Search

The recognition of color differences in solar cells with complex textures is a significant challenge in cell manufacturing. Traditional methods fail to detect the color difference effectively. Deep learning models have exhibited promise in many engineering fields. A multi-component attention-based convolution approach is proposed for the surface inspection based on the feature information in different color spaces. Wavelet entropy is employed to represent the information of different components, remove redundant components and extract effective feature information. Additionally, a residual attention mechanism is developed to capture local features with contextual semantic information. The best network structure is determined by evaluating the layer depth of the basic model and convolution kernel size. A multi-component network model is constructed based on the formed structure to improve the ability to distinguish different color difference features. Experimental results indicated that the proposed approach exhibits competitive performance. The research solution provides guidance for applications of deep learning to improve the quality of solar cells in manufacturing. [ABSTRACT FROM AUTHOR]

Convolutional Neural Networks have promoted development of the fault diagnosis in the machine prognostics and health management. However, the existing methods have limited applicability under strong noisy conditions. We propose an attention-based ConvNeXt with a parallel multiscale dilated convolution residual module for the rotor fault diagnosis. The parallel multiscale dilated convolution residual module is firstly introduced to filter noise and extract multiscale discriminative features intelligently. The multi-head attention module is then embedded in ConvNeXt for global discriminative features adaptively. Focal Loss is finally implemented for identifying hard-to-classify samples to further improve the diagnostic accuracy. The application results demonstrate the superiority and robustness of our method in two case studies. [ABSTRACT FROM AUTHOR]

Predicting the remaining useful life of manufacturing tools based on condition maintenance is a critical task in prognostics and health management systems. Existing methods have limitations in modeling variable-length sequence data and extracting sufficient local and global dependencies. To address these issues, a temporal convolutional approach with residual multi-head attention mechanism is proposed. The temporal convolution efficiently models variable-length sequence data and extracts long-term dependency features in the time dimension. The residual multi-head attention module captures local and global dependency features in the sequence features and extracts the tool key degradation features, which further improves the performance of the model. To validate the predictive performance of the temporal convolutional approach with residual multi-head attention mechanism, two manufacturing tool datasets are used in the remaining useful life experiments. The results demonstrate the effectiveness and superiority of the proposed method compared to other advanced methods. [ABSTRACT FROM AUTHOR]

As a typical combinatorial optimization problem, two-dimensional (2D) irregular layout is a common problem in the engineering field. No-fit polygon (NFP) is a common geometric tool for solving layout problems. Although it is necessary to calculate NFP frequently in the process of layout, the search efficiency of the NFP generation algorithm is affected. The existing unfitting polygon algorithms have difficulties in dealing with complex contours, especially irregular contours with cavities. The large-scale layout process takes the long time with the low material utilization. A vector superposition NFP (VS-NFP) method is proposed in this paper to improve the solution efficiency of 2D irregular layout problems with complex contours. Based on the VS-NFP, an improved mixed integer programming (MIP) model is built. The model increases compression constraints for high solution search efficiency. A hybrid algorithm based on the VS-NFP and MIP is proposed to solve the problem. Comparing with the existing methods, the proposed method shortens the search time and improves the material utilization. The proposed method is verified in the application of large-scale layout problems. [ABSTRACT FROM AUTHOR]

Orthonormal basis function (OBF) decomposition method and minimum entropy (ME) detector are two typical methods of magnetic anomaly detection. The OBF detector works effectively only when the assumptions of the target signal and noise are appropriate, and the ME detector is limited by low signal-to-noise ratio (SNR). In order to improve magnetic anomaly detection performance in the case of low SNR and no prior information of the target signal, we proposed a novel detector by using stochastic resonance (SR) method and named it SR detector in the paper. The SR detector consists of a bistable SR system and a corresponding receiver. Firstly, the noise is used to enhance the magnetic anomaly signal with the help of SR system, instead of being suppressed in a traditional method; then the anomaly signal can be detected more effectively by the receiver. Experimental results show that the SR detector did work well, its detection probability approximated to 70% even if the input SNR was −3 dB, and was about 100% if the input SNR was 0dB, when the false alarm rate (FAR) was 1.5%. Furthermore, the SR detector provided higher detection probability than the ME detector under the same basis. In a word, due to the good detection performance and simple implementation, the SR detector would be more attractive in practice. [ABSTRACT FROM AUTHOR]

The tension–compression asymmetry of metallic glasses has shown that the Mohr–Coulomb yield criterion can better determine the fracture characteristics of metallic glasses than the traditional Tresca and Von Mises criterion. Available cutting models based on the traditional yield criterion cannot describe the cutting mechanism of metallic glasses. An orthogonal cutting mechanics model of metallic glasses based on the Mohr–Coulomb yield criterion taking into account the influence of normal stress and temperature on shear plane is established. The analytical expressions for cutting force, shear angle and average temperature rise on shear plane are derived. Orthogonal turning experiments were designed and carried to validate the developed model. The analytically calculated cutting forces from the developed model considering cutting temperature rise were compared with the measured forces and the results show that a small difference of 7.7% was found, which is contrary to the prediction from traditional model based on von Mises yield criterion (nearly 60%) and the prediction from the model based on M–C without considering cutting temperature rise (more than 70%). The comprehensive comparison results show that the developed analytical model can accurately predict cutting process for metallic glasses. [ABSTRACT FROM AUTHOR]