Search

Coaxial electrostatic spinning (co‐electrostatic spinning) technology has greatly expanded the versatility of the preparation of core–shell polymer nanofibers and has found a wide range of applications in the environmental and biological fields. Here we present a method for the preparation of coaxial nanofibers using polyacrylonitrile (PAN) and polyurethane (PU) as raw materials. It was found that the tensile strength ranges from 2.14 to 4.07 MPa with the increasing spinning speed of the nucleated PU layer, and the elongation at break was up to 95.09% for M6:4, which was three times higher than that of the original MPAN (30.54%), and the toughness of the nanofiber film was also significantly improved. Finally, the oil/water separation capacity of the coaxial nanofiber membrane was investigated, and the results showed that the separation fluxes for various oil compounds ranged from 2380.18 to 3130.17 L·m−2·h−1, with separation efficiencies above 99%. This study not only investigates the effect of different flow rates of core (PU)/shell (PAN) on the performance of coaxial electrostatic spun nanofiber membranes, but also provides a new insight into the coaxial electrostatic spinning process. [ABSTRACT FROM AUTHOR]

With the exponential growth of high dimensional unlabeled data, unsupervised feature selection (UFS) has attracted considerable attention due to its excellent performance in machine learning. Existing UFS methods implicitly assigned the same attribute score to each sample, which disregarded the distinctiveness of features and weakened the clustering performance of UFS methods to some extent. To alleviate these issues, a novel UFS method is proposed, named unsupervised feature selection with latent relationship penalty term (LRPFS). Firstly, latent learning is innovatively designed by assigning explicitly an attribute score to each sample according to its unique importance in clustering results. With this strategy, the inevitable noise interference can be removed effectively while retaining the intrinsic structure of data samples. Secondly, an appropriate sparse model is incorporated into the penalty term to further optimize its roles as follows: (1) It imposes potential constraints on the feature matrix to guarantee the uniqueness of the solution. (2) The interconnection between data instances is established by a pairwise relationship situation. Extensive experiments on benchmark datasets demonstrate that the proposed method is superior to relevant state-of-the-art algorithms with an average improvement of 10.17% in terms of accuracy. [ABSTRACT FROM AUTHOR]

Endless rolling urgently requires an increase in the casting speed of continuous casting. For the continuous casting process of a high-casting-speed billet, the heat flux of the mold would be much higher, requiring a stronger cooling performance and longer mold life to match the high-speed casting. Mold material, thickness, and slot structure have a great influence on the casting speed. To design a more efficient billet casting mold, a three-dimensional thermal-stress-coupled analysis model of a 150 mm × 150 mm mold was established in this research to analyze the thermal state of a mold with high casting speed; in addition, the material, thickness, and water slot structure, which pertain to the mold cooling performance, were also studied. The results show that the billet mold of Cu-Ag with a thinner thickness and right-corner water slot is better in terms of casting speed. Regarding the material, the Cu-Ag mold has a higher thermal conductivity efficiency; its hot surface temperature is 4.89 °C lower, its equivalent stress is 7 MPa lower, and its longitudinal deformation is 0.0023% lower compared with the deoxidized phosphorus copper mold. Regarding the thickness, the thinner mold has a 60.76 °C lower hot surface temperature, its equivalent stress is 340 MPa lower, and its longitudinal deformation is 0.0443% lower compared with the thicker mold. For the water slot structure, the mold with the right-angled water slot has a 2.895 °C lower hot surface temperature, its equivalent stress is 37 MPa lower, and its longitudinal deformation is 0.0039% lower compared with the mold with a rounded-corner water slot. [ABSTRACT FROM AUTHOR]

The present work studied various models for predicting turbulence in the problem of injecting a fluid microjet into the boundary layer of a turbulent flow. For this purpose, the one-equation Spalart–Allmaras (SA), two-equation k–ε and k–ω, multi-equation transition k-kL–ω, transition shear stress transport (SST), and Reynolds stress models were used for solving the steady microjet into the turbulent boundary layer, and their results are compared with experimental results. Comparing the results indicated that the steady solution methods performed sufficiently we for this problem. Furthermore, it was found that the four-equation transition SST model was the most accurate method for predicting turbulence in this problem. This model predicted the velocity along the x-axis in near- and far-jet locations with about 1% and 5% average errors, respectively. It also outperformed the other methods in predicting Reynolds stresses, especially at the center (nearly 5% error). Moreover, the modified four-equation transition SST model has improved the system's performance in predicting the studied parameters by utilizing Sørensen correlations in predicting R e θ t (the transition momentum thickness Reynolds number), F l ength (an empirical correlation that controls the length of the transition region), and R e θ c (the critical Reynolds number where the intermittency first starts to increase in the boundary layer). [ABSTRACT FROM AUTHOR]

Semi-supervised non-negative matrix factorization (NMF) has achieved successful results due to the significant ability of image recognition by a small quantity of labeled information. However, there still exist problems to be solved such as the interconnection information not being fully explored and the inevitable mixed noise in the data, which deteriorates the performance of these methods. To circumvent this problem, we propose a novel semi-supervised method named DLRGNMF. Firstly, dual latent space is characterized by the affinity matrix to explicitly reflect the interrelationship between data instances and feature variables, which can exploit the global interconnection information in dual space and reduce the adverse impacts caused by noise and redundant information. Secondly, we embed the manifold regularization mechanism in the dual graph to steadily retain the local manifold structure of dual space. Moreover, the sparsity and the biorthogonal condition are integrated to constrain matrix factorization, which can greatly improve the algorithm's accuracy and robustness. Lastly, an effective alternating iterative updating method is proposed, and the model is optimized. Empirical evaluation on nine benchmark datasets demonstrates that DLRGNMF is more effective than competitive methods. [ABSTRACT FROM AUTHOR]