Search results

The flow phenomenon of particle image velocimetry has revealed the transition process of the complex multi-scale vortex between the boundary layer and mainstream region. Nonetheless, present computational fluid dynamics methods inadequately distinguish the discernable flows in detail. A multi-physical field coupling model, which was applied in rotor-stator fluid machinery (Umavathi, 2015; Syawitri et al., 2020), was put forward to ensure the identification of multi-scale vortexes and the improvement of performance prediction in torque converter.

A newly-developed multi-physical field simulation framework that coupled the scale-resolving simulation method with a dynamic modified viscosity coefficient was proposed to comparatively investigate the influence of energy exchange on thermal and flow characteristics and the description of the flow field in detail.

Regardless of whether quantitative or qualitative, its description ability on turbulence statistics, pressure-streamline, vortex structure and eddy viscosity ratio were visually experimentally and numerically analyzed. The results revealed that the modification of transmission medium viscous can identify flows more exactly between the viscous sublayer and outer boundary layer. Compared with RANS and large eddy simulation, a stress-blended eddy simulation model with a dynamic modified viscosity coefficient, which was further used to achieve blending on the stress level, can effectively solve the calculating problem of the transition region between the near-wall boundary layer and mainstream region.

This indeed provides an excellent description of the transient flow field and vortex structure in different physical flow states. Furthermore, the experimental data has proven that the maximum error of the external performance prediction was less than 4%.

An improved model was applied to simulate and analyze the flow mechanism through the evolution of vortex structures in a working chamber, to deepen the designer with a fundamental understanding on how to reduce flow losses and flow non-uniformity in manufacturing.

Fine-tuning pre-trained language models (PLMs), e.g. SciBERT, generally require large numbers of annotated data to achieve state-of-the-art performance on a range of NLP tasks in the scientific domain. However, obtaining fine-tuning data for scientific NLP tasks is still challenging and expensive. In this paper, the authors propose the mix prompt tuning (MPT), which is a semi-supervised method aiming to alleviate the dependence on annotated data and improve the performance of multi-granularity academic function recognition tasks.

Specifically, the proposed method provides multi-perspective representations by combining manually designed prompt templates with automatically learned continuous prompt templates to help the given academic function recognition task take full advantage of knowledge in PLMs. Based on these prompt templates and the fine-tuned PLM, a large number of pseudo labels are assigned to the unlabelled examples. Finally, the authors further fine-tune the PLM using the pseudo training set. The authors evaluate the method on three academic function recognition tasks of different granularity including the citation function, the abstract sentence function and the keyword function, with data sets from the computer science domain and the biomedical domain.

Extensive experiments demonstrate the effectiveness of the method and statistically significant improvements against strong baselines. In particular, it achieves an average increase of 5% in Macro-F1 score compared with fine-tuning, and 6% in Macro-F1 score compared with other semi-supervised methods under low-resource settings.

In addition, MPT is a general method that can be easily applied to other low-resource scientific classification tasks.

Corporate social responsibility (CSR) and innovation are the two main approaches firms utilize to promote sustainable development. However, as yet, scholars have reached no consensus regarding their precise impact on construction firm performance (CFP), hindering efforts to implement effective sustainable development strategies that improve CFP. In view that a simple linear relationship may not be sufficient to capture their precise pattern, this study aims to unveil the nonlinear impact of CSR and innovation on CFP, especially when construction firms take up a distinct competitive position.

This study first proposed four hypotheses to establish a new theoretical model by incorporating CSR, innovation, CFP and construction firms' competitive position (CFCP). Then the model was tested by using 292 annual observations collected from 75 construction firms in China. A multiple regression model analysis was carried out to analyze the survey data and validate the hypotheses.

The results reveal that both CSR and innovation have a U-shaped impact on the price-to-book ratio of a construction firm, a specific CFP measure. CFCP negatively moderates the U-shaped relationship between CSR and CFP, but positively moderates the U-shaped relationship between innovation and CFP.

This study goes beyond a simple linear view, instead of unveiling the nonlinear U-shaped effects of CSR and innovation on CFP that deepen the understanding of their complex relationships in the construction industry and makes construction firms aware that CSR and innovation can only improve performance if they reach a certain level. The moderating role of CFCP provides important implications for construction firms seeking to adopt appropriate competitive strategies related to social responsibility and innovation that both promote CFP and achieve sustainable development.

