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.

The demand for new high-performance materials in industries such as aviation, navigation, nuclear power etc., is continuously increasing. Indentation technology, as a non-destructive testing technique, is widely used to test the mechanical properties of new materials. Generally, the mechanical properties of material can be described by material constitutive relation model. The parameters of constitutive relation model can be recognized via inverse measurement technology which combines indentation measurement and finite element method. The inversion measurement process is actually to continuously adjust the constitutive parameters of the materials using optimization algorithms to minimize the difference between the indentation test results and the simulation results. However, in the process of inversion calculation, the single optimization algorithms have various deficiencies, such as weak global search ability, slow search speed, low computational accuracy, etc. In response to the above problems, a joint optimization algorithm methodology is proposed in this paper. First, the characteristics of several typical optimization algorithms and their applicability in material parameter identification are compared and analyzed, including quasi-Newton method, conjugate gradient method, Gaussian-Newton method, Levenberg-Marquardt method, complex method and genetic algorithm. Second, the joint optimization algorithm is established based on the single algorithms by integrating the advantages of different algorithms. Finally, the effectiveness of the joint optimization algorithm is verified by an example, which measures the constitutive relationship of elastic-plastic material. The result shows that the joint optimization algorithm can significantly improve the efficiency and the accuracy in the inversion identification.