1689 papers
This study proposes a graph neural network-based framework using multi-source data to identify and classify fine-grained rural settlement units in Dangyang, China, revealing their spatial network structure and key drivers to enable differentiated governance beyond traditional administrative village scales.
This study experimentally demonstrates that for thick-walled PP-R injection molding, conventionally machined CuBe alloy molds outperform selective laser melting (SLM) molds with conformal cooling across all economic metrics, offering superior cycle time reduction and cost savings despite a higher initial investment, while both advanced concepts successfully produce parts meeting functional qualification standards.
This paper presents a parameterized, physics-informed neural network that efficiently predicts transient thermal fields and melt pool geometry in SS316L Laser Powder Bed Fusion by combining FEM-generated data with heat conduction regularization to overcome the computational limitations of conventional simulations.
This paper analyzes the channel characteristics and performance of long-distance magnetic induction-based through-the-earth communication by establishing an equivalent circuit model, evaluating noise and frequency responses under various conditions, and determining optimal modulation schemes to enhance system reliability.
This paper proposes a position-weighted sequence-kernel Bayesian optimization framework that efficiently designs composite laminate stacking sequences under manufacturing constraints by leveraging a specialized distance metric for Gaussian process surrogates to maximize buckling performance while minimizing candidate redundancy.
This paper proposes FSMSN, a lightweight multi-scale network that integrates physical fault frequency priors into channel attention mechanisms to achieve robust, high-accuracy bearing fault diagnosis under severe noise interference with significantly fewer parameters than existing deep learning baselines.
This study physically characterized a facial vibrotactile stimulation device by quantifying the site-dependent mechanical displacement and non-uniform vibration at the contact surface, establishing that the stimulus cannot be accurately described by a single nominal value prior to clinical evaluation.
This paper proposes a stable, modular, and adaptive multi-model evaluation system that integrates dynamic, static, and hybrid approaches to accurately assess green building energy efficiency across diverse climates and scales, achieving a hybrid model error rate of under 6% while supporting both individual building optimization and large-scale portfolio management.
This study utilizes TCAD simulations to demonstrate that increasing the gate-to-channel distance in InAlAs/InGaAs/InP HEMTs generally reduces noise spectral densities and gate-to-drain capacitance while increasing output conductance and gate-to-source capacitance, whereas varying the gate bias produces distinct, often opposing, effects on current and voltage noise characteristics.
This paper proposes the RECINAT framework to assess and enhance the resilience of Nordic power grids against natural hazards like landslides by utilizing multi-objective optimization to strategically deploy Virtual Power Plants, thereby reducing economic losses and improving system redundancy.