MRC Laboratory of Molecular Biology
CUED - Cambridge University Engineering DepartmentNot a member yet
45551 research outputs found
Sort by
Grand canonical approach to modeling hydrogen trapping at vacancies in alpha-Fe
No full textVacancies in iron are hydrogen traps, important in the understanding of hydrogen embrittlement of steel. We present a grand canonical approach to computing the trap occupancy as a function of both temperature and hydrogen concentration from practically zero to super-saturation. Our method couples a purpose-made machine-learned H-Fe potential, which enables rapid sampling with near density functional theory accuracy, with a statistical mechanical calculation of the trap occu- pancy using the technique of nested sampling. In contrast to the conventional assumption (based on Oriani theory) that at industrially relevant hydrogen concentrations and ambient conditions vacancy traps are are fully occupied, we find that vacancy traps are less than fully occupied under these conditions, necessitating a reevaluation of how we think about “mobile hydrogen” in iron and steel
A phase field model for elastic-gradient-plastic solids undergoing hydrogen embrittlement
No full textWe present a gradient-based theoretical framework for predicting hydrogen assisted fracture in elastic-plastic solids. The novelty of the model lies in the combination of: (i) stress-assisted diffusion of solute species, (ii) strain gradient plasticity, and (iii) a hydrogen-sensitive phase field fracture formulation, inspired by first principles calculations. The theoretical model is numerically implemented using a mixed finite element formulation and several boundary value problems are addressed to gain physical insight and showcase model predictions. The results reveal the critical role of plastic strain gradients in rationalising decohesion-based arguments and capturing the transition to brittle fracture observed in hydrogen-rich environments. Large crack tip stresses are predicted, which in turn raise the hydrogen concentration and reduce the fracture energy. The computation of the steady state fracture toughness as a function of the cohesive strength shows that cleavage fracture can be predicted in otherwise ductile metals using sensible values for the material parameters and the hydrogen concentration. In addition, we compute crack growth resistance curves in a wide variety of scenarios and demonstrate that the model can appropriately capture the sensitivity to: the plastic length scales, the fracture length scale, the loading rate and the hydrogen concentration. Model predictions are also compared with fracture experiments on a modern ultra-high strength steel, AerMet100. A promising agreement is observed with experimental measurements of threshold stress intensity factor Kth over a wide range of applied potentials
Is automation stealing manufacturing jobs? Evidence from South Africa's apparel industry
No full textThere are growing fears that automation will lead to major job displacement and increasing unemployment, particularly in labour-intensive manufacturing. This is especially worrying for developing countries because of the importance of labour-intensive manufacturing to economic development. The purpose of this paper is to evaluate the threat of automation to employment, focusing on the manufacturing sector. It does so by critically reviewing studies that evaluate the current and future impact of automation on employment, as well as using industry-specific evidence from the apparel industry in South Africa. Through our literature review, we find that many studies fail to acknowledge: (1) the full range of factors that determine the net impact of automation on employment and; (2) many country-specific and industry-specific barriers to adopting new automation technologies, particularly in developing countries. A qualitative case study such as this one could therefore be a valuable contribution to the literature. The apparel industry has been chosen as a case because many forecast studies predict that the industry will suffer huge job losses due to automation. South Africa has been chosen as a case because the current literature is lacking case study evidence in the context of developing countries, and because South Africa is among those few developing countries adopting automation technologies in the apparel industry. Our evidence draws on 26 interviews with firm managers in the South African apparel industry, as well as with government and union representatives. We find that the overall impact of automation on unemployment has been negligible and is predicted to continue to be negligible. But in some instances, increased automation has and is predicted to increase employment by improving productivity at the firm level
Uplift Resistance of Enlarged Base Pile Foundations
No full textPile foundations with enlarged bases are effective either in expansive soils or when piles are liable to attract tensile loading. In this paper, the uplift behaviour of enlarged base piles (referred to as under-reamed piles in the text) in layered strata has been experimentally investigated, using half-space small-scale 1g models, and compared with theoretical predictions developed using upper and lower bound theorems. The failure mechanisms observed were characterised using particle image velocimetry (PIV) techniques. It was observed that an upper stratum of clay overlying sand increased the post-peak uplift capacity when compared to two-layered sand, but had little effect at small strain. The small strain failure mechanisms observed were similar, consisting of a balloon-shaped bulb above the pile base, before extending vertically and radially out from the pile with increased displacement. This extension was seen to be wider in the clay than in the sand. These results show that when considering the design of under-reamed pile foundations, anchoring the base of the pile in a stronger stratum at greater depth is beneficial for increasing uplift capacity. At small strains, the overlying stratum has little consequence on the peak uplift resistance. However, at larger strains it can influence the post-peak uplift capacity, which would be an important consideration for possible situations of large imposed displacements
Mechanised Tunnel Excavation Through an Instrumented Site in Rome: Class A Predictions and Monitoring Data
No full textContract T3 of Line C of Rome underground is currently under construction. The running tunnels, excavated using two EPB shields, cross the historical centre of the city, potentially interacting with the existing monuments of great historic value. For this reason, two fully instrumented control sections were established at the beginning of this contract, in representative ground conditions. In addition to a greenfield section, an embedded barrier of bored piles running parallel to the tunnel axis was monitored, to study its effectiveness for settlement mitigation. Class A predictions of the passage of the tunnels through the control sections were carried out using a recently developed advanced numerical procedure, modelling in detail the main physical processes occurring around the shield, including cutter-head overcut, shield tapering and tail void grouting. Hypoplastic laws, calibrated with all the available data from the geotechnical investigation, were employed to model the behaviour of the soil. In this paper, the numerical results are compared with preliminary data gathered during the passage of the TBMs, showing a good qualitative agreement and suggesting strategies to improve the predictions
