Search

AbstractThe advent of non-pillar mining technology of self-formed roadway based on roof cutting theory (NPMTSFRRCT) has revolutionized the method of tackling the difficulties posed by hard suspended roofs in mining engineering. The design of roof cutting parameters plays a crucial role in determining the roof fracture characteristics. The principles of roof cutting parameter design have been analysed, with a focus on the key and challenging aspect of the collapse of single-layer thick hard rock under the condition of a thin immediate roof. A mechanical model of roof fracture has been established, and the impact of immediate roof thickness, roof cutting height, roof cutting angle, and main roof thickness on roof fracture has been analysed. A numerical model based on the UDEC software has been created, and the roof fracture, stress, and displacement variation characteristics have been studied using the failure criterion of polygonal blocks. The results of the theoretical analysis have been verified, and it was found that the roof fracture in partial roof cutting occurs in the form of hinge bite, while complete roof cutting results in step sinking. Engineering practice has shown that the deformation of the roadway surrounding rock has been effectively controlled.

The construction industry is a major energy consumer and carbon emitter, and identifying the key drivers for its green transition has attracted increasing attention. Although government subsidies are one of the most effective and direct ways to induce a green transition, few academics have examined their effects at a micro level. Therefore, this study used the Chinese construction industry as an example to study the influence of subsidies on its green transition. Given the ambiguity of the green transition concept, this study employed the number of green patents and Environmental, Social, and Governance (ESG) ratings to represent the narrow sense and the wide sense of green transition, respectively. According to the empirical findings, subsidies can successfully induce green technology innovation and thus facilitate a green transition. The results of heterogeneity analysis show that government subsidies have a significant incentive-based effect solely on state-owned firms, but an insufficient effect on private and other enterprises. Furthermore, while government subsidies have little effect on ESG ratings, they can promote green transition of enterprises by increasing ESG ratings. The government should increase the types of subsidy packages available to enterprises, while attaching more importance to social responsibility.

Pot and field trials were conducted to explore the combined effect of biochar (BC) with topdressing silicon (Si) on Cd uptake by rice and grain yield in Cd-contaminated paddy soil. The treatments, including BC applied before transplanting (TBC), topdressing Si applied in the soil at the jointing stage (JSi) and BC combined with topdressing Si (TBC + JSi), were designed in a complete random block, and treatment without application of BC and Si was used as a control (CK). Results showed that Cd concentration in milled rice treated with TBC + JSi was decreased by 34.62%, 22.73% and 10.53%, respectively, when compared to CK, TBC and JSi, with the concentration being only 0.17 mg·kg−1. At rice maturity, available Cd in the soil was reduced by 7.98% (TBC), 4.76% (JSi) and 6.02% (TBC + JSi) when compared with CK, while the concentrations of total Cd were 32.07% (TBC), 27.85% (JSi) and 35.44% (TBC + JSi) higher than CK. Moreover, BC and Si increased the Cd sequestrated by leaves markedly, especially for TBC + JSi, which was much higher than TBC and JSi. Therefore, the transfer of Cd from leaf to milled rice was greatly decreased by TBC + JSi. In addition, a synergy effect of TBC + JSi on rice yield was also found. Compared with CK, the grain yields of TBC, JSi and TBC+ JSi were increased by 8.35%, 8.20% and 18.74%, respectively. Nutrient contents in soil and rice plants were also elevated by the application of BC and Si to a certain extent; for example, the contents of nitrogen (N), phosphorus (P), potassium (K) and Si in soil treated with TBC + JSi were raised by 8.96–60.03% when compared with CK. Overall, the combined application of BC with topdressing Si not only increases soil nutrients significantly, promotes their uptake by rice and boosts grain yield, but also effectively inhibits Cd transfer and reduces its accumulation in rice, which ultimately guarantees milled rice security. These results also imply that the combined application of biochar with topdressing silicon might be considered as an effective agronomic measure to decrease the milled-rice Cd in Cd-contaminated paddy soil, which would guarantee food security.

