The fatigue life evaluation of metallic materials plays an important role in ensuring the safety and long service life of metal structures. To further improve the accuracy and efficiency of the ultra-high-cycle fatigue life prediction of metallic materials, a new prediction method using machine learning was proposed. The training database contained the ultra-high-cycle fatigue life of different metallic materials obtained from fatigue tests, and two fatigue life prediction models were constructed based on the gradient boosting (GB) and random forest (RF) algorithms. The mean square error and the coefficient of determination were applied to evaluate the performance of the two models, and their advantages and application scenarios were also discussed. The ultra-high-cycle fatigue life of GCr15 bearing steel was predicted by the constructed models. It was found that only one datapoint of the GB model exceeded the triple error band, and the RF model had higher stability. The network model coefficient of determination and mean square error for the GB and RF models were 0.78, 0.79 and 0.69, 3.79, respectively. Both models could predict the ultra-high-cycle fatigue life of metallic materials quickly and effectively.

Purpose: The purpose of this paper is to study the drop reliability of ball-grid array (BGA) solder joints affected by thermal cycling. Design/methodology/approach: The drop test was made with the two kinds of chip samples with the thermal cycling or not. Then, the dyeing process was taken by these samples. Finally, through observing the metallographic analysis results, the conclusions could be found. Findings: It is observed that the solder joint cracks which were only subjected to drop loads without thermal cycling appeared near the BGA package pads. The solder joint cracks which were subjected to drop loads with thermal cycling appear near the printed circuit board pads. Originality/value: This paper obtains the solder joint cracks picture with drop test under the thermal cycling.

In order to enhance the surface properties of GCr15 bearing steel, a TiAlCN coating with a low friction coefficient, high hardness, and excellent adhesion was fabricated. The TiAlCN multilayer coating was deposited onto the GCr15 bearing steel surface using magnetron sputtering technology, and optimal coating parameters were achieved by adjusting the number of layers, sputtering power of the graphite target, and coating duration. The experimental results showed that adding Cr/CrN as a transition layer between GCr15 bearing steel and TiAlCN significantly improved multiple properties of the coating. Adding carbon atoms caused TiAlN to dissolve into a TiAlCN structure, enhancing multiple properties of the coating. With the increase in the sputtering power of the graphite target material, the hardness, friction, and wear performance of the coating showed a trend of first increasing and then decreasing. The hardness of the coating gradually increased with time, and the friction coefficient and wear amount first decreased and then increased. When the sputtering power of the graphite target material was 100 W and the coating time was 4800 s, the coating performance was optimal. The hardness was 876 HV, the friction coefficient was 0.42, the wear amount was 1 × 10⁻⁴ g, and the wear rate was 2.8 × 10⁻⁶ g/m·N under optimal process parameter conditions.

The main focus of this work is the successful deposition of hard and wear-resistant TiAlCN coating on the surface of GCr15 bearing steel by means of magnetron sputtering technology. The phase composition of the chromium nitride transition layer was monitored by precisely controlling the nitrogen (N₂) flow rate to strengthen the bonding between the TiAlCN coating and the GCr15 bearing steel surface. It was found that coating performance reached the optimal state at a N₂ flow rate of 40 sccm, yielding a hardness of 23.3 GPa, a friction coefficient of only 0.27, and a wear rate of 0.19 × 10⁻⁸ mm³/N·m.