Search

Geospatial cellular automata (Geo-CA) models have encountered challenges in computing efficiency and scalability when conducting large-scale land use change simulation applications. Parallel computing has proven to be effective to address these challenges. However, parallelization strategies for existing parallel Geo-CA models are always optimized for specific models and lack applicability to others. Besides, most parallel Geo-CA models focus on parallelizing land use change simulations, ignoring land use transition rule mining. Moreover, there is a lack of effective parallel strategies for demand-constrained land use change simulation on distributed heterogeneous architectures. This study proposes a parallel framework on hybrid parallel computing architectures applicable for raster-based Geo-CA models to enhance their computing efficiency and scalability while maintaining simulation accuracy. The framework provides parallelization strategies for both the land use transition rule mining for multiple land use types and the demand-constrained land use change simulation on distributed heterogeneous architectures. The framework was employed to parallelize two contemporary Geo-CA models, i.e., PLUS and MCCA. Experiments showed that the parallelized models achieved significant improvements in computing efficiency and scalability, confirming the effectiveness of the proposed framework for large-scale land use change simulation studies. [ABSTRACT FROM AUTHOR]

Oil-immersed power transformers generate characteristic gases after failure,Therefore, it is necessary to analyse and monitor the soluble gases in transformer oil samples. In this study, the DFT calculation method was used to study the adsorption properties of H2, CO, C2H2, and C2H4 gases in oil on both intrinsic MoTe2 and Rh-doped MoTe2 films. In order to analyse the adsorption characteristics, this paper first obtains the most stable Rh-doped MoTe2 monolayer model through the modelling and computational analysis of different doping sites, then, the adsorption of these gases on the material surface is studied by analysing adsorption energy, charge transfer, total state density, parting density, energy band structure, differential charge density map, molecular front orbital and desorption time, and finally concludes that Rh-MoTe2 monolayer film is the ideal material for hydrogen sensing elements, due to the extremely long desorption time of CO gas, indicating that CO gas is hard to desorption on the surface of adsorbent,which shows that this process has played a certain role in promoting CO removal. Highlights Study focused on gas adsorption using DFT: Rh-MoTe2 ideal for hydrogen sensing and effective for CO removal, intrinsic MoTe2 ineffective and required metal doping. Rh doping enhanced conductivity and gas adsorption capacity in MoTe2. Rh-MoTe2 was suitable for detecting H2, C2H2, and C2H4 gases. Desorption times vary with temperature, with CO removal particularly effective at higher temperatures. Findings suggest potential applications for Rh-MoTe2 in gas sensing and removal. [ABSTRACT FROM AUTHOR]

Inspired by its potential applications of organic luminescence and fluorescence probe materials for 2-(2′-hydroxyphenyl)benzazoles (HBX) derivatives, this study mainly investigates the excited-state behaviours of a novel 2-benzooxazol-2-yl-6-thiophen-2-yl-phenol (BYTYP) fluorophore. Theoretical exploration has been conducted on the solvent-dependent interactions of excited-state intramolecular hydrogen bonding and the process of excited-state intramolecular proton transfer (ESIPT) for BYTYP. By combining optimised geometrical modifications, infrared (IR) vibrational spectra and the core-valence bifurcation (CVB) index, hydrogen bonding strengthening can be confirmed. Predicting the bond energy (EHB), we assert that nonpolar solvents are more conducive to enhancing hydrogen bonding interactions. The ESIPT tendency of BYTYP is further elucidated by charge reorganisation resulting from photoexcitation. By exploring potential energy surfaces (PESs) and identifying transition states (TS), we have uncovered the solvent-polarity-controlled ESIPT behaviours. We hope these insights into excited-state dynamics will facilitate the design and development of novel fluorescent sensors in future. [ABSTRACT FROM AUTHOR]