Quantitative study on the elastic and plastic thermomechanical coupling effect of solid materials has important theoretical significance for monitoring the deformation state and stability evaluation of engineering components. In this study, the sample is made of polymethyl methacrylate (PMMA), and uniaxial compression experiments are carried out. The sample experiences the elastic state and then the plastic state until the failure of the sample. The thermoelastic effect and thermoplastic effect in the deformation process of PMMA materials are monitored by infrared thermography and digital speckle system (DIC). That is, the relationship between the reversible elastic temperature change ΔTe and average stress change Δσm, and the relationship between the irreversible plastic temperature rise ΔTp and the dissipated energy ΔUp during the failure deformation process. The results show that when the sample is in the elastic regime, there is only elastic temperature change ΔTe inside the sample. When the sample enters the plastic deformation regime, the total temperature change ΔT includes elastic temperature change ΔTe and plastic temperature change ΔTp, and the temperature change has an obvious memory effect on the previous maximum load. When the previous maximum load exceeded, ΔTp increases rapidly. When the load P is removed, ΔTe disappears, and ΔTp remains. Combined with the theory of thermodynamics and solid mechanics, starting from the 3D stress state, the approximate linear incremental relationship between Te and σm, and the accurate exponential total quantity relationship are derived. The incremental relationship and the full relationship combined with thermodynamic theory and the energy conservation principle, the relationship between the overall plastic temperature rise ΔTp and local plastic temperature rise ΔTpmax of the sample and the energy dissipation value ΔUp experienced by the sample and the local dissipation specific energy Δup is obtained. The theoretical calculation results of ΔTe and ΔTp are in good agreement with the experimental results. The research results can provide a new method for monitoring and identifying the stress and deformation state of engineering components through the temperature field.

One of the best options for managing energy markets and power system performance optimization is precise Electricity Price Forecasting (EPF). In detecting the prices of energy, some factors are of stochastic behavior which has made an unwieldy task. In this study, carrying out Tensor Canonical Correlation Analysis (TCCA) is the first activity to choose the operative parameters in EPF issues and removing redundant factors with low correlation ranks. The second activity is making use of a Deep Neural Network (DNN) with a novel Stacked Pruning Sparse Denoising Autoencoder (SPSDAE) to decrease the noise of data sets with different sources individually. The third activity is that a novel method which is called the Maximum Separation Subspace (MASES) in Sufficient Dimension Reduction (SDR) with Categorical Response. This method is used to detect the outstanding properties of the input data and removing the uncertainty characteristics. The new multi-modal combined (MMC) method is the final step that needs to be taken to predict the day-ahead (one day in advance) cost of electricity. The proposed method is investigated based on the data that were collected from Australia The combination of various predicting models and suggested methods might develop the function of predicting possibility by supplying divergent types of compactness performance. The combined model comprised of three predicting possibilities called beta circulation fitting, scattered Bayesian learning, core compactness prediction. Predicting data respectively are used as a yardstick to compare the efficiency of forecasted results and consequently assess the proposed method. Results of this study confirmed a decrease of effectual error criterion and accuracy enhancement in EPF field as the advantageous effects of this proposed SDR structure.

Extractive Question Answering, also known as machine reading comprehension, can be used to evaluate how well a computer comprehends human language. It is a valuable topic with many applications, such as in chatbots and personal assistants. End-to-end neural-network-based models have achieved remarkable performance on these tasks. The most frequently used approach to extract answers with neural networks is to predict the answer’s start and end positions in the document, independently or jointly. In this paper, we propose another approach that considers all words in an answer jointly. We introduce an encoder-decoder model to learn from all words in the answer. This differs from previous works. which usually focused on the start and end and ignored the words in the middle. To help the encoder-decoder model to perform this task better, we employ evaluation-based reinforcement learning with different reward functions. The results of an experiment on the SQuAD dataset show that the proposed method can outperform the baseline in terms of F1 scores, offering another potential approach to solve the extractive QA task.

In order to mater the reasonable layout basis of the terminal mining line position in the repeated mining of close-distance thick coal seams, taking Yan mine as the engineering background, we conducted theoretical analyses, numerical simulations, and field measurements to study the action mechanism of the stress arch. The results show that (1) with repeated mining, the shape of the left and right half arches of the stress arch changes in the order of time and space until it is stable; (2) the shape of the stress arch is highly related to the distribution of abutment pressure in the working face. If the shape is unchanged, the distribution of abutment pressure is unchanged; (3) in the final mining stage of repeated mining, when the shape of the right half arch is stable, the difference stagger distance of the terminal mining line has little effect on the distribution of abutment pressure of the working face where the front arch foot is located; (4) when the internal stagger distance between 3216 working face and terminal mining line of 4216 working face is greater than 22 m or the external stagger distance is greater than 30 m, 3216 working face is located in a relatively safe position. This study clarifies the key factors for the layout of terminal mining lines in close-distance thick coal seams, which can provide a scientific basis for similar projects.

