Simultaneous fault type and severity identification is critical for timely maintenance actions to prevent accidents from industrial machinery. The former can indicate occurrences of specific faults, and the latter can track early fault evolutions. Existing methods generally assume training and testing data are drawn from the same data distribution. However, in real industries, due to the change of working conditions, domain shift phenomenon can be triggered. The existing intelligent diagnosis methods are less effective in such scenarios for lack of domain adaptation ability. To address such problems, a novel two-branch domain adaptation network is developed. A deep convolutional neural network with two branches, as the main hierarchical architecture, is designed to handle two recognition tasks. The maximum mean discrepancy based multi-kernel learning is embedded to reduce the distribution discrepancy between the source domain and target domain. As such, the domain-invariant features with a hierarchical structure can be effectively extracted and the fault types and fault severities can be recognized at the same time. Experiments on a bearing dataset, a gear dataset and a motor bearing dataset are carried out to validate the effectiveness of the proposed approach. The results demonstrate that the proposed method can effectively perform fault type and severity identification simultaneously and obviously outperforms other state-of-the-art methods.
We report on the fabrication of ion exchangeable microstructures by femtosecond laser direct writing of an ion exchange photopolymer, poly(2-acrylamido-2-methyl-1-propanesulfonic acid) (PAMPS). The resultant microstructures are negatively charged in aqueous solution, and can adsorb positively charged species, such as metal ions, nanoparticles, and proteins by electrostatic interaction, forming functional components for chip functionalization. In addition, it is possible to modify the microstructures with positively charged species that make the microstructures sensitive to negatively charged species. As a typical example, a crossed 3D microvessel functionalized with antibodies was fabricated, which reveals great potential for organ-on-a-chip systems. The fabrication of ion exchangeable microstructures holds great promise for flexible chip functionalization.
Compound eyes are unique optical imaging systems that consist of numerous separate light-sensitive units (ommatidia). Attempts have been made to produce artificial compound eyes via advanced 3D nanotechnologies. Among them, femtosecond laser direct writing (FsLDW) technology has emerged as an effective strategy due to its distinct advantages in 3D designable and high precision fabrication capability. However, the point-by-point scanning process results in a very low fabrication efficiency, limiting the practical applications of the FsLDW technology. To solve this problem, we propose a high-efficiency method for the mass production of 3D artificial compound eyes using a photopolymer template fabricated by FsLDW. The resultant 3D SU-8 compound eye templates could be used to replicate polydimethylsiloxane (PDMS) compound eyes many times (over 50 times) with a highly improved efficiency (nearly 20 times higher than the efficiency of direct fabrication using the point-by-point FsLDW). The PDMS replicas showed good focusing and imaging performances. We anticipate that this method may serve as an enabler for the mass production of 3D artificial compound eyes and promote their practical applications in the near future.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.