As a process for producing seamless tubes, the tandem skew rolling (TSR) process was proposed. In order to study deformation characteristics and mechanism on tubes obtained by the TSR process, a numerical simulation of the process was analyzed using Deform-3D software. Simulation results demonstrated the distribution of stress, strain, velocity, and temperature of a seamless tube in the stable stage during the TSR process. Actual experiments of carbon steel 1045, high strength steel 42CrMo, and magnesium alloy AZ31 were carried out in a TSR testing mill. The results demonstrated that the TSR process is qualified for producing tubes of high quality, with an accuracy of ±0.2 mm in wall thickness and ±0.35 mm in diameter. This process is suitable for manufacturing seamless tubes that are difficult to deform or that have been deformed in a narrow range of temperature.

Surface quality control of hot-dip galvanized sheets is a critical research topic in the metallurgical industry. Zinc dross, the most common surface defect in the hot-dip galvanizing process, significantly affects the sheet’s service performance. In this manuscript, a novel magnetic flux leakage (MFL) detection method was proposed to detect zinc dross defects on the surface of hot-dip galvanized steel sheets. Instead of using exciting coils in traditional methods, a tiny permanent magnet with a millimeter magnitude was employed to reduce the size and weight of the equipment. Additionally, a high-precision tunnel magnetoresistance (TMR) sensor with a sensitivity of 300 mV/V/Oe was selected to achieve higher detection accuracy. The experimental setup was established, and the x-axis direction (sample movement direction) was determined as the best measurement axis by vector analysis through experiments and numerical simulation. The detection results indicate that this novel MFL detection method could detect industrial zinc dross with an equivalent size of 400 μm, with high signal repeatability and signal-to-noise ratio. In the range of 0–1200 mm/s, the detection speed has almost no effect on the measurement signal, which indicates that this novel method has higher adaptability to various conditions. The multi-path scanning method with a single probe was used to simulate the array measurement to detect a rectangular area of 30 × 60 mm. Ten zinc dross defects were detected across eight measurement paths with 4 mm intervals, and the positions of these zinc dross defects were successfully reconstructed. The research results indicate that this novel MFL detection method is simple and feasible. Furthermore, the implementation of array measurements provides valuable guidance for subsequent in-depth research and potential industrial applications in the future.

To investigate the forming mechanisms and force conditions of a tube blank during tandem skew rolling (TSR), force analysis diagrams were created for each stage of the process. Force equilibrium and rotation conditions for the forming coordination of each rolling stage were calculated and obtained. The finite element method (FEM) and experimental research method were used, and four groups of process parameters were designed for the comparative analysis. The finite element simulations reveal the deformation characteristics of the tube during the rolling process, and provide a reference for the experiment. The results showed that Groups 1 and 3 successfully completed the rolling process, resulting in hollow tubes with good surface quality. Compared to Group 1, the torsion observed in Group 2 is caused by a decrease in the roll gap of the rolling section, which reduces the volume of the metal passing through the rolling section. In comparison to Group 3, the tensile cracking observed in Group 4 was attributed to an increase in the roll rotational speed of the rolling section, which accelerated the axial metal flow and caused overtension in the piercing section. Further analysis of the rolling force curves from Group 3 revealed that the rolling force on the entry side of the piercing section rollers was greater than that on the exit side. From the analysis of the structure and tensile mechanical property, the fiber structures of Groups 1 and 3 are consisting of ferrite and pearlite, while the finer microstructure of Group3 corresponds to a higher yield strength and tensile strength.

Based on the third generation neural network spiking neural network, this paper optimizes and improves a classification and coding method, and proposes an image recognition method. Firstly, the read image is converted into a spike sequence, and then the spike sequence is encoded in groups and sent to the neurons in the spike neural network. After learning and training for many times, the quantization standard code is obtained. In this process, the spike sequence transformation matrix and dynamic weight matrix are obtained, and the unclassified data are output through the same matrix for image recognition and classification. Simulation results show that the above methods can get correct coding and preliminary recognition classification, and the spiking neural network can be applied.