Under the condition of fully mechanized top coal caving in close-distance coal seams, the surrounding rock of the stopping space easily loses stability during the withdrawal of mining equipment in the working face because the lower coal seam working face is located under the goaf and the overburden rock has a large range of complex interaction. Field investigation, theoretical analysis, laboratory experiment, similar simulation experiment, numerical simulation, and field industrial tests are used to carry out the research on the stability and control of the surrounding rock in the large section stopping space under the goaf in this paper. The research conclusions are as follows. (1) It is determined that the lower coal seam working face can only stop mining under the goaf, and the reasonable stopping position under the goaf should ensure that the key block fracture line of the main roof is behind the support. (2) The interaction law between the main roof’s key blocks of the upper and lower coal seams is analyzed, and the catastrophic conditions for sliding instability and rotary instability of the main roof’s key blocks of the upper and lower coal seams are obtained. (3) “Anchorage with push and pull equipment-Embedded anchorages and trays” integral anchoring technology is developed. The dimensions of the push and pull equipment are determined. (4) Through numerical simulation of the distribution characteristics of the anchor cable pre-stress field, the asymmetric control scheme of “Partition long and short anchor cables + Integral polyurethane mesh + Embedded anchorages and trays for roof protection” is determined. The rock pressure observation shows that the withdrawal of the working face equipment is implemented safely.

Logistics is playing an important role in China with the rapid growth of the digital economy, and has caused large quantities of carbon emissions as an energy-intensive industry. Due to the extreme imbalance of land urbanization, the performance of carbon emissions reduction in logistics is significantly different among regions. This paper establishes a new indicator to describe the carbon emissions catch-up effect and decomposes the impact of land urbanization into 4 driving factors, thereby identifying the impact of land urbanization on carbon emissions catch-up effect in detail. The results indicate that: (1) at the national level, the catch-up effect of carbon emissions in logistics showed three stages of “catching up-lagging behind-catching up”, which was consistent with the development of logistics. (2) At the regional level, the land urbanization-related factors had significant but different impacts on the catch-up effect of carbon emissions. The spatial expansion and road network density effect were the main inhibitors for catch-up effect of the eastern region, and spatial structure effect was the main inhibitor for central and western regions. (3) At the provincial level, the impact of land urbanization-related factors on provinces at different development stages was different. Finally, some policy implications are proposed to achieve carbon neutrality targets.

As carbon emissions are increasing due to the development of economy, low-carbon supply chain plays an important role in carbon emissions reduction and the dual-channel supply chain has become a hit because online shopping is developing rapidly. Therefore, this paper builds a Stackelberg game model led by the manufacturer in a dual-channel supply chain to examine the reaction of the government under centralized or decentralized decisions-making structures with different low-carbon strategies. The result shows that the government can achieve higher profits by taking incentive or punitive measures for centralized decision-making supply chain no matter they invest in emissions reduction or not. Moreover, for decentralized decision-making mode, increasing low-carbon subsidies for retailers can achieve a win-win result between the supply chain and the government; and, finally, channel competition is good for improving the supply chain and social benefits. Therefore, the government is responsible for taking reasonable subsidy policies, formulate industry’s low-carbon standards, and properly guide competition between supply chain members to achieve higher profits.

The promotion of industrial environmental efficiency (IEE) has attracted considerable attention from scholars and policymakers. Previous studies have focused on the factors driving IEE without considering the leader-follower relationship between the two stages of IEE and have used aggregated indicators to detect the effect of urbanization on IEE. In this study, we open the “black box” of IEE and regard it as a serial system comprised of the economic development (ED) stage and the environmental governance (EG) stage. We select typical provinces belonging to the ED-oriented regions and EG-oriented regions for our analysis. We quantify IEE by using the slack-based model (SBM), decompose the effect of urbanization into four typical factors and detect the effect of the urbanization-related factors by using a tobit regression model. Next, we compare the effect of the urbanization-related factors in the different regions. The results show that overall, in 2011–2015, IEE in the different regions was low and fluctuating, and it increased in ED-oriented regions while remaining stable in EG-oriented regions. The IEE of the ED-oriented regions was always higher than that of the EG-oriented regions, with the gap increasing over time. The IEE of the two stages varied in each type of region. Urbanization-related factors had different effects on the IEE of the different stages in the two types of region, and the same factor had significantly different effects in different regions.

