Modeling Complex Networks Based on Deep Reinforcement Learning

The network topology of complex networks evolves dynamically with time. How to model the internal mechanism driving the dynamic change of network structure is the key problem in the field of complex networks. The models represented by WS, NW, BA usually assume that the evolution of network structure is driven by nodes’ passive behaviors based on some restrictive rules. However, in fact, network nodes are intelligent individuals, which actively update their relations based on experience and environment. To overcome this limitation, we attempt to construct a network model based on deep reinforcement learning, named as NMDRL. In the new model, each node in complex networks is regarded as an intelligent agent, which reacts with the agents around it for refreshing its relationships at every moment. Extensive experiments show that our model not only can generate networks owing the properties of scale-free and small-world, but also reveal how community structures emerge and evolve. The proposed NMDRL model is helpful to study propagation, game, and cooperation behaviors in networks.

Reutilizing Waste Iron Tailing Powders as Filler in Mortar to Realize Cement Reduction and Strength Enhancement

Recently, the massive accumulation of waste iron tailings powder (WITP) has resulted in significant environmental pollution. To solve this problem, this paper proposes an original mortar replacement (M) method to reuse waste solids and reduce cement consumption. In the experiment, the author employed an M method which replaces water, cement, and sand with WITP under constant water/cement and found that the strength development can be significantly improved. Specifically, a mortar with 20% WITP replacement can obtain a 30.95% improvement in strength development. To study the internal mechanism, we performed experiments such as thermogravimetric analysis (TGA), mercury intrusion porosimetry (MIP), and SEM. The results demonstrate that the nucleation effect and pozzolanic effect of WITP can help promote cement hydration, and MIP reveals that WITP can effectively optimize pore structure. In addition, 1 kg 20% WITP mortar reduced cement consumption by 20%, which saves 19.98% of the economic cost. Comprehensively, our approach achieves the effective utilization of WITP and provides a favorable reference for practical engineering.

Tunable dual plasmon-induced transparency and slow-light analysis based on monolayer patterned graphene metamaterial

We propose a novel terahertz metamaterial structure based on patterned monolayer graphene. This structure produces an evident dual plasmon-induced transparency (PIT) phenomenon due to destructive interference between bright and dark modes. Since the Fermi level of graphene can be adjusted by an external bias voltage, the PIT phenomenon can be tuned by adjusting the voltage. Then the coupled-mode theory (CMT) is introduced to explore the internal mechanism of the PIT. After that, we investigate the variation of absorption rate at different graphene carrier mobilities, and it shows that the absorption rate of this structure can reach 50%, which is a guideline for the realization of graphene terahertz absorption devices. In addition, through the study of the slow-light performance for this structure, it is found that its group index is as high as 928, which provides a specific theoretical basis for the study of graphene slow-light devices.

Piezoelectric catalytic, photocatalytic and adsorption capability and selectivity removal of various dyes and mixed dye wastewater by ZnO nanoparticles

The C-O functional group decorated ZnO nanoparticles with high UV absorption and VIS/NIR reflectance were synthesized by a simple wet chemistry method using various chelating agents. This study attempts to explore the internal mechanism of the piezoelectric catalytic activity, photocatalytic activity and adsorption performance of ZnO nanoparticles. The phase purity, particle size, optical band gap and photocatalytic activity of ZnO nanoparticles showed strong chelating agent - dependent behavior. The ZnO nanoparticles prepared by using EDTA as a chelating agent exhibits smallest particle size, highest photocatalytic activity for the degradation of methyl orange, methylene blue and rhodamine B, high adsorption capacity for the adsorption of Congo red and high vibration-catalytic performance for the vibration degradation of rhodamine B. The synergies mechanism among piezoelectric catalysis, photocatalysis and adsorption capacity of ZnO nanoparticles are discussed on the basis of the experimental results.

