1,720,982 research outputs found
Hybrid Monte Carlo studies of high temperature superconductors
No full textIn this thesis we have developed a Hybrid Monte Carlo simulation of the vortex state in layered high-temperature superconductors. A set of potentials that govern vortex behaviour are derived from the Lawrence-Doniach free-energy functional which incorporate (i) intra-layer coupling (ii) inter-layer Josephson and electromagnetic interactions. We develop an extensive set of system observables that enable detailed studies of the structural properties of the vortex state. Naïve truncation of the long range intra-layer potential is shown to cause incorrect physical behaviour. We present two methods to overcome the problem. The first smoothes the potential and the second performs an in-plane infinite lattice summation for the intra-layer interactions, which provides a minimum 20,000 speed-up over previous methods. We present results of the numerical B-T phase diagram in the pure and pinned system and obtain good agreement with available experimental and theoretical results.Significant hysteresis is observed in the melting properties of the system and we implement the Hybrid Monte Carlo (HMC) method for the first time in such a system to overcome this. The correlation time in the system and the rate of transitions between solid and liquid states are both shown to improve by a factor of 5 over the Monte Carlo (MC) method. We perform HMC simulations on a simple, well-studied model (Ryu, 1996b) and show that the HMC method accurately simulates the system. Finally we investigate the effects of a phenomenological pinning surface upon the melting properties of this system, and demonstrate that the effects of introducing disorder into the system are consistent with experimental and other numerical studies
Learning and Intelligent Optimization. 4th International Conference LION 4
No full textIn the design of complex engineering systems involving multiple disciplines it is critical that the interactions between the subsystems of the problem are accountedfor. Only by considering the fully coupled system can an optimal design emerge. Formal multidisciplinary design optimization (MDO) methods [1] fall into two broad categories; 1) monolithic formulations where a single optimizer addresses the whole problem and 2) multilevel methods where the problem is decomposed along disciplinary lines and optimization takes place at both a system and domain level. The single optimizer approach is simple to implement but can scale poorly for larger problems and increasing number of disciplines. It may also prove problematic in an industrial setting to bring all of the domain analysis tools under the control of a single optimizer. Multilevel architectures promote discipline autonomy. The system level is responsible for managing interactions between disciplines. Such an approach allows design teams to work in relative isolation based upon targets set at the system level. If MDO methods are tobe accepted in an industrial context they must support this form of distributed design optimization for both organizational and computational reasons. In thiswork a related approach is proposed; that of replacing the formal system level optimizer with an expert system to reason over information from the domains and make decisions about changes to the common design variables vector or bounds. Such an approach sacrifices, possibly elusive, guarantees of convergence for potentially attractive returns in the enterprise
Monte Carlo simulation of layered high temperature superconductors
No full textWe present results from Monte Carlo simulations of the vortex state in layered high temperature superconductors. We use a set of potentials derived from the Lawrence-Doniach free-energy functional which incorporates (i) intra-layer coupling (ii) inter-layer Josephson and electromagnetic interactions. We have employed a novel technique for performing an in-plane infinite lattice summation for the intra-layer interactions. This provides a minimum 50,000 times faster speed-up in the simulations over previous naïve methods which add periodic images in shells of increasing radius. We present results of the numerical B-T phase diagram in the pure system and obtain good agreement with available experimental/theoretical results
Earth system model optimisation using response surface modelling in the Grid-enabled GENIE framework
No full textModels frequently contain parameters that are poorly constrained by observations or are a product of simplification during model formulation. In the case of Earth System Models (ESMs), which typically comprise several earth system components more commonly modelled separately, there can be many such parameters. Consequently, with so many degrees of freedom it can be difficult to tune a model to observations in
preparation for experiments.The Grid-enabled GENIE framework provides an interface to a Design Search and Optimisation package which we exploit to adopt a Response Surface Modelling (RSM) approach to tune an ESM comprising a 3D ocean, sea-ice and a simplified 2D atmosphere (GENIE c-GOLDSTEIN). This approach uses a small (order 100) number of computationally-expensive simulations to sample multi-dimensional (order 12) parameter space with respect to a [model - observations] error function. These samples are pooled to generate a response surface (using a Kriging method) to estimate the behaviour of this error function across parameter space. By finding the minimum point on this surface, simulating at this point, and then adding the new simulation to the pool to regenerate the response surface, the error minimum is successively refined.Further work evaluates the performance of this method by attempting to recover model parameters in twin studies. The consequences of the choice of error function (e.g. state variables versus fluxes) are also examined
