Search

The application of machine learning (ML) to corrosion research has become an important trend in corrosion science in recent years. In this paper, the feature extraction method for corrosion data and the ML algorithms commonly used (including artificial neural networks, support vector machines, ensemble learning and other widely used algorithms) in corrosion field is introduced. Then, the characteristics of different algorithms and their application scenarios in the corrosion prediction are summarized. Finally, the development trend of ML in material corrosion field is prospected.

For small aspherical molds, it is difficult for the existing polishing method to take into account the correction of the surface error and the control of the uniformity of the surface roughness (SR) distribution, because the polishing tool is always larger than the small mold. Therefore, we used viscoelastic polyester fiber cloth to wrap the small steel ball as a polishing tool to adapt to the surface shape change of the aspherical mold, and designed a semi-flexible small polishing disc tool with microstructure, which can better adapt to the curvature change of aspherical surface and obtain better SR Ra. At the same time, a combined polishing method of constant speed and variable speed for screw feed was proposed to improve the uniformity of SR distribution in the paper. Then, a series of theoretical analysis and experimental verification were carried out in this paper to predict the tool influence function (TIF) of the two polishing tools and the effectiveness of the combined polishing method. In the experiment, a TIF bandwidth of about 0.46 mm was obtained with a small spherical polishing tool, which favors the surface shape correction of the small aspherical mold. The experiment of uniform removal with a small polishing disc tool was carried out to quickly reduce the Ra. Finally, the surface quality of the aspherical mold was effectively improved, combined with the constant speed and variable speed polishing modes of screw feed of the small spherical polishing tool and the smoothing effect of the small polishing disc tool. The peak valley (PV) of two small aspherical molds with an optical effective diameter less than 13 mm converged from 0.3572 μm and 0.2075 μm to 0.1282 μm and 0.071 μm, respectively. At the same time, the SR dispersion coefficient was reduced from 27.9% and 41.6% to 14.2% and 12.7%, respectively. The study provides a good solution for the surface quality control of small aspherical molds.

In terms of longer hardening time of emulsified asphalt seal mixture, a rapid hardening seal mixture was prepared by adding cement and carbon fibre to the Styrene–Butadiene-Rubber (SBR)-modified em...

To suppress the mid-high-frequency error of small optical tungsten carbide aspheric molds, it is proposed to quickly select the critical process parameters by simulating the residual error after convolution of the tool influence function (TIF). After polishing for 10.47 min by the TIF, two simulation optimizations, RMS and Ra, converge to 9.3 and 5.347 nm, respectively. Their convergence rates are improved by 40% and 7.9%, respectively, compared to ordinary TIF. Then, a faster and more high-quality multi-tool combination smoothing suppression method is proposed, and the corresponding polishing tools are designed. Finally, the global Ra of the aspheric surface converges from 5.9 to 4.5 nm after smoothing for 5.5 min with a disc-shaped polishing tool with a fine microstructure and maintains an excellent low-frequency error (PV 0.0781 µm).