Measurement pitting morphology characteristic of corroded steel surface and fractal reconstruction model

Abstract

Corrosion is one of inevitable problems of constructional steel exposed to atmospheric environment in the long time, of which the quantification usually plays a key role in the accurate safety evaluation of corroded steel structure. Aiming to determine the reasonable quantitative method of corrosion, the morphological parameters were thoroughly investigated based on the surface measurements of 6 groups of corroded steel specimens obtained by accelerated corrosion test. By using the statistical analysis, the corrosion pitting parameters and their evolution rules were systematically analyzed and discussed. The distribution laws of pitting depth, width and spacing distance, and their statistical characteristic parameters (i. e., the expectation and the variance) were revealed. And then, their probability density functions were given at different pitting shape parameter. At last, the fractal character of corroded specimen surface was also researched. The results showed that all the pitting parameters obeyed the chi-square distribution (χ2). The corrosion pit evolved from a single corrosion pit to a double-stepped corrosion pit, and finally to be a single-stepped style. With increase of corrosion degree, the multi-fractal dimension (D) of corroded surface generally increased. Besides, the value of D can be well predicted by the combined effect of all characteristic parameters of corrosion pit. Finally, based on the W-M function and considering the relationship between the characteristic parameters and fractal dimension, and noise reduction and continuous treatment in consideration, the fractal reconstruction model of corroded steel surface was proposed, which can be utilized to accurately simulate the corroded surface of constructional steel.