A5016222684- Railway Engineering and Dynamics
- Structural Health Monitoring Techniques
- Probabilistic and Robust Engineering Design
- Mechanical stress and fatigue analysis
- Hydraulic and Pneumatic Systems
- Machine Learning in Materials Science
- Advanced ceramic materials synthesis
- Railway Systems and Energy Efficiency
- Civil and Geotechnical Engineering Research
- Adhesion, Friction, and Surface Interactions
- Engineering Applied Research
- Vibration and Dynamic Analysis
- Underwater Acoustics Research
- Materials Engineering and Processing
- Mechanical and Thermal Properties Analysis
- Marine animal studies overview
- Structural Engineering and Vibration Analysis
- Railway Systems and Materials Science
- Advanced Measurement and Metrology Techniques
- Iterative Learning Control Systems
- Protein Structure and Dynamics
- Evacuation and Crowd Dynamics
- Computational Drug Discovery Methods
- Infrastructure Maintenance and Monitoring
- Wind Energy Research and Development
Harvard University
2022-2024
Harvard University Press
2022-2023
Robert Bosch (United Kingdom)
2022
University of Oslo
2019
Chalmers University of Technology
2008-2017
University of Bristol
2014-2016
SSPA (Sweden)
2015
Swedish Defence Research Agency
2010
Smart High Tech (Sweden)
2004-2007
University of Southampton
2004
A simultaneously accurate and computationally efficient parametrization of the potential energy surface molecules materials is a long-standing goal in natural sciences. While atom-centered message passing neural networks (MPNNs) have shown remarkable accuracy, their information propagation has limited accessible length-scales. Local methods, conversely, scale to large simulations but suffered from inferior accuracy. This work introduces Allegro, strictly local equivariant deep network...
This literature survey discusses the state-of-the-art in research on why out-of-round railway wheels are developed and damage they cause to track vehicle components. Although term can be attributed a large spectrum of different wheel defects, focus here is with long wavelengths, such as so-called polygonalization 1-5 harmonics (wavelengths) around circumference. Topics dealt include experimental detection wheel/rail impact loads, mathematical models predict development consequences wheels,...
Summary High-frequency train-track interaction and mechanisms of wheel/rail wear that is non-uniform in magnitude around/along the running surface are surveyed. Causes, consequences suggested remedies to relieve problems discussed for three types irregular wear: (1) short-pitch rail corrugation on tangent tracks large radius curves, (2) wheel as caused by tread braking, (3) polygonalisation. The state-of-the-art modelling dynamic conjunction with prediction reviewed.
The influence of different types railway wheel out-of-roundness (OOR) on the vertical dynamic wheel-rail contact force and track response is investigated through extensive field tests numerical simulations. from a freight train, provided with number severe tread damage, studied. Two axle loads are used in combination train speeds range 30-100km/h. defects wheelflats, local spalls due to rolling fatigue cracking, long polygonal wheels (periodic OOR). was measured using strain gauge based...
Abstract Machine learning interatomic force fields are promising for combining high computational efficiency and accuracy in modeling quantum interactions simulating atomistic dynamics. Active methods have been recently developed to train efficiently automatically. Among them, Bayesian active utilizes principled uncertainty quantification make data acquisition decisions. In this work, we present a general workflow, where the field is constructed from sparse Gaussian process regression model...
Abstract Metal surfaces have long been known to reconstruct, significantly influencing their structural and catalytic properties. Many key mechanistic aspects of these subtle transformations remain poorly understood due limitations previous simulation approaches. Using active learning Bayesian machine-learned force fields trained from ab initio calculations, we enable large-scale molecular dynamics simulations describe the thermodynamics time evolution low-index mesoscopic surface...
Ionic liquids (ILs) are an exciting class of electrolytes finding applications in many areas from energy storage to solvents, where they have been touted as "designer solvents" can be mixed precisely tailor the physiochemical properties. As using machine learning interatomic potentials (MLIPs) simulate ILs is still relatively unexplored, several questions need answered see if MLIPs transformative for ILs. Since often not pure, but either together or contain additives, we first demonstrate...
Abstract Understanding the electrical and thermal transport properties of materials is critical to design electronics, sensors, energy conversion devices. Computational modeling can accurately predict material but, in order be reliable, requires accurate descriptions electron phonon states their interactions. While first-principles methods are capable describing spectrum each carrier, using them compute still a formidable task, both computationally demanding memory intensive, requiring...
This work brings the leading accuracy, sample efficiency, and robustness of deep equivariant neural networks to extreme computational scale. is achieved through a combination innovative model architecture, massive parallelization, models implementations optimized for efficient GPU utilization. The resulting Allegro architecture bridges accuracy-speed tradeoff atomistic simulations enables description dynamics in structures unprecedented complexity at quantum fidelity. To illustrate...
Dropped head in parkinsonism has been attributed to dystonia or unbalanced muscle rigidity. To our knowledge, isolated neck extensor myopathy with described only one patient.To assess the occurrence of extension weakness resulting dropped patients and explore whether drop might be consequence myopathy.All who were evaluated because Department Neurology hospital between January 1, 1997, December 31, 1999, found have both studied. The underwent clinical examination, blood tests including...
Current societal requirements necessitate the effective delivery of complex projects that can do more while using less. Yet, recent large-scale project failures suggest our ability to successfully deliver them is still at its infancy. Such be seen arise through various failure mechanisms; this work focuses on one such mechanism. Specifically, it examines likelihood a sustaining catastrophe, as triggered by single task and delivered via cascading process. To so, an analytical model was...
Numerical models that can be used to evaluate crack initiation in rails owing rolling contact loads are of great value. Very few on rail fatigue assessments exist the literature consider non-linear stress and strain response caused by both global dynamic track local wheel-rail loads. In present investigation, such a tool has been developed using finite element (FE) models. The calculate governing conditions as residual stresses, plastic strains orientation planes for initiation. ability...
Estimating city evacuation time is a nontrivial problem due to the interaction between thousands of individual agents, giving rise various collective phenomena, such as bottleneck formation, intermittent flow, and stop-and-go waves. We present mean field approach draw relationships road network spatial attributes, number evacuees, resultant estimate (ETE). Using volunteered geographic information, we divide 50 United Kingdom cities into total 704 catchment areas (CAs) which define an area...
Three well-known ratchetting models for metals with different hardening rules were calibrated using uniaxial experimental data from Bower (1989) [J. Mech. Phys. Solids, 31, pp. 455–470], and implemented in the FE code ABAQUS (Hibbitt et al., 1997 [ABAQUS Version 5.7]) to predict results a tension-torsion specimen. The integrated numerically by implicit Backward Euler rule, material parameters via optimization data. algorithmic tangent stiffnesses of derived obtain efficient implementations....
Underwater surveillance in a harbor is typically performed using active sonar systems. The performance of an can drastically change due to rapid variations the sound propagation. This paper presents method for detecting divers with open circuit breathing systems passive acoustics. authors have previously reported on acoustic diver detection that employs two hydrophones. proposed uses single hydrophone, resulting simpler and cheaper system. works by whitening background noise mitigating...