Thermal modeling and design optimization of PCB vias and pads
No full textMiniature power semiconductor devices mounted on printed circuit boards (PCBs) are normally cooled by means of PCB vias, copper pads, and/or heatsinks. Various reference PCB thermal designs have been provided by semiconductor manufacturers and researchers. However, the recommendations are not optimal, and there are some discrepancies among them, which may confuse electrical engineers. This paper aims to develop analytical thermal resistance models for PCB vias and pads, and further to obtain the optimal design for thermal resistance minimization. First, the PCB via array is thermally modeled in terms of multiple design parameters. A systematic parametric analysis leads to an optimal trajectory for the via diameter at different PCB specifications. Then, an axisymmetric thermal resistance model is developed for PCB thermal pads where the heat conduction, convection, and radiation all exist; due to the interdependence between the conductive/radiative heat transfer coefficients and the board temperatures, an algorithm is proposed to fast obtain the board-ambient thermal resistance and to predict the semiconductor junction temperature. Finally, the proposed thermal models and design optimization algorithms are verified by computational fluid dynamics simulations and experimental measurements
Stochastic trust region inexact Newton method for large-scale machine learning
No full textNowadays stochastic approximation methods are one of the major research direction to deal with the large-scale machine learning problems. From stochastic first order methods, now the focus is shifting to stochastic second order methods due to their faster convergence and availability of computing resources. In this paper, we have proposed a novel Stochastic Trust RegiOn Inexact Newton method, called as STRON, to solve large-scale learning problems which uses conjugate gradient (CG) to inexactly solve trust region subproblem. The method uses progressive subsampling in the calculation of gradient and Hessian values to take the advantage of both, stochastic and full-batch regimes. We have extended STRON using existing variance reduction techniques todeal with the noisy gradients and using preconditioned conjugate gradient (PCG) as subproblem solver, and empirically proved that they do not work as expected, for the large-scale learning problems. Finally, our empirical results prove efficacy of the proposed method against existing methods with bench marked datasets
A Single- and Three-Phase Grid Compatible Converter for Electric Vehicle On-Board Chargers
No full textThis paper proposes a voltage-source converter for an on-board Electric Vehicle (EV) charger which is compatible with both the single- and three-phase (1-ϕ and 3-ϕ) grids. The classic 3-ϕ active AC-DC rectifier circuit is used for both the 1-ϕ and 3-ϕ connection, but a new control scheme and LCL filter are designed to address the double-line frequency power pulsation issue caused by a 1-ϕ grid without using bulky DC capacitors. The third leg of the circuit is utilized to control the power pulsation in conjunction with stored energy in the LCL filter between the grid and charger rectifier. Neither additional active nor passive components are required. For the 3-ϕ connection, the rectifier is under balanced operation; when connected with the 1- ϕ grid, all three legs are controlled cooperatively as a 3-ϕ rectifier but under unbalanced operation to recreate the 1-ϕ voltage. Hence advantages from the 3-ϕ rectifier such as space vector pulse width modulation (SVPWM) and Y/Δ transformation can be utilized to increase utilization of DC-link voltage and filter capacitance respectively. The operation principle, control, and LCL filter design are reported and validated by both simulation and experiments of a 3-kW porotype
Machine learning and artificial neural network accelerated computational discoveries in materials science
No full textArtificial intelligence (AI) has been referred to as the “fourth paradigm of science,” and as part of a coherent toolbox of data-driven approaches, machine learning (ML) dramatically accelerates the computational discoveries. As the machinery for ML algorithms matures, significant advances have been made not only by the mainstream AI researchers, but also those work in computational materials science. The number of ML and artificial neural network (ANN) applications in the computational materials science is growing at an astounding rate. This perspective briefly reviews the state-of-the-art progress in some supervised and unsupervised methods with their respective applications. The characteristics of primary ML and ANN algorithms are first described. Then, the most critical applications of AI in computational materials science such as empirical interatomic potential development, ML-based potential, property predictions, and molecular discoveries using generative adversarial networks (GAN) are comprehensively reviewed. The central ideas underlying these ML applications are discussed, and future directions for integrating ML with computational materials science are given. Finally, a discussion on the applicability and limitations of current ML techniques and the remaining challenges are summarized. This article is categorized under: Computer and Information Science > Chemoinformatics. Structure and Mechanism > Computational Materials Science. Computer and Information Science > Computer Algorithms and Programming. Software > Molecular Modeling
A Combined Cycle Gas Turbine Model for Heat and Power Dispatch Subject to Grid Constraints
No full textThis paper investigates an optimal scheduling method for the operation of combined cycle gas turbines (CCGT). The objective is to minimize the CO2 emissions while supplying both electrical and thermal loads. This paper adopts a detailed model of the units in order to relate the heat and power outputs. The grid constraints as well as system losses are considered for both the electrical and thermal systems. Finally, the optimal power dispatch lies on the hybridization of a mixed integer linear programing scheduling with a greedy search method. Different sets of simulations are run for a small 5-bus test case and a larger model of Jurong Island in Singapore. Several load levels are considered for the heat demand, and the impact of the steam pipe capacities is highlighted