Search

Intelligent machinery fault diagnosis is one of the key technologies for the transformation and competitiveness of traditional factories. Complex production environments make it difficult to maintain good prediction performance using traditional methods. This paper proposes a deep convolutional neural network combined with an adaptive environmental noise method to achieve robust fault classification. The proposed method uses six-dimensional physical signals for data fusion and feature fusion, extracts obvious features and enhances subtle features, and uses continuous wavelets and Gramian angular fields to transform signals with different physical and frequency characteristics into time–frequency maps and two-dimensional images. The fusion technology of different signals can provide comprehensive features for fault prediction, improving upon the blind spots of traditional methods to extract features, and then perform prediction and classification through deep convolutional neural networks. In the experiment, the tool failure classification of the dental milling machine is used as a verification case. The results show that the prediction accuracy of the proposed method is nearly 100%, much better than other comparison methods. In addition, white noise was added in the experiment to verify the noise immunity of the model. The results show that the accuracy of the proposed method is 99%, which is better than other comparison methods in terms of accuracy and robustness, proving the effectiveness of the proposed method for fault diagnosis and classification. [ABSTRACT FROM AUTHOR]

Whether altered white matter microstructural property of autistic people also exists in non-autistic siblings is uncertain. The microstructures of a neural tract may not be consistent throughout the whole track. We assessed 38 cognitive-able autistic males (aged 15.8 ± 4.4 years), 39 non-autistic siblings (16.5 ± 5.7 years), and 78 age- and sex-matched non-autistic comparison people (14.4 ± 5.3 years) using tract-based automatic analysis of diffusion spectrum imaging and threshold-free cluster-weighted method. First, we identified segments within the right frontal aslant tract, frontostriatal tract, and thalamic radiation to precentral areas in both autistic people and non-autistic siblings that differed from those in non-autistic comparison people. Second, segments within bilateral cingulate gyri and callosal fibers connecting superior temporal lobes differed between autistic people and non-autistic comparison people but not between siblings and non-autistic comparison people. Third, segments within the left inferior longitudinal fasciculus and callosal fibers connecting precuneus showed increased generalized fractional anisotropy in non-autistic siblings. Our findings suggest microstructural properties of some potential neural segments that were similar between autistic people and their non-autistic siblings may serve as intermediate phenotypes of autism, facilitating further etiological searching for autism. Meanwhile, increased microstructural properties in unaffected siblings alone might indicate compensatory processes in the light of genetic predisposition for autism. White matter is the neural pathway that connects neurons in different brain regions. Although research has shown white matter differences between autistic and non-autistic people, little is known about the properties of white matter in non-autistic siblings. In addition, past studies often focused on the whole neural tracts; it is unclear where differences exist in specific segments of the tracts. This study identified neural segments that differed between autistic people, their non-autistic siblings, and the age- and non-autistic people. We found altered segments within the tracts connected to anterior brain regions corresponding to several higher cognitive functions (e.g. executive functions) in autistic people and non-autistic siblings. Segments connecting to regions for social cognition and Theory of Mind were altered only in autistic people, explaining a large portion of autistic traits and may serve as neuroimaging markers. Segments within the tracts associated with fewer autistic traits or connecting brain regions for diverse highly integrated functions showed compensatory increases in the microstructural properties in non-autistic siblings. Our findings suggest that differential white matter segments that are shared between autistic people and non-autistic siblings may serve as potential "intermediate phenotypes"—biological or neuropsychological characteristics in the causal link between genetics and symptoms—of autism. These findings shed light on a promising neuroimaging model to refine the intermediate phenotype of autism which may facilitate further identification of the genetic and biological bases of autism. Future research exploring links between compensatory segments and neurocognitive strengths in non-autistic siblings may help understand brain adaptation to autism. [ABSTRACT FROM AUTHOR]