The non-contact and directional nature of acoustic fault diagnosis offers a distinct advantage for fault detection in complex environments, ensuring the safe and stable operation of rotating machinery. However, challenges such as complex working conditions and data imbalance significantly hinder the effectiveness of acoustic-based diagnostic systems. To address these issues, this paper proposes a Collaborative Perception Information Aggregation Guidance Network (CPAGN). In the proposed method, a Stem layer is utilized at the front end to extract nonlinear acoustic features. Subsequently, the collaborative perception layer integrates local frequency information extraction, multi-receptive field temporal information fusion, and redundancy reduction, thereby capturing the time-frequency variations of acoustic signals under complex operating conditions and enhancing condition awareness. Finally, inter-layer feature correlation computation is employed to generate global channel guidance, facilitating the effective interaction and fusion of global information. The CPAGN successfully extracts critical acoustic features and identifies essential fault information, demonstrating remarkable robustness and generalizability in handling data imbalance. To validate the effectiveness of CPAGN, comparative experiments were conducted on two datasets, comparing it with several existing deep learning methods. The results indicate that CPAGN excels at capturing fault information from acoustic signals, achieving superior diagnostic performance in complex working conditions and imbalanced data scenarios.
1 August 2022 in Journal on Multimodal User Interfaces
Electrochemical deuteration utilizing D2O is a promising room-temperature strategy for synthesizing deuterated molecules. At present, high Faradic efficiencies (FEs) at high reaction rates are extraordinarily significant but highly challenged. Here, high-curvature low-coordinated copper nanotips (LC-Cu NTs) are designed to enable electroreductive deuteration of α-deuterio aryl acetonitriles, in situ generated from fast α-H/D exchange of acetonitriles in D2O, to α,β-deuterio arylethyl primary amines with 90% FE and 0.11 mmol h-1 cm-2, greatly outperforming the reported data and other Cu counterparts. The increased electric fields of the nanotips concentrate nitriles and K+(D2O)n, and low-coordination sites promote nitriles and D2O adsorption, thus facilitating nitrile deuteration with an excellent reaction rate. The higher coupling energy barrier of active hydrogen atoms at low-coordination sites restricts the D2 evolution, accounting for the outstanding FE. The wide substrate scope, easy gram-scale synthesis, and facile preparation of d4-melatonin with enhanced antitumor and antioxidation effects highlight its great promise. Furthermore, the application of LC-Cu NTs in other deuteration reactions with improved reaction rates and FEs rationalizes the design concept.
In complex product form design, cultural element integration can enhance design uniqueness and identity. Cultural stimuli—visual stimuli with cultural characteristics—are associated with cognitive processes involved in creativity. However, because cultural stimuli integrate semantic concept and visual form, the cognitive and neural patterns linking these dimensions to design creativity remain to be explored. This study examines these relationships in form design of rail vehicles, presenting cultural stimuli with varying semantic distances (SD) and analogical distances (AD) to 24 industrial-design students and teachers (18 students, 6 teachers). Multimodal data were collected, including eye-tracking, electroencephalography (EEG), expert sketch evaluations, and think-aloud protocols. Results indicate that: (1) cultural stimuli enhanced creativity ratings relative to no-stimulus conditions; (2) creativity exhibited a nonlinear trend with SD, peaking at medium SD for novelty; (3) AD moderated the SD effect, with significant SD effects under far-AD conditions that diminished under near-AD conditions. Specifically, far-AD stimuli (conceptual pictures) elicited N400 amplitudes in the frontal and central regions, indexing semantic integration demands. Near-AD stimuli (design examples), by contrast, showed enhanced P200 and P300 without corresponding N400 increases, consistent with structural matching. These findings are consistent with two distinct patterns observed between SD-AD configurations and design ideation: far AD with medium SD was associated with novelty and semantic integration, while near AD with medium SD was linked to completeness and structural matching. This framework advances understanding of stimulus-driven design cognition within the present sample and context and provides a theoretical foundation for future AI-assisted design applications. Future research should extend these patterns to diverse cultural contexts and investigate causal mechanisms through experimental manipulation.
This paper proposes a functional carriage design and an evaluation index system to improve the operational efficiency of high-speed medical trains. Hierarchical task analysis and human-machine-environment analysis were applied to model the transfer task and the functional modules of the medical train. The functional module configuration was obtained by performing a correlation analysis between the task and function. The relationship between carriages was elucidated by analysing material, personnel and information flow, and an optimal grouping diagram was obtained. Based on this design method, an innovative 6-carriage grouping design scheme was proposed. A functional evaluation index system for the carriage design was constructed, and the 6-carriage design was compared with the conventional 8-carriage design to verify the usability of the design method. The results showed that the 6-carriage high-speed trains can be flexibly configured to suit the changing task environment and are generally better than the 8-carriage design. This study provides theoretical and methodological support for constructing efficient and rational functional carriages for high-speed medical trains.