The aim of this study is to identify the dynamic explicit and implicit information factors which displayed on the webpage of platforms that influence backers' investment decision-making behavior. We analyze the connections among these factors by collecting the longitudinal dataset from reward-based crowdfunding platform. Based on ELM model, we establish Fixed Estimation Panel Data Model respectively according to explicit and implicit factors and take Funding Status (crowdfunding results) as the moderating variable to observe the goal gradient effect. Results indicate that most variables in the central route affect backers' investment behavior positively, while most variables in the periphery route have a negative impact on backers' investment behavior. The Funding Status has a significant negative moderating effect on the explicit variables, and has no significant moderating effect on the implicit information variables of the project. In addition, we upgrade the econometric method used by previous scholars, which could improve the accuracy of the FE model. Furthermore, we find strong support for the herding effect in reward-based crowdfunding and the intensity tends to decrease before the funding goal draws near.

Fake news has been spreading intentionally and misleading society to believe unconfirmed information; this phenomenon makes it challenging to identify fake news based on shared content. Fake news circulation is not only a current issue, but it has been disseminated for centuries. Dealing with fake news is a challenging task because it spreads massively. Therefore, automatic fake news detection is urgently needed. We introduced TB-BCG, Topic-Based BART Counterfeit Generator, to increase detection accuracy using deep learning. This approach plays an essential role in selecting impacted data rows and adding more training data. Our research implemented Latent Dirichlet Allocation (Topic-based), Bidirectional and Auto-Regressive Transformers (BART), and Cosine Document Similarity as the main tools involved in Constraint @ AAAI2021-COVID19 Fake News Detection dataset shared task. This paper sets forth this simple yet powerful idea by selecting a dataset based on topic and sorting based on distinctive data, generating counterfeit training data using BART, and comparing counterfeit-generated text toward source text using cosine similarity. If the comparison value between counterfeit-generated text and source text is more than 95%, then add that counterfeit-generated text into the dataset. In order to prove the resistance of precision and the robustness in various numbers of data training, we used 30%, 50%, 80%, and 100% from the total dataset and trained it using simple Long Short-Term Memory (LSTM) and Convolutional Neural Network (CNN). Compared to baseline, our method improved the testing performance for both LSTM and CNN, and yields are only slightly different.

Based on the existing classification research of rockburst dynamic disasters, a new rockburst model is introduced. The model specimens with inclusion rock are designed and produced, and the indoor test of inclusion rockburst with or without cementation, different inclusion angles, and prefabricated cracks of different lengths is carried out. The failure process and displacement variation of rock specimens are recorded and measured by the high-speed camera system and digital image correlation method, and based on this, the characteristics and laws of rockburst are analyzed. The results show that the failure process of inclusion rock can be divided into several periods, such as crushing and extrusion of the cement body, and the cracks in the upper left tip, the lower right tip, the lower left tip, and the upper tip are expanded and broken in different orders. The dynamic ejection failure of the rock fragments occurs when the degree of failure is severe, and the static failure occurs when the degree of failure is weak. The inclusions have cementation, and the better the cementation, the greater the stress loading required in the failure, the greater the energy accumulated, the more severe the failure, and the higher the fractal dimension of the fragments. The larger the angle of the inclusion tip, the greater the stress loading in failure; the greater the change of the u and v displacements, the more the displacement of the inclusions changed; and the more the energy accumulated, the higher the fractal dimension of the fragment. The smaller the crack length, the greater the stress loading in failure, and the more the energy accumulated, the higher the fractal dimension of the fragments in failure. With more number of cracks and the more severe expansion before failure, the failure of rock mass can be transformed from the dynamic to the static.

In some online social network services (SNSs), the members are allowed to label their relationships with others, and such relationships can be represented as the links with signed values (positive or negative). The networks containing such relations are named signed social networks (SSNs), and some real-world complex systems can be also modeled with SSNs. Given the information of the observed structure of an SSN, the link prediction aims to estimate the values of the unobserved links. Noticing that most of the previous approaches for link prediction are based on the members’ similarity and the supervised learning method, however, research work on the investigation of the hidden principles that drive the behaviors of social members are rarely conducted. In this paper, the deep belief network (DBN)-based approaches for link prediction are proposed. Including an unsupervised link prediction model, a feature representation method and a DBN-based link prediction method are introduced. The experiments are done on the datasets from three SNSs (social networking services) in different domains, and the results show that our methods can predict the values of the links with high performance and have a good generalization ability across these datasets.

Identification of protein binding sites is critical for studying the function of the proteins. In this paper, we proposed a method for protein binding site prediction, which combined the order profile propensities and hidden Markov support vector machine (HM-SVM). This method employed the sequential labeling technique to the field of protein binding site prediction. The input features of HM-SVM include the profile-based propensities, the Position-Specific Score Matrix (PSSM), and Accessible Surface Area (ASA). When tested on different data sets, the proposed method showed promising results, and outperformed some closely relative methods by more than 10% in terms of AUC.