Search

In order to study the influence of defect geometric parameters (width, thickness) on the electromagnetic characteristics of the space around the defect in the process of eddy current testing of metal spheres, based on the principle of eddy current testing, a metal sphere detection model with different defect parameters was established by COMSOL software for simulation. The distribution of the magnetic field around the defect under different parameters is analyzed with the help of the butterfly diagram composed of the radial magnetic induction intensity and the axial magnetic induction intensity. The results show that: When the width of the corrosion defect is different, the butterfly diagram will change differently. When the width is small, the butterfly diagram basically does not change. When the width reaches a certain value, the butterfly diagram begins to change significantly. The intersection of the graph and the initial butterfly graph is defined as the influence node, and the width of the defect can be quantitatively analyzed according to the sequence of the influence nodes. The influence of the thickness of the corrosion defect on the magnetic field on the surface of the sphere is different under different defect widths. The larger the thickness of the corrosion defect, the greater the influence on the magnetic field near the center of the defect, and the deeper the depression of the butterfly diagram.

Sufficient labeled fault samples are the key to ensuring the performance of deep learning diagnostic models. However, in practical engineering applications, machinery and equipment operate normally most of the time, and it is difficult to collect enough fault data. In contrast, through the kinetics analysis method, simulation data sets of various fault types can be easily obtained. However, there is a difference in distribution between simulation data and real data. The deep learning diagnosis model trained directly with simulation data lacks versatility and cannot be applied to fault diagnosis of real data. To this end, this paper proposes a simulation data-driven deep transfer learning fault diagnosis method, which applies the fault diagnosis knowledge in the simulation data to the real data fault diagnosis task. The effectiveness of the method is verified by experiments on the two-stage planetary gearbox in the Drivetrain Diagnostics Simulator (DDS) test bench.

In this paper, we take a novel two-translation parallel mechanism based on variable length parallelogram structure as the research object, and carry out the research on the dynamic characteristics of the mechanism. Based on the principle of virtual work, the dynamic model of the new two-translation parallel mechanism is established. By numerically solving the dynamic model, the dynamic behavior of the new mechanism with impact characteristics is found, which provides a theoretical basis for optimizing this new mechanism.