Search

The goaf treatment of underground metal mines is an important link in mining, and it is particularly important to master the laws of overlying rock strata and surface movement of goaf. In this paper, Persistent Scatterer Interferometric Synthetic Aperture Radar (PS-InSAR) technology is used to monitor the surface subsidence of the Taibao lead-zinc mine, and the surface subsidence laws of goaf-closure, partial-filling, and full-filling treatments are analyzed by the time-series method. The findings indicate that the surface subsidence of the closed goaf is solely governed by the pillars, with the quality of these pillars playing a pivotal role in controlling such subsidence. Factors like stope span also influence the surface subsidence of partially filled goaf. Prior to compaction, it is primarily the pillars that control surface subsidence; however, after compaction, filling and pillars jointly regulate this phenomenon. Notably, in filled goaf, the quality of both roof and pillars significantly impacts surface subsidence. Before compaction occurs, control over surface subsidence is not evident, yet post-compaction, the filling is effective and tends to stabilize this process. The research findings are significant in enhancing goaf’s treatment efficacy, mitigating surface damage and minimizing ecological environmental impact.

The particle shape has significant effects on the slip and rotation of particles in the shear of geomaterials, which is an important factor in the deformation and strength of geomaterials. This paper employed particle flow code (PFC3D) to simulate the simple shear test, and ellipsoidal particles with different aspect ratios were prepared to study the effects of particle shape on the mechanical behavior and fabric evolution of granular materials under complex stress paths. The numerical results show that the particle shape has a significant effect on the peak strength, dilatancy, non-coaxiality, and other mechanical properties of granular materials. The contact fabric evolves from orthotropy to transverse isotropy under the principal stress axes rotation. This paper will provide a reference for natural granular materials with different shapes in the study of mechanical behavior and the micro-constitutive model.

The stacking of impermissible materials in the disposal of dry fly ash is unprecedented in the last 40 years of power plant management in China, and their effect on the stability of the whole facility is uncertain. Due to the lack of relevant treatment experience, a more comprehensive method such as numerical modeling must be adopted for the final design. This paper set up a borehole database from geological logging data to obtain the distribution of the coal gangue solid waste. Then, it established an accurate three-dimensional mesh model through Rhino. Based on elastic–plastic mechanics, the finite difference code Flac3D 6.0 was employed to study the risk of the coal gangue as a dam foundation. A comparative analysis of the influence of the displacement method and the composite foundation method on subdam deformation and differential subsidence was conducted. The simulation revealed that the composite foundation method showed the best reductions: 70.57% in shear failure, 97.83% in tension failure, and 22.63% in maximum subsidence. Ultimately, the optimum stone column diameter of 0.5 m and the spacing of 6 m were proposed due to the standard deviation.

Considering the influence of fabric on the critical state of sandy soil, an anisotropic hypoplastic constitutive model is developed by introducing the anisotropic critical state line that takes into account the rotation of the principal stress axes. With the introduction of new anisotropic state variables defined by the joint invariants of the stress tensor and fabric tensor, the critical state equation of sandy soil is established to describe the effects of three factors, namely, anisotropic parameters, stress states, and the relationship between principal stresses and fabric directions, on the critical state. The mechanical response of sandy soil under different deposition angles can be described by considering the rotation of principal stresses relative to the fabric. The application range of Wu et al.’s isotropic hypoplastic model (2017) is extended by incorporating the effect of principal stress rotation on the stress–strain relationship of sandy soil. Based on a series of Toyoura sand plane strain tests, the effects of void ratio, confining pressure, and principal stress axis rotation angle on anisotropic strength and deformation characteristics are simulated under low confining pressure. Furthermore, a comparison with Wu’s transversely isotropic hypoplastic model (1998) is made regarding their simulation performances. The proposed model exhibits a balanced performance when simulating the variation of anisotropy in both strength and deformation with respect to the rotation angle, without being overestimated within a certain range of rotation angles. The prediction results demonstrate, to a certain degree, the validity and effectiveness of the proposed model.

As a special road-building material widely distributed in desert areas, critical state soil mechanics is used to study the mechanical properties of sand and make up for the lack of research on its engineering characteristics. A series of drained and undrained triaxial compression tests with a loading rate of 0.12 mm/min medium-density aeolian sands taken from Tengger Desert in the northwest of China was carried out to obtain the three-dimensional state boundary surface. The test results reveal that the strength gained from drained and undrained tests increased, respectively, linearly and non-linearly with the increase of the effective confining pressure. Affected by the variation of pore pressure and shear rate, the undrained strength was higher than the drained strength at low effective confining pressures, and the two types of strengths tend to be consistent when the effective confining pressure becomes greater than 800 kPa. The volumetric changes of the aeolian specimens transition from dilatation to contraction when the effective confining pressures increase. The investigation of the strength, deformation and failure characteristics gives rise to the shape parameters of its state boundary surface, which provides not only a basis for the constitutive modelling of the aeolian sand, but also a reference for roadbed construction and other foundation engineering in desert areas.

