Search

Hard petroleum asphalt, which is a high modulus asphalt binder material, is used for improving asphalt pavement rutting distress. This property is attributed to the good high-temperature rheological characteristics of hard petroleum asphalt. The study investigates the performance of hard petroleum asphalt and Styrene-Butadiene-Styrene polymer modified asphalt (SBSPMA) based on three aspects: asphalt oil source, asphalt grade, and aging degree. The short-term and long-term aging of four hard petroleum asphalts and SBSPMA of comparative samples were measured using the rolling thin film oven test (RTFOT) and the pressure aging test (PAV). Firstly, the viscosity-temperature curve of hard petroleum asphalt was obtained using the Brookfield viscosity test at different temperatures combined with the improved REFUTAS formula. The complex shear modulus (G∗), phase angle (δ), and rutting factor (G∗/sin δ) of each asphalt sample were then measured by the dynamic shear rheometer test (DSR). Furthermore, the temperature sensitivity of each asphalt sample was calculated using the complex modulus index method (GTS). The compaction and paving temperature ranges of each asphalt were then confirmed. The results showed that the high-temperature rheological characteristics of hard petroleum asphalt were similar to those of SBSPMA. The temperature sensitivity of hard petroleum asphalt was basically the same as that of SBSPMA, while the influence of aging on the G∗ of hard petroleum asphalt was greater than that of SBSPMA. In addition, the hard petroleum asphalt construction temperature was lower than that of SBSPMA.

To make steel slag being reasonably used in geotechnical backfilling projects or soft foundation treatment projects, three kinds of steel slag such as fine, coarse, and gravel steel slag were studied through particle analysis tests, relative density tests, and specific gravity tests to obtain basic physical parameters. Considering the influence of relative density, gradation, and other factors, constant head permeability tests of pure steel slag and variable head permeability tests of modified silt soil with different mixing contents of steel slag were carried out to test permeability coefficients under various working conditions. Prediction formulas on the permeability coefficients of the three kinds of pure steel slag and steel slag-treated silt soil were, respectively, deduced. It was concluded that the permeability coefficient of pure steel slag was greatly influenced by particle size and relative density, similar to the case of permeability coefficients of fly ash and fine sand in their dense states, and the larger the relative density was, the smaller the permeability coefficient was. The permeability coefficient of steel slag-treated silt soil increased with increasing of mixing content of steel slag, showing that steel slag can obviously improve the permeability performance of silt soil. Research results provide reference for design and construction on the application of steel slag in roadbed backfill, steel slag modifying silt soil, and other projects.

Waste steel slag is a recycled industrial solid waste and is sometimes used as backfills, but the dynamic shear properties of waste steel slag have not been researched at present. To understand the dynamic shear properties of waste steel slag, the resonance column tests were carried out to investigate the dynamic shear modulus and damping ratio of the steel slag considering the effect of the relative density of steel slag and confining pressure. The maximum dynamic shear modulus was discussed, and the dynamic model of the steel slag was constructed by the Hardin–Drnevich model. The regression analysis of the normalized dynamic shear modulus ratio and damping ratio of the steel slag showed that the dynamic model was very fit for the steel slag. The comparisons of the dynamic shear properties of the steel slag with those of Nanjing fine sand and Fujian standard sand showed that the steel slag had similar dynamic shear resistance to Fujian standard sand and had potential to become a substitute for sand in the practical geotechnical engineering in the future.

It is difficult to directly measure the dynamic elastic modulus of soils when the strain is less than 10−4, though the dynamic shear modulus can be easily obtained by the free-vibration testing method. Some improvements were adapted to the traditional free-vibration column for measuring dynamic elastic modulus of soil precisely with lateral vibration in small strain (10−4∼10−6). Differing from the traditional lateral vibration theory, a new dynamic elastic modulus testing method based on bending vibration equation of cantilever beam was put forward based on the improved free-vibration column of GZZ-50. Firstly, some descriptions about calibration process of needed testing parameter (moment of inertia of drive board) were made, and then, Poisson’s ratio of standardized column was used to confirm the measured results. Four main improvements were applied, including the shape of drive board which made the bending vibration equation solved, laser displacement sensors for correcting specimen’s height parameter with uncontacted manner which can reduce the influence of specimen, a hoop with plug used to make the split mould be removed, and air duct used when installing drive board to increase the success rate of installing specimen. Finally, standard process of installing specimen was described, which can be used as reference of similar test.

Different kernels cause various class discriminations owing to their different geometrical structures of the data in the feature space. In this paper, a method of kernel optimization by maximizing a measure of class separability in the empirical feature space with sparse representation-based classifier (SRC) is proposed to solve the problem of automatically choosing kernel functions and their parameters in kernel learning. The proposed method first adopts a so-called data-dependent kernel to generate an efficient kernel optimization algorithm. Then, a constrained optimization function using general gradient descent method is created to find combination coefficients varied with the input data. After that, optimized kernel PCA (KOPCA) is obtained via combination coefficients to extract features. Finally, the sparse representation-based classifier is used to perform pattern classification task. Experimental results on MSTAR SAR images show the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]