Search for: All records

Abstract Adhesive bonding of composite materials has become increasingly crucial for advanced engineering applications, offering unique advantages for lightweight and high-performance designs. This study presents a novel framework, physics-informed failure mode proportion prediction (PIFMP) model, for predicting failure mode proportions in composite adhesive joints, addressing critical gaps in understanding mixed-mode failure behaviors. In contrast to conventional approaches that focus solely on force or stress prediction, this research integrates important parameters from multistage manufacturing processes (MMPs) and simulation data into a physics-informed machine learning (PIML) framework, enabling proactive failure prediction and design optimization. The proposed framework unifies data-driven machine learning models with features derived from finite element analysis (FEA), incorporating cohesive zone modeling (CZM) to capture the physical dynamics of adhesive behavior under lap shearing. By embedding FEA-based physics features into the machine learning process and leveraging a time-series transformer model to analyze the temporal progression of interfacial damage and separation, the framework ensures predictive accuracy and physics-informed consistency, enabling precise analysis of failure mechanisms. The empirical study validates the effectiveness and the reliability of the framework, demonstrating enhanced predictive performance through cross-validation. The work establishes a foundational approach for failure analysis and provides a robust basis for future advancements. 

The link between surface tectonic plates and mantle slabs is fundamental for paleo-tectonic reconstructions and for our understanding of mantle dynamics. Many seismic tomography-based studies have assumed vertical slab sinking and projected mantle features to the surface to reconstruct paleo-trench locations or explain tectonic features. Here, we used a slab-unfolding approach that does not require assumptions about sinking paths or rates to re-interpret the seismic structure of the Lesser Antilles slab underneath the Caribbean. A recent study invoked mainly vertical slab sinking and a highly folded and deformed slab to explain seismic Caribbean mantle structures. However, our results show that the upper-mantle Lesser Antilles slab structure can be better explained by limited intra-slab deformation and up to ~900 km lateral slab transport towards the northwest after subduction. Our results indicate that such lateral slab transport can occur even with probable weaknesses in the slab that originate from a subducted fossil ridge-transform system. We ascribe the lateral slab transport in the mantle to a kinematic connection with the North American plate, which has migrated northwestward since the Eocene. 

Cryoelectron tomography directly visualizes heterogeneous macromolecular structures in their native and complex cellular environments. However, existing computer-assisted structure sorting approaches are low throughput or inherently limited due to their dependency on available templates and manual labels. Here, we introduce a high-throughput template-and-label-free deep learning approach, Deep Iterative Subtomogram Clustering Approach (DISCA), that automatically detects subsets of homogeneous structures by learning and modeling 3D structural features and their distributions. Evaluation on five experimental cryo-ET datasets shows that an unsupervised deep learning based method can detect diverse structures with a wide range of molecular sizes. This unsupervised detection paves the way for systematic unbiased recognition of macromolecular complexes in situ. 

Abstract The importance of a low-viscosity asthenosphere underlying mobile plates has been highlighted since the earliest days of the plate tectonics revolution. However, absolute asthenospheric viscosities are still poorly constrained, with estimates spanning up to 3 orders of magnitude. Here we follow a new approach using analytic solutions for Poiseuille-Couette channel flow to compute asthenospheric viscosities under the Caribbean. We estimate Caribbean dynamic topography and the associated pressure gradient, which, combined with flow velocities estimated from geologic markers and tomographic structure, yield our best-estimate asthenospheric viscosity of (3.0 ± 1.5)*10 18 Pa s. This value is consistent with independent estimates for non-cratonic and oceanic regions, and challenges the hypothesis that higher-viscosity asthenosphere inferred from postglacial rebound is globally-representative. The active flow driven by Galapagos plume overpressure shown here contradicts the traditional view that the asthenosphere is only a passive lubricating layer for Earth’s tectonic plates.