The current popular image processing technologies based on convolutional neural network have the characteristics of large computation, high storage cost and low accuracy for tiny defect detection, which is contrary to the high real-time and accuracy, limited computing resources and storage required by industrial applications. Therefore, an improved YOLOv4 named as YOLOv4-Defect is proposed aim to solve the above problems.

On the one hand, this study performs multi-dimensional compression processing on the feature extraction network of YOLOv4 to simplify the model and improve the feature extraction ability of the model through knowledge distillation. On the other hand, a prediction scale with more detailed receptive field is added to optimize the model structure, which can improve the detection performance for tiny defects.

The effectiveness of the method is verified by public data sets NEU-CLS and DAGM 2007, and the steel ingot data set collected in the actual industrial field. The experimental results demonstrated that the proposed YOLOv4-Defect method can greatly improve the recognition efficiency and accuracy and reduce the size and computation consumption of the model.

This paper proposed an improved YOLOv4 named as YOLOv4-Defect for the detection of surface defect, which is conducive to application in various industrial scenarios with limited storage and computing resources, and meets the requirements of high real-time and precision.

This study purposes to study the influence of artificial freezing on the liquefaction characteristics of Nanjing sand, as well as its mechanism.

was studied through dynamic triaxial tests by means of the GDS dynamic triaxial system on Nanjing sand extensively discovered in the middle and lower reaches of the Yangtze River under seismic load and metro train vibration load, respectively, and potential hazards of the two loads to the freezing construction of Nanjing sand were also identified in the tests.

The results show that under both seismic load and metro train vibration load, freeze-thaw cycles will significantly reduce the stiffness and liquefaction resistance of Nanjing sand, especially in the first freeze-thaw cycle; the more freeze-thaw cycles, the worse structural behaviors of silty-fine sand, and the easier to liquefy; freeze-thaw cycles will increase the sensitivity of Nanjing sand's dynamic pore pressure to dynamic load response; the lower the freezing temperature and the effective confining pressure, the worse the liquefaction resistance of Nanjing sand after freeze-thaw cycles; compared to the metro train vibration load, the seismic load in Nanjing is potentially less dangerous to freezing construction of Nanjing sand.

The research results are helpful to the construction of the artificial ground freezing of the subway crossing passage in the lower reaches of the Yangtze River and to ensure the construction safety of the subway tunnel and its crossing passage.

This paper aims to report the influence of hexamethylenetetramine (HMTA) on phosphate coatings formed on AZ31 magnesium alloys.

These phosphate coatings were obtained by immersing magnesium alloys in phosphate baths with HMTA. The morphology and composition of the phosphate coatings were investigated via scanning electron microscopy, energy dispersive spectrometry and X-ray diffraction.

The phosphate coatings were mainly composed of CaHPO₄·2H₂O. The HMTA concentration in the phosphate bath influenced the crystallization and corrosion resistance of the phosphate coating.

The polarization curve shows that the anti-corrosion qualities of the phosphate coating were optimal when the HMTA concentration was 1.0 g/L in the phosphate bath. Electrochemical impedance spectroscopy (EIS) shows that the electrochemical impedances increased gradually when the HMTA concentration varied from 1.0 to 3.0 g/L.

In this paper the interconnectedness among financial institutions and the level of systemic risks of four types of Chinese financial institutions are investigated.

By the means of RAS algorithm, the interconnection among financial institutions are illustrated. Different methods, including Linear Granger, Systemic impact index (SII), vulnerability index (VI), CoVaR, and MES are used to measure the systemic risk exposures across different institutions.

The results illustrate that big banks are more interconnected and hold the biggest scales of inter-bank transactions in the financial network. The institutions which have larger size tend to have more connection with others. Insurance and security companies contribute more to the systemic risk where as other institutions, such as trusts, financial companies, etc. may bring about severe loss and endanger the financial system as a whole.

Since other institutions with low levels of regulation may bring about higher extreme loss and suffer the whole system, it deserves more attention by regulators considering the contagion of potential risks in the financial system.

This study builds a valuable contribution by examine the systemic risks from the perspectives of both interconnection and tail risk measures. Furthermore; Four types financial institutions are investigated in this paper.