Search

Ultrasonic rolling is an effective technique for enhancing surface integrity, and surface integrity is closely related to fatigue performance. The process parameters of ultrasonic rolling critically affect the improvement of surface integrity. This study proposes an optimization method for process parameters by combining machine learning (ML) with the NSGA-II. Five ML models were trained to establish relationships between process parameters and surface residual stress, hardness, and surface roughness by incorporating feature augmentation and physical information. The best-performing model was selected and integrated with NSGA-II for multi-objective optimization. Ultrasonic rolling tests based on a uniform design were performed, and a dataset was established. The objective was to maximize surface residual stress and hardness while minimizing surface roughness. For test specimens with an initial surface roughness of 0.54 µm, the optimized process parameters were a static pressure of 900 N, a spindle speed of 75 rpm, a feed rate of 0.19 mm/r, and rolling once. Using optimized parameters, the surface residual stress reached −920.60 MPa, surface hardness achieved 958.23 HV, surface roughness reduced to 0.32 µm, and contact fatigue life extended to 3.02 × 107 cycles, representing a 52.5% improvement compared to untreated specimens and an even more significant improvement over without parameter optimization.

The demand for titanium alloy has been increasing in various industries, including aerospace, marine, and biomedical fields, as they fulfilled the need for lightweight, high-strength, and corrosion-resistant material for modern manufacturing. However, titanium alloy has relatively low hardness, poor wear performance, and fatigue properties, which limits its popularization and application. These disadvantages could be efficiently overcome by surface strengthening technology, such as the ultrasonic surface rolling process (USRP). In this study, the true thermo-mechanical deformation behavior of Ti-6Al-4V was obtained by dynamic mechanical experiment using a Hopkinson pressure bar. Moreover, USRP was applied on the Ti-6Al-4V workpiece with different parameters of static forces to investigate the evolution in surface morphology, surface roughness, microstructure, hardness, residual stress, and fatigue performance. The strain rate and temperature during the USRP of Ti-6Al-4V under the corresponding conditions were about 3000 s−1 and 200 °C, respectively, which were derived from the numerical simulation. The correlation between the true thermo-mechanical behavior of Ti-6Al-4V alloy and the USRP parameters of the Ti-6Al-4V workpiece was established, which could provide a theoretical contribution to the optimization of the USRP parameters. After USRP, the cross-sectional hardness distribution of the workpiece was shown to initially rise, followed by a subsequent decrease, ultimately to matrix hardness. The cross-sectional residual compressive stress distribution of the workpiece showed a tendency to initially reduce, then increase, and finally decrease to zero. The fatigue performance of the workpiece was greatly enhanced after USRP due to the effect of grain refinement, work hardening, and beneficial residual compressive stress, thereby inhibiting the propagation of the fatigue crack. However, it could be noted that the excessive static force parameter of USRP could induce the decline in surface finish and compressive residual stress of the workpiece, which eliminated the beneficial effect of the USRP treatment. This indicated that the choice of the optimal USRP parameters was highly crucial. This work would be conducive to achieving high-efficiency and low-damage USRP machining, which could be used to effectively guide the development of high-end equipment manufacturing.

Oxide-dispersion- and hard-particle-strengthened (ODS) laser-cladded single-layer multi-tracks with a Ni-based alloy composition with 20 wt.% μm-WC particles and 1.2 wt.% nano-Y2O3 addition were produced on ultra-high-strength steel in this study. The investigation of the composite coating designed in this study focused on the reciprocating friction and wear workpiece surface under heavy load conditions. The coating specimens were divided into four groups: (i) Ni-based alloy, nano-Y2O3, and 2 μm-WC (2 μm WC-Y/Ni); (ii) Ni-based alloy with added 2 μm-WC (2 μmWC/Ni); (iii) Ni-based alloy with added 80 μm-WC (80 μmWC/Ni); and (iv) base metal ultra-high-strength alloy steel 30CrMnSiNi2A. Four conclusions were reached: (1) Nano-Y2O3 could effectively inhibit the dissolution of 2 μm-WC. (2) It can be seen from the semi-space dimensionless simulation results that the von Mises stress distribution of the metal laser composite coating prepared with a 2 μm-WC particle additive was very uniform and it had better resistance to normal impact and tangential loads than the laser coating prepared with the 80 μm-WC particle additive. (3) The inherent WC initial crack and dense stress concentration in the 80 μm-WC laser coating could easily cause dislocations to accumulate, as shown both quantitatively and qualitatively, resulting in the formation of micro-crack nucleation. After the end of the running-in phase, the COF of the 2 μm-WC-Y2O3/Ni component samples stabilized at the minimum of the COF of the four samples. The numerical order of the four COF curves was stable from small to large as follows: 2 μm-WC-Y2O3/Ni, 2 μm-WC/Ni, 80 μm-WC/Ni, and 30CrMnSiNi2A. (4) The frictional volume loss rate of 2 μm-WC-Y2O3/Ni was 1.3, which was significantly lower than the corresponding values of the other three components: 2.4, 3.5, and 13.

As a new surface treatment technology, laser shock peening (LSP) is a multi-point overlay process of single-point laser shock. In this study, the finite element method (FEM) was used to build a model of single-point laser shock, and the model was verified by experiments. The difference in residual stresses between the experimental and simulated results was less than 20%. Then, the effects of the stress field and deformation of 20CrMnTi with different laser shock parameters were simulated and analyzed. According to the mechanical response of 20CrMnTi to different laser shock parameters, the optimal shocking process parameters for single-point shocking via LSP were determined to be a shock energy of 5 J, a laser pulse width of 20 ns, and an impact number of 5. Lastly, a simulation of multi-point laser shock was performed with the optimal parameters, and the residual compressive stress values of multi-point impacts are close to those of single-point impacts under the same process conditions.

To quantitatively characterize the pyramidal texture of monocrystalline silicon cells and to optimize the parameters of the texturing process, the relative standard deviation Sh was proposed to quantitatively characterize the uniformity of the pyramidal texture. Referring to the definition and calculation of the standard deviation in mathematical statistics, Sh was defined as the standard deviation of the pyramid relative height hi after normalization of the pyramid height Hi of monocrystalline silicon wafer surfaces. Six different silicon cells, with different pyramidal textures, were obtained by applying different texturing times. The relationships between Sh and the photoelectric characteristics were analyzed. The feasibility of quantitatively characterizing the uniformity of the pyramidal texture using Sh was verified. By fitting the Sh curve, the feasibility of optimizing the texturing process parameters and predicting the photoelectric characteristics using Sh was verified. The experimental and analytical results indicate that, when the relative standard deviation Sh was smaller, the uniformity of the pyramidal texture obtained by texturing was better. The photoelectric conversion efficiency (PCE) of the silicon cells monotonically increased with decreasing Sh. The silicon cell obtained by texturing with 2% tetramethylammonium hydroxide (TMAH) solution for 18.1 min had a textured surface with a minimum of Sh, the reflectivity of the silicon cell reached its minimum value of 2.28%, and the PCE reached its maximum value of 19.76%.