An Adaptive, Finite Difference Solver for the Nonlinear Poisson-Boltzmann Equation with Applications to Biomolecular Computations - Volume 13 Issue 1 - Mohammad Mirzadeh, Maxime Theillard, Asdís Helgadöttir, David Boy, Frédéric Gibou

A comprehensive catalogue of historical earthquakes, with accurate epicentres and harmonised magnitudes is a crucial resource for seismic hazard mapping. Here we update and combine catalogues from several sources to compile a catalogue of earthquakes in and near Iceland, in the years 1900–2019. In particular the epicentres are based on local information, whereas the magnitudes are based on teleseismic observations, primarily from international online catalogues. The most reliable epicentre information comes from the catalogue of the Icelandic Meteorological Office, but this is complemented with information from several technical reports, scientific publications, and newspaper articles. The catalogue contains 1281 moment magnitude (M_w) ≥4 events, and the estimated completeness magnitude is M_w 5.5 in the first years, going down to M_w 4.5 for recent years. The largest magnitude is M_w 7.0. Such merging of local data and teleseismic catalogues has not been done before for Icelandic earthquakes, and the result is an earthquake map with much more accurate locations than earlier maps. The catalogue also lists 5640 additional earthquakes on the Mid-Atlantic Ridge, north of 43^∘, with both epicentres and magnitudes determined teleseismically. When moment magnitudes are not available, proxy M_w values are computed using χ² regression, normally on the surface-wave magnitude but exceptionally on the body-wave magnitude. Magnitudes of M_w≥4.5 have associated uncertainty estimates. The actual combined seismic moment released in the Icelandic earthquakes is found to be consistent with the moment estimated using a simple plate motion model, indicating that the seismic activity of the catalogue period might be typical of any 120-year time span. The catalogue is named ICEL-NMAR, and it is available online at http://data.mendeley.com (last access: 19 July 2021).

Numerical simulations were performed with the open-source CFD software OpenFOAM to investigate the ability of several configurations of short-length twisted tube geometries with non-circular cross section connected to tubes with circular cross section to induce a swirling flow. The heat transfer and the pressure drop linked to the generated swirling flow are also calculated. The swirling flow is modeled using a k-ω SST turbulence model with a low-Reynolds approach. It is shown that a short-length twisted tube with an elliptical cross section (STE) is able to generate a swirling flow, but its intensity greatly depends on its twist pitch and its aspect ratio. The lower the aspect ratio, the higher the swirl intensity. For a Reynolds number ranging from 10,000 to 100,000, the results reveal that compared to a plain tube, the STE with the lowest aspect ratio achieves enhancing the heat transfer from 22 to 90% at the cost of an increased pressure drop of, respectively, 63 and 129%. The second part of the study is focused on a short-length twisted tube with a three-lobed cross section, and the results reveal that the generated swirling flow is even more intense than with the STE and that the heat transfer enhancement goes from 30 to 105% at the cost of an increased pressure drop from 137 to 180%.

Turbulent flow is a complex and vital phenomenon in fluid dynamics, as it is the most common type of flow in both natural and artificial systems. Traditional methods of studying turbulent flow, such as computational fluid dynamics and experiments, have limitations such as high computational costs, experiment costs, and restricted problem scales and sizes. Recently, artificial intelligence has provided a new avenue for examining turbulent flow, which can help improve our understanding of its flow features and physics in various applications. Strained turbulent flow, which occurs in the presence of gravity in situations such as combustion chambers and shear flow, is one such case. This study proposes a novel data-driven transformer model to predict the velocity field of turbulent flow, building on the success of this deep sequential learning technique in areas such as language translation and music. The present study applied this model to experimental work by Hassanian et al., who studied distorted turbulent flow with a specific range of Taylor microscale Reynolds numbers 100<Reλ<120. The flow underwent a vertical mean strain rate of 8 s−1 in the presence of gravity. The Lagrangian particle tracking technique recorded every tracer particle's velocity field and displacement. Using this dataset, the transformer model was trained with different ratios of data and used to predict the velocity of the following period. The model's predictions significantly matched the experimental test data, with a mean absolute error of 0.002–0.003 and an R2 score of 0.98. Furthermore, the model demonstrated its ability to maintain high predictive performance with less training data, showcasing its potential to predict future turbulent flow velocity with fewer computational resources. To assess the model, it has been compared to the long short-term memory and gated recurrent units model. High-performance computing machines, such as JUWELS-DevelBOOSTER at the Juelich Supercomputing Center, were used to train and run the model for inference.