Search

The interpolation, prediction, and feature analysis of fine-gained air quality are three important topics in the area of urban air computing. The solutions to these topics can provide extremely useful information to support air pollution control, and consequently generate great societal and technical impacts. Most of the existing work solves the three problems separately by different models. In this paper, we propose a general and effective approach to solve the three problems in one model called the Deep Air Learning (DAL). The main idea of DAL lies in embedding feature selection and semi-supervised learning in different layers of the deep learning network. The proposed approach utilizes the information pertaining to the unlabeled spatio-temporal data to improve the performance of the interpolation and the prediction, and performs feature selection and association analysis to reveal the main relevant features to the variation of the air quality. We evaluate our approach with extensive experiments based on real data sources obtained in Beijing, China. Experiments show that DAL is superior to the peer models from the recent literature when solving the topics of interpolation, prediction, and feature analysis of fine-gained air quality. [ABSTRACT FROM AUTHOR]

• A meso-scale model was developed for glass mortar with three phases considered. • Two most important physical characteristics of glass aggregate were reflected. • Neglecting ITZ leads to overestimation of compressive strength and elastic modulus. • Aggregate shape was found to have minor impact on the numerical results. • Influence of various aggregate properties was investigated by sensitivity analysis. Waste glass has become one of the most attractive candidates for replacing natural aggregate in mortar/concrete. Although researchers have conducted extensive experiments to investigate the macroscopic behaviour of mortar/concrete containing glass aggregate, certain mechanical behaviours are difficult to capture. These include damage initiation and propagation, and the effects of the intrinsic characteristics of glass aggregate, which are important in building applications. In this study, a two-dimensional meso-scale model was developed to numerically investigate the macroscopic and mesoscopic behaviour of glass mortar under uniaxial compression loading. Glass mortar was modelled as a three-phase composite including the surrounding normal mortar, glass aggregate and interfacial transition zone. The developed model was validated using experimental observations of the behaviour of glass mortar under uniaxial compression loading. Two prominent physical characteristics of glass aggregate were reflected in the developed meso-scale model, namely the angular aggregate shape and weaker interfacial transition zone. Based on the simulation results, it was found that both characteristics affected the mechanical response, while the aggregate shape was observed to be less significant. A comprehensive sensitivity analysis was performed to investigate the effects of various aggregate's properties, including random distribution, volume fraction, particle size, and elongation ratio, on the mechanical response of mortar. Overall, the developed meso-scale model for glass mortar is expected to provide a better understanding of the failure mechanisms of glass mortar/concrete under uniaxial compression loading. [ABSTRACT FROM AUTHOR]