In this study, the improved forward rod-backward can combined extrusion test was performed for investigation of the tribological properties of Ti-6Al-4V titanium alloy under cold and warm forging. 1.6 and 2.1 µm-thick oxide films were formed on the test piece by atmospheric oxidation treatment. A friction test was performed by changing the temperature of the test piece from 20 to 600 °C without lubricant. As a result, the oxide film formed by atmospheric oxidation treatment was found to prevent seizure within a certain length of punch strokes at 20 °C, and the friction coefficient became approximately 0.15. However, at high temperatures, the decohesion of the oxide film increases with temperature and seizure finally occurred. Therefore, the effect of inhibiting the decohesion of oxide films is limited. We concluded that thickening the oxide film can be effective in inhibiting decohesion.

In the design of the closed die forging process, a process designer must decide the process conditions such as the number of stages and the die surface shape in each stage. When the target product cannot be forged by a one-stage process, a multi-stage process is necessary. There are numerous combinations of the number of stages and the die surface shape in each stage, which is called the “process layout” in this paper. Therefore, it takes a long time to determine the process layout by trial and error. In this study, to design a draft of the process layout automatically, we investigated the technology for process design automation, focusing on axisymmetric disk-shaped products. The design automation technology was developed by combining the proposed method for die surface generation, finite element method, and an optimization algorithm. In addition, to resolve the issue about the increase in the number of process layout candidates when the target product is vertical asymmetric, we investigated a method of reducing their number. As a result of verification for severally vertically symmetric and vertically asymmetric products, we found that process layouts that satisfy the forging load restriction and forged shape accuracy could be automatically designed.

Peen forming has been widely applied to aircraft wings. In conventional peen forming using spherical shots, axisymmetric plastic strain causes spherical deformation. A new peen forming method, vibration peening, using a rectangular solid pin was developed. In this method, the curvature radius R_x in the x-direction (short side of the pin) was smaller than the curvature radius R_z in the z-direction (long side of the pin). In vibration peening, the pin is projected onto the specimen using a hammer with a reciprocating motion. For a plate of aluminum alloy A7075-T6, the effects of the pin movement (x-direction speed v_x and z-direction travel pitch P_z ) and the parallel length L of the pin tip on 1/R_x and 1/R_z were investigated. To predict 1/R_x and 1/R_z from forming conditions, machine learning regression analysis (support vector regression) was performed using the experimental results of v_x , P_z , L, 1/R_x and 1/R_z as training data. Regression analysis verified that R_x and R_z were approximately predictable under the unlearned condition. Therefore, regression analysis can be used to select the appropriate forming conditions (v_x , P_z , L ) for achieving the target curvature (1/R_x and 1/R_z ).