Digital Print Synthesis Based on Image Processing and Interactive Technology

Fire is a common disaster, which causes major threats and losses to human life and property. Countries around the world have been committed to the study of the mechanism and internal mechanism of fires, with the goal of preventing fires from occurring and minimizing the losses caused by fires. Among the many methods, fire detection technology is an effective method to prevent and reduce the occurrence of fire. This article focuses on the research of the fire detection system based on artificial intelligence technology, improves the accuracy of the fire detection system by introducing artificial intelligence technology into the fire detection system, and uses experiments to verify the error rate of the artificial intelligence technology fire detection system. The experimental results show that the system’s detection of fire is not very different from the actual situation, and the error rate is within 10%. Then compared with the traditional detection system, the detection performance is relatively high, and the error rate can be reduced by one time.

Nano-Modified Meta-Aramid Insulation Paper with Advanced Thermal, Mechanical, and Electrical Properties

Molecular dynamics simulations were used to analyze the internal mechanism for the observed improvement in performance of nano-modified meta-aramid insulation paper from a microscopic point of view. The results showed that the k-polyphenylsilsesquioxane(PPSQ) modified meta-aramid insulation paper was superior to b-PPSQ modified meta-aramid insulation paper in terms of its thermal stability and mechanical and electrical properties. The analysis of microscopic parameters showed that the stiffness of k-PPSQ was less than that of b-PPSQ, and the hydroxyl groups on the open-loop system were more likely to enter the dispersed system, resulting in higher bonding strength, meta-aramid fiber chains between k-PPSQ molecules, and the formation of hydrogen bonds. Additionally, the nano-enhancement effects of k-PPSQ and b-PPSQ resulted in various improvements, including a reduction in pores between molecules in the blend model, an increase in the contact area, the formation of interfacial polarization, and a reduction in defects at the interface.

Effects of Aging on the Dry Shrinkage Cracking of Lime Soils with Different Proportions

In this study, the reasonable aging time of lime soils with different ratios was determined to investigate the effect of aging on the internal mechanism of the dry shrinkage cracking of lime soil. To this end, the effects of aging time, lime content, and particle size on the volume crack rate, expansion and shrinkage rate, particle size distribution, and pH were analyzed using a dry–wet cycle, screening, and pH tests. In addition, the changes in the particle structure of the samples and the formation of new substances were analyzed. The results revealed that the volume crack and expansion shrinkage rates of the sample initially decreased, and then stabilized with increasing aging time. In addition, the aging time of the sample increased with increasing lime content and particle size. Further, at the initial aging stage, the volume crack and expansion/shrinkage rates of the sample increased with increasing lime content and particle size. With an increase in the aging time, the pH increased and then decreased significantly to 0.57–1.1% at the reasonable aging time. These indicate that the pH exhibited a significant effect on the reasonable aging time, and this will provide useful insights for the restoration of lime soil sites.

The Triggering Mechanism of Short Video Customer Inspiration – Qualitative Analysis Based on the Repertory Grid Technique

It is believed that stimulating the inspiration of short video consumers might be an effective way to attract and maintain the attention of consumers so that they are willing to respond positively to short video ads. Therefore, in order to explore the source of customer inspiration in short video and its cognitive psychological process, the text and grid data collected from an interview among 25 short video users have been qualitatively analyzed by Kelly Grid Technology in order to construct the formation path model of short video customer inspiration, and find out its source, triggering mechanism, and influencing factors. It is found that the inspiring informational content characteristics include richness, reliability, vividness, and fluency of emotional content characteristics, fun, novelty, and narrative. However, the characteristics of commercial content in short video ads hinder the inspiration of consumers. The study also reveals that an internal mechanism of inspiration stimulation is built on some cognitive processes (i.e., presence, processing fluency, perceived innovation, perceived convenience) generated by informational content, and emotional responses by emotional content (i.e., curiosity, surprise, enjoyment, etc.). In addition, it is shown that personal involvement enhances the relationship between the inspiring content characteristics and consumer inspiration. As a result, customer inspiration and engagement in short video ads are highly enriched. Findings provide implications for short video platforms and online marketers.