A visual dialog task entails an agent engaging in a multiple round conversation about an image. Notably, one of the main issues is capturing the semantic associations of multiple inputs, such as the questions, dialog history, and image features. Many of the techniques use a token or a sentence granularity semantic representation of the question and dialog history to model semantic associations; however, they do not perform collaborative modeling, which limits their efficacy. To overcome this limitation, we propose a multi-granularity semantic collaborative reasoning network to properly support a visual dialog. It employs different granularity semantic representations of the question and dialog history to collaboratively identify the relevant information from multiple inputs based on attention mechanisms. Specifically, the proposed method collaboratively reasons the question-related information from the dialog history based on its granular semantic representations. Then, it collaboratively locates the question-related visual objects in the image by leveraging refined question representations. The experimental results conducted on the VisDial v.1.0 dataset verify the effectiveness of the proposed method, showing the improvements of the best normalized discounted cumulative gain score from 59.37 to 60.98 with a single model, from 60.92 to 62.25 with ensemble models, and from 63.15 to 64.13 with performing multitask learning.

Different from the other roadbed material, the unique mechanical properties of aeolian sand bring great difficulties to the construction and maintenance of desert highways. However, the main attention was usually paid to the engineering properties of aeolian sand, such as collapsibility, strong permeability, and poor gradation. To investigate the shear behaviour of aeolian sand under different engineering conditions, the drained and undrained tests were performed on aeolian sand with relatively large range of density and confining pressure. Under this condition, both the drained and undrained tests tend to the same critical state line, and the shear behaviour of aeolian sand is directly dependent on its density. Under the undrained condition, the q-ε1 curves and the effective stress paths in triaxial stress space exhibit four types of undrained shear behaviour, such as flow, limited flow, strain hardening, and strain softening. Meanwhile all the specimens exhibit three types of failure, such as flow slip, bulging failure, and shear bands. In the q-p’ plane, the analogous drained and undrained stress paths can be followed by aeolian sand with same initial relative density but different confining pressures, and there are two critical state lines due to the generation of shear bands for dense sand. In addition, the critical state lines in e-lnp’ plane decrease with increasing initial relative density Dr, that is, the material constant eᴦ decreases with increasing Dr, and the λ is also not constant but decreases with the increase in Dr. The results suggest that the strength behaviours of aeolian sand can be fitted by a straight line considering relative density and confining pressure and that two empirical formulas are established to describe this feature.

Based on the principle of stereology to describe void fabric, the fabric tensor is redefined by the idea of normalization, and a novel quantitative description method for the orthotropic fabric of granular materials is presented. The scan line is described by two independent angles in the stereo space, and the projection of the scan line on three orthogonal planes is used to determine the plane tensor. The second-order plane tensor can be described equivalently by two invariants, which describe the degree and direction of anisotropy of the material, respectively. In the three-dimensional orthogonal space, there are three measurable amplitude parameters on the three orthogonal planes. Due to the normalized definition of tensor in this paper, there are only two independent variations of the three amplitude parameters, and any two amplitude parameters can be used to derive the three-dimensional orthotropic fabric tensor. Therefore, the same orthorhombic anisotropy structure can be described by three fabrics, which enriches the theoretical description of orthotropy greatly. As the geometric relationship of the stereoscopic space scan line changes, the three sets of orthotropic fabrics degenerate into different forms of transversely isotropic and isotropic fabrics naturally and have a clear physical meaning. The novel fabric tensor is quantitatively determined based on mathematical probability and statistics. The discrete distribution of voids in space is projected as a scalar measurable parameter on a plane. This parameter is related to the macroscopic constitutive relationship directly and can be used to describe the effect of microscopic voids on the macroscopic phenomenon of materials.

With the rapid development of social networks, spam bots and other anomaly accounts’ malicious behavior has become a critical information security problem threatening the social network platform. In order to reduce this threat, the existing research mainly uses feature-based detection or propagation-based detection, and it applies machine learning or graph mining algorithms to identify anomaly accounts in social networks. However, with the development of technology, spam bots are becoming more advanced, and identifying bots is still an open challenge. This paper proposes a new semi-supervised graph embedding model based on a graph attention network for spam bot detection in social networks. This approach constructs a detection model by aggregating features and neighbor relationships, and learns a complex method to integrate the different neighborhood relationships between nodes to operate the directed social graph. The new model can identify spam bots by capturing user features and two different relationships among users in social networks. We compare our method with other methods on real-world social network datasets, and the experimental results show that our proposed model achieves a significant and consistent improvement.

In recent years, privacy leakage events in large-scale social networks have become increasingly frequent. Traditional methods relying on operators have been unable to effectively curb this problem. Researchers must turn their attention to the privacy protection of users themselves. Privacy metrics are undoubtedly the most effective method. However, social networks have a substantial number of users and a complex network structure and feature set. Previous studies either considered a single aspect or measured multiple aspects separately and then artificially integrated them. The measurement procedures are complex and cannot effectively be integrated. To solve the above problems, we first propose using a deep neural network to measure the privacy status of social network users. Through a graph convolution network, we can easily and efficiently combine the user features and graph structure, determine the hidden relationships between these features, and obtain more accurate privacy scores. Given the restriction of the deep learning framework, which requires a large number of labelled samples, we incorporate a few-shot learning method, which greatly reduces the dependence on labelled data and human intervention. Our method is applicable to online social networks, such as Sina Weibo, Twitter, and Facebook, that can extract profile information, graph structure information of users’ friends, and behavioural characteristics. The experiments show that our model can quickly and accurately obtain privacy scores in a whole network and eliminate traditional tedious numerical calculations and human intervention.

