®®®® SIIA Público

Título del libro: Eccomas Congress 2016 - Proceedings Of The 7th European Congress On Computational Methods In Applied Sciences And Engineering
Título del capítulo: Statistical and geometrical characteristics of randomly rough surfaces used for contact simulations

Autores UNAM:
RAFAEL SCHOUWENAARS; VICTOR HUGO JACOBO ARMENDARIZ;
Autores externos:

Idioma:
Inglés
Año de publicación:
2016
Palabras clave:

Computational methods; Deformation; Fractal dimension; Fractals; Mechanics; Random processes; Spectroscopy; Surface measurement; Contact Mechanics; Fractal surfaces; Geometrical characteristics; Mid-point algorithms; Spectral methods; Statistical convergence; Surface characteristics; Weierstrass-mandelbrot functions; Finite element method


Resumen:

Over the last decades, it has been established that rough surfaces of technological materials can be described by means of random processes with fractal character. This means that the numerical simulation of contact mechanics will depend on the size of the simulated surface or the lower cut-of distance (i.e. mesh size) of the simulation; moreover, each simulation only presents a single case out of an infinite spectrum of possible results. This work explores the problem of statistical convergence by focusing on the surface characteristics which are relevant for contact mechanics applications, i.e. the amount of simulations required to obtain a reasonable estimate of the surface characteristics as a function of the size of the simulated surface. This analysis is applied to three classical methods for the generation of fractal surfaces, i.e. the Midpoint Displacement algorithm, the generalised Weierstrass-Mandelbrot function and a Random Spectral method. Apart from geometrical characteristics, the simulated surfaces will be evaluated by means of a numerical modification of the Greenwood-Williamson model. It will be seen that the Weierstrass-Mandelbrot function has several shortcomings which may affect advanced contact simulations. More generally, the statistical variation between individual surface simulations introduces a considerable spread on the results, as seen in the distribution of calculated fractal dimensions or load-contact area curves.


Entidades citadas de la UNAM: