3D model reconstruction using neural gas accelerated on GPU

In this work, we propose the use of the neural gas (NG), a neural network that uses an unsupervised Competitive Hebbian Learning (CHL) rule, to develop a reverse engineering process. This is a simple and accurate method to reconstruct objects from point clouds obtained from multiple overlapping views using low-cost sensors. In contrast to other methods that may need several stages that include downsampling, noise filtering and many other tasks, the NG automatically obtains the 3D model of the scanned objects. To demonstrate the validity of our proposal we tested our method with several models and performed a study of the neural network parameterization computing the quality of representation and also comparing results with other neural methods like growing neural gas and Kohonen maps or classical methods like Voxel Grid. We also reconstructed models acquired by low cost sensors that can be used in virtual and augmented reality environments for redesign or manipulation purposes. Since the NG algorithm has a strong computational cost we propose its acceleration. We have redesigned and implemented the NG learning algorithm to fit it onto Graphics Processing Units using CUDA. A speed-up of 180× faster is obtained compared to the sequential CPU version.

Implementation of non-local boundary layer schemes in the Regional Atmospheric Modeling System and its impact on simulated mesoscale circulations

This paper proposes the implementation of different non-local Planetary Boundary Layer schemes within the Regional Atmospheric Modeling System (RAMS) model. The two selected PBL parameterizations are the Medium-Range Forecast (MRF) PBL and its updated version, known as the Yonsei University (YSU) PBL. YSU is a first-order scheme that uses non-local eddy diffusivity coefficients to compute turbulent fluxes. It is based on the MRF, and improves it with an explicit treatment of the entrainment. With the aim of evaluating the RAMS results for these PBL parameterizations, a series of numerical simulations have been performed and contrasted with the results obtained using the Mellor and Yamada (MY) scheme, also widely used, and the standard PBL scheme in the RAMS model. The numerical study carried out here is focused on mesoscale circulation events during the summer, as these meteorological situations dominate this season of the year in the Western Mediterranean coast. In addition, the sensitivity of these PBL parameterizations to the initial soil moisture content is also evaluated. The results show a warmer and moister PBL for the YSU scheme compared to both MRF and MY. The model presents as well a tendency to overestimate the observed temperature and to underestimate the observed humidity, considering all PBL schemes and a low initial soil moisture content. In addition, the bias between the model and the observations is significantly reduced moistening the initial soil moisture of the corresponding run. Thus, varying this parameter has a positive effect and improves the simulated results in relation to the observations. However, there is still a significant overestimation of the wind speed over flatter terrain, independently of the PBL scheme and the initial soil moisture used, even though a different degree of accuracy is reproduced by RAMS taking into account the different sensitivity tests.

Thermodynamic calculations for the salt crystallisation damage in porous built heritage using PHREEQC

This work considers the crystallisation mechanisms of the most common and aggressive salts that generate stress in porous building stones as a result of changing ambient conditions. These mechanisms include the salt crystallisation that result from decreasing relative humidity and changes in temperature and, in hydrated salts, the dissolution of the lower hydrated form and the subsequent precipitation of the hydrated salt. We propose a new methodology for thermodynamic calculations using PHREEQC that includes these crystallisation mechanisms. This approach permits the calculation of the equilibrium relative humidity and the parameterization of the critical relative humidity and crystallisation pressures for the dissolution–precipitation transitions. The influence of other salts on the effectives of salt crystallisation and chemical weathering is also assessed. We review the sodium and magnesium sulphate and sodium chloride systems, in both single and multicomponent solutions, and they are compared to the sodium carbonate and calcium carbonate systems. The variation of crystallisation pressure, the formation of new minerals and the chemical dissolution by the presence of other salts is also evaluated. Results for hydrated salt systems show that high crystallisation pressures are possible as lower hydrated salts dissolve and more hydrated salts precipitate. High stresses may be also produced by decreasing temperature, although it requires that porous materials are wet for long periods of time. The presence of other salts changes the temperature and relative humidity of salt transitions that generates stress rather than reducing the pressure of crystallisation, if any salt has previously precipitated. Several practical conclusions derive from proposed methodology and provide conservators and architects with information on the potential weathering activity of soluble salts. Furthermore, the model calculations might be coupled with projections of future climate to give as improved understanding of the likely changes in the frequency of phase transitions in salts within porous stone.