Transient simulations of the last 22,000 years, with a fully dynamic atmosphere in the GENIE earth-system framework
No full textThis paper presents and discusses an ensemble of transient model simulations from the Last Glacial Maximum to present-day. The model includes a fully dynamic, primitive equation atmosphere (the Reading IGCM), computed vegetation (TRIFFID), and a slab-ocean and seaice. The atmospheric model is more akin to a low-resolution GCM than traditional EMICS, and yet is fast enough for long ensemble simulations to be carried out. The model is tuned in a purely objective manner, using a genetic algorithm, which perturbs 30 tunable parameters in the model to find the best fit to a prescribed pre-industrial climate.The control deglaciation experiment has good agreement with data at the Last glacial Maximum and mid-Holocene. The deglaciation ensembles are over initial conditions, physical processes, and tunable model parameters. The ice-sheets are prescribed, and changes in oceanic heat transport are neglected, and yet the model exhibits rapid transitions in many of the ensemble members. These are attributable to the interaction of the dynamic atmosphere with the sea-ice, and are not observed when the ocean and sea-ice surface temperatures are prescribed. The timing of these transitions is sensitive to the initial conditions, pointing to the chaotic nature of the climate system.The simulations have been carried out making use of GRID technologies, developed as part of the GENIE project
Computational modelling of high temperature superconductors
No full textWe are performing Monte-Carlo and Molecular Dynamics simulations of the vortex state in high temperature superconductors using parallel computers. The model for these highly anisotropic cuprate materials consists of a stack of two-dimensional layers in which point-like vortex pancakes interact with each other. Both in-plane vortex-vortex and multi-layer three/four body interactions are included. By varying the temperature and imposed field we study the fundamental behaviour of the vortex lines in the material. We also investigate the effects of pinning centres, which correspond to oxygen vacancies. Our overall aim is to determine the phase diagram of the superconductor numerically and compare this with experimental results
The role of ocean and atmosphere feedbacks in maintaining bi-stability of the thermohaline circulation (abstract paper presented at 4th EGU General Assembly, April 2007)
No full textWe have used the Grid ENabled Integrated Earth system modelling (GENIE) framework to undertake a systematic search for bi-stability of the thermohaline circulation (THC) with a 3-D dynamical atmosphere model (IGCM) coupled to a 3-D frictional geostrophic ocean model (GOLDSTEIN) and slab sea-ice. Three different ocean grids are used: (i) 36x36x8 longitude-sine (latitude), (ii) 72x72x16 longitude-sine(latitude), (iii) 64x32x8 longitude-latitude. In all cases, the IGCM is run at T21 resolution with 7 vertical levels and surface grid (iii). We contrast this with earlier work using an energy-moisture balance atmosphere model (EMBM) and ocean resolution (i).For each model version, we constructed an ensemble of runs in which we vary atmospheric freshwater transport from the Atlantic to Pacific. The resulting ensembles are run toward equilibrium and then restarts are used to search parameter space for regions of THC bi-stability. A total of 407,000 model years were simulated in 3 months by using UK Grid computing resources, including 6 nodes of the National Grid Service, and additional clusters in Norwich, Southampton and Bristol.We find bi-stability of the THC despite significant, quasi-periodic variability in its strength driven by variability in the dynamical atmosphere. The position and width of the hysteresis loop depends on the choice of surface grid (longitude-latitude or equal area), but is less sensitive to changes in ocean resolution. For the same ocean resolution, the region of bi-stability is broader with the IGCM than with the EMBM.Feedbacks involving both ocean and atmospheric dynamics are found to promote THC bi-stability. THC switch-off leads to increased export of salt at the Southern boundary of the Atlantic that tends to maintain the off state. THC switch-off can also generate net freshwater input to the Atlantic from the atmosphere that tends to maintain the off state. The ocean feedback is present in all resolutions, across most of the bi-stable region, whereas the atmosphere feedback is strongest in resolution (iii) and around the transition where the THC off state is disappearing. Here the ocean response reverses, promoting THC switch-on by reducing Atlantic salt export, but the atmosphere counteracts this by increasing net freshwater input. This appears to maintain some bistability even when the THC does switch on - weaker and stronger THC on states can be distinguished under the same boundary conditions and different initialisations of the model
- …
