At present, the detection accuracy of bolt-loosening diagnoses is still not high. In order to improve the detection accuracy, this paper proposes a fault diagnosis model based on the TSCNN model, which can simultaneously extract fault features from vibration signals and time-frequency images and can precisely detect the bolt-loosening states. In this paper, the LeNet-5 network is improved by adjusting the size and number of the convolution kernels, introducing the dropout operation, and building a two-dimensional convolutional neural network (2DCNN) model. Combining the advantages of a one-dimensional convolutional neural network (1DCNN) with wide first-layer kernels to suppress high-frequency noise, a two-stream convolutional neural network (TSCNN) is proposed based on 1D and 2D input data. The proposed model uses raw vibration signals and time-frequency images as input and automatically extracts sensitive features and representative information. Finally, the effectiveness and superiority of the proposed approach are verified by practical experiments that are carried out on a machine tool guideway. The experimental results show that the proposed approach can effectively achieve end-to-end bolt-loosening fault diagnoses, with an average recognition accuracy of 99.58%. In addition, the method can easily achieve over 93% accuracy when the SNR is over 0 dB without any denoising preprocessing. The results show that the proposed approach not only achieves high classification accuracy but also has good noise immunity.

Pipe flanging is needed extensively in the industrial field, and its technological requirements are complicated and changeable. Currently the special flanging die and equipment are less flexible, while the whole set of equipment cost is high. As a result, pipe flanging technology with strong adaptability is necessary in single-piece or small-batch production. In this paper, a multi-pass progressive flanging process of round metal pipes with ball-end tool bar as forming tool is introduced, and dieless flanging of pipe can be realized by means of numerical control machine. In the dieless flanging process, the feedback and control of flanging force is a key factor for the flanging quality to meet design requirements. Therefore, this paper studies the change of flanging force in the multi-pass progressive flanging process of round metal tubes by combining theory, simulation and experiments. Based on the process of multi-pass progressive flanging of round pipes, an analytical model of the forming force of multi-pass progressive flanging of round pipes is established by taking the micro-element contacting tool bar with pipe as the research object. With the commercial ABAQUS finite element software, the multi-pass progressive flanging process of round pipes is simulated, and the evolution data of stress and strain field of the pipe during flanging process are obtained. Finally, a multi-pass progressive flanging test system is set up, and the three-dimensional forming force on the tool bar in flanging is measured. The flanging force measured by analytical calculation, finite element simulation and experiment is in good agreement. When the flanging angle is small, the radial forming force is the largest, followed by the axial forming force. As the flanging angle increases, the difference between radial forming force and axial forming force decreases gradually. When the flanging angle is greater than 30 to 40 degrees, the axial forming force becomes the largest, followed by the radial forming force.

Visual Simultaneous Localization and Mapping(SLAM) technology is widely used in the autonomous navigation of mobile robots. However, in the face of increas