮廓分析中的应用进行了详细说明。注：本文件中规定的频谱分析不能表达所研究的路面轮廓的所有特性。特别是，轮廓不对称的影响，例如“正”和“负”轮廓某些功能质量的差异，不能用功率谱密度来表达，因为它忽略了信号的任何不对称性（见附件 B）。 [外文原描述]： This document describes the methods that are available to perform a spectral analysis of a pavement surface profile. It specifies a method for performing spatial frequency analysis (or texture wavelength analysis) of two-dimensional surface profiles that describe the pavement texture amplitude as a function of the distance along a straight or curved trajectory over the pavement. It also details an alternative (non-preferred) method to obtain these spectra: a) constant-percentage bandwidth obtained by digital filtering (normative method); b) constant narrow bandwidth frequency analysis by means of discrete Fourier transform (DFT), followed by a transformation of the narrow-band spectrum to an octave- or one-third-octave-band spectrum (informative). The result of the frequency analysis will be a spatial frequency (or texture wavelength) spectrum in constant-percentage bandwidth bands of octave or one-third-octave bandwidth. The objective of this document is to standardize the spectral characterization of pavement surface profiles. This objective is pursued by providing a detailed description of the analysis methods and related requirements for those who are involved in pavement characterization but are not familiar with general principles of frequency analysis of random signals. These methods and requirements are generally applicable to all types of random signals; however, they are elaborated in this document for their use in pavement surface profile analysis. NOTE The spectral analysis as specified in this document cannot express all characteristics of the surface profile under study. In particular, the effects of asymmetry of the profile, e.g. the difference of certain functional qualities for “positive” and “negative” profiles cannot be expressed by the power spectral density, as it disregards any asymmetry of the signal (see Annex B).