The popularity of social networks provides people with many conveniences, but their rapid growth has also attracted many attackers. In recent years, the malicious behavior of social network spammers has seriously threatened the information security of ordinary users. To reduce this threat, many researchers have mined the behavior characteristics of spammers and have obtained good results by applying machine learning algorithms to identify spammers in social networks. However, most of these studies overlook class imbalance situations that exist in real world data. In this paper, we propose a heterogeneous stacking-based ensemble learning framework to ameliorate the impact of class imbalance on spam detection in social networks. The proposed framework consists of two main components, a base module and a combining module. In the base module, we adopt six different base classifiers and utilize this classifier diversity to construct new ensemble input members. In the combination module, we introduce cost sensitive learning into deep neural network training. By setting different costs for misclassification and dynamically adjusting the weights of the prediction results of the base classifiers, we can integrate the input members and aggregate the classification results. The experimental results show that our framework effectively improves the spam detection rate on imbalanced datasets.

Recently, the number of people who are members of multiple online social networks simultaneously has increased. However, if these people share everything with others, they risk their privacy. Users may be unaware of the privacy risks involved with sharing their sensitive information on a network. Currently, there are many research efforts focused on social identity linkage (SIL) on multiple online social networks for commercial services, which exacerbates privacy issues. Many existing studies consider methods of encrypting or deleting sensitive information without considering if this is unreasonable for social networks. Meanwhile, these studies ignore privacy awareness, which is rudimentary and critical. To enhance privacy awareness, we discuss a user privacy exposure measure for users who are members of multiple online social networks. With this measure, users can be aware of the state of their privacy and their position on a privacy measurement scale. Additionally, we propose a straightforward method through our framework to reduce information loss and foster user privacy awareness by using spurious content for required fields.

This paper mainly studies the colorization of near-infrared (NIR) images. Image colorization methods cannot be extended to NIR image colorization since the wavelength band of the NIR image exceeds the visible light spectral range and it is often linearly independent of the luminance of the RGB image. Furthermore, a symmetric codec, which cannot guarantee the ability of the encoder to extract features, is often used as the main frame of the network in both CNN-based colorization networks and CycleGAN-based colorization networks. In order to deal with the investigated problem, we propose a novel NIR colorization method using asymmetric codec (ACD) and pixel-level fusion. ACD is designed to improve the feature extraction ability of the encoder by allowing the information to enter deeper into the model and learning more non-redundant information. In addition, the global and local feature fusion networks (GLFFNet) are embedded between the encoder and the decoder to improve the prediction of the subtle color information of the image. The ACD and GLFFNet together constitute the colorization network (ColorNet) in this paper. Bilateral filtering and weighted least squares filtering (BFWLS) are used to fuse the pixel-level information of the input NIR image into the raw output image of the ColorNet. Finally, an intensive comparison analysis based on common datasets is conducted to verify superiority over existing methods in qualitative and quantitative visual assessments.

Materials such as silicon, copper, gold, and aluminum exhibit strong absorption and scattering characterization under short-pulsed laser irradiation. Due to the photoelastic effect and thermoelastic relaxation, the focal area may induce a local modulation in the refractive index, which can be detected with the intensity reflection coefficient perturbation. Normally, the thermal effect causes a weak refractive index change and is negligible, compared with the pressure-induced effect in most photoacoustic analytical systems. In this study, we present a theoretical model with the whole process of absorbed energy conversion analysis for the refractive index perturbation induced by both thermal effect and photoacoustic pressure. In this model, data analysis was carried out on the transformation of the energy absorbed by the sample into heat and stress. To prove the feasibility of this model, numerical simulation was performed for the photothermal and photoacoustic effects under different incident intensities using the finite element method. Experiment results on silicon and carbon fiber verified that the refractive index change induced by the photothermal effect can be detected and be incorporated with pressure-induced refractive index change. The simulation results showed very good agreement with the results of the experiments. The main aim of this study was to further understand the absorption and conversion process of short-pulsed light energy and the resulting photothermal and photoacoustic effects.

In photoacoustic imaging, the use of arrayed laser sources brings several advantages. Acoustic waves can be generated with flexible control of wavefronts, bringing functionality such as ultrasonic beam steering and focusing. The use of arrays reduces the optical intensity while increasing the strength of the ultrasonic wave, bringing the advantages of improved signal-to-noise ratio (SNR) while avoiding laser-induced damage. In this paper, we report a numerical model for studying the generation and shaping of acoustic wavefronts with laser arrays. The propagation of mechanical waves, photoacoustically generated by thermal expansion, is simulated and discussed in detail. In addition, a partially delayed distributed array is studied both theoretically and quantitatively. The developed model for wavefront control through time-delayed laser pulses is shown to be highly suited for the optimization of laser array generation schemes.