Search

The increasing amount of network data and the development of big data technology have brought development opportunities and challenges to recommendation systems. Model-based collaborative filtering has become a mainstream algorithm in recommendation systems. For example, the Funk singular- value decomposition (FunkSVD) algorithm. However, in big data calculations, data sparseness and iterative oscillations often affect the accuracy of the FunkSVD algorithm. Moreover, when the data volume is in units of GB or more, the FunkSVD algorithm runs slowly and is ineffective. Therefore, we propose an improved FunkSVD algorithm (IFABG) based on the root-mean-square prop (RMSProp) and graphics processing unit (GPU) to solve this problem. First, we used RMSProp algorithm to improve the traditional FunkSVD algorithm, alleviate data sparseness and iterative shock, and improve the prediction accuracy of the algorithm. Next, we implemented parallelization of the improved FunkSVD algorithm in the GPU, which increased the calculation speed of the algorithm. Finally, we verify the IFABG algorithm using the Movielens dataset. The experimental results show that the IFABG algorithm is very suitable for processing sparse data, and it alleviates the iterative shock. The prediction accuracy rate is approximately 30% higher than that of the traditional FunkSVD algorithm. The experimental results also demonstrate that the IFABG algorithm has a good acceleration effect. For the same size dataset, the IFABG algorithm is faster than the traditional FunkSVD, and the acceleration ratio can be as high as 19.27.

Aiming at the low classification accuracy of unbalanced data sets, an improved SMOTE over-sampling algorithm ACC-SMOTE (Ant Colony Clustering Synthetic Minority Oversampling Technology) based on ant colony clustering is proposed. On the one hand, the improved ant colony clustering algorithm is used to divide a small number of samples into different sub-clusters, fully considered the imbalance between inter-cluster and intra-cluster data, and SMOTE algorithm is used to oversample the samples according to the proportion of sub-clusters, to reduce the imbalance of intra-class data. On the other hand, Tomek Links data cleaning technology is used to correct the oversampled samples in time, the quality of synthetic samples is guaranteed by eliminating noise in data sets and overlapping samples generated by sampling methods. The training data set and the test data set used in this paper are both UCI data sets. The experimental results show that this algorithm can significantly improve the classification accuracy of a few classes, thus improving the classification performance of the classifier.

In response to the challenges of intermediate poisoning and the high cost of noble metal catalysts in the hydrogen evolution reaction (HER), we develop a Ru-doped SnO2 catalyst. This Ru–SnO2 catalyst has the characteristics of low Ru loading and a hollow structure, which endow it with good electrocatalytic activity and stability for the HER. [ABSTRACT FROM AUTHOR]

Pt‐based intermetallics are recognized as effective catalysts for oxygen reduction reaction (ORR) in fuel cells. However, the synthesis of intermetallics often requires prolonged annealing and the effect of different crystal structures on ORR still need to be investigated. Herein, L12‐Pt3Mn and L10‐PtMn intermetallics with Pt‐skin are rapidly synthesized through an emerging periodic carbothermal shock method to investigate their ORR‐activity difference. The formation of L12‐Pt3Mn and L10‐PtMn can be well‐controlled by the Pt/Mn feeding ratio, pulse cycles, and temperature. Electrocatalytic investigations show that L12‐Pt3Mn presents a more positive half‐wave potential (0.91 V vs RHE) and onset potential (1.02 V vs RHE) than those of L10‐PtMn. The mass activity and specific activity of L12‐Pt3Mn are respectively fourfold and threefold greater than those of L10‐PtMn. Theoretical calculations indicate that the L12‐Pt3Mn has a more substantial work function than L10‐PtMn, thereby conferring L12‐Pt3Mn with an increased electron density allocation for catalytic involvement. This electron confinement imparts L12‐Pt3Mn with a Pt d‐band center lower than that of L10‐PtMn, consequently attenuating the adsorption of strongly bonded *O intermediates during the rate‐determining step. This study not only employs a straightforward method for intermetallic preparation but also elucidates the discrepancies in ORR activity across intermetallics with distinct structures. [ABSTRACT FROM AUTHOR]

In the era of big data, the complexity of data is increasing. Problems such as data imbalance and class overlap pose challenges to traditional classifiers. Meanwhile, the importance of imbalanced data has become increasingly prominent, it is necessary to find appropriate methods to enhance classification performance of classifiers on such datasets. In response, this paper proposes a mixed sampling method (ISODF-ENN) based on iterative self-organizing (ISODATA) denoising diffusion algorithm and edited nearest neighbors (ENN) data cleaning algorithm. The algorithm first uses iterative self-organizing clustering algorithm to divide minority class into different sub-clusters, then it uses denoising diffusion algorithm to generate new minority class data for each sub-cluster, and finally it uses ENN algorithm to preprocess majority class data to remove the overlap with the minority class data. Each sub-cluster is oversampled according to sampling ratio, so that the oversampled minority class data also conforms to the distribution of original minority class data. Experimental results on keel datasets demonstrate that the proposed method outperforms other methods in terms of F-value and AUC, effectively addressing the issues of class imbalance and class overlap. [ABSTRACT FROM AUTHOR]

Ruthenium dioxide (RuO2) has been considered a benchmark electrocatalyst for the oxygen evolution reaction, according to the volcano theory. However, the comparatively strong adsorption of RuO2 and oxygenated intermediates restricts its performance during the actual oxygen evolution reaction (OER) process. Here, we have developed a rutile-structured ruthenium–tin solid-solution oxide (Ru0.6Sn0.4O2) by using a competitive adsorption strategy to accelerate the OER kinetics. Sn was introduced into RuO2 to construct a solid-solution oxide structure, given the similar ionic radii, identical cation valence, and the same crystal structure for the oxides. As a consequence, the Ru0.6Sn0.4O2 catalyst displayed a low overpotential of 245 mV, reaching 10 mA cm−2 current density and a low Tafel slope of 61.80 mV dec−1 in 1.0 M KOH electrolyte. Meanwhile, theoretical calculations revealed that the incorporation of Sn components decreased the anti-bonding spin states of the Ru d orbitals, which modulated the energy barrier of the rate-determining step, thus accelerating the OER kinetics. This approach extends the ideas for designing efficient solid-solution oxide electrocatalysts. [ABSTRACT FROM AUTHOR]

We propose a discontinuous least squares finite element method for solving the Helmholtz equation. The method is based on the L2$$ {L}^2 $$ norm least squares functional with the weak imposition of the continuity across the interior faces as well as the boundary conditions. We minimize the functional over the discontinuous polynomial spaces to seek numerical solutions. The wavenumber explicit error estimates to our method are established. The optimal convergence rate in the energy norm with respect to a fixed wavenumber is attained. The least squares functional can naturally serve as a posteriori estimator in the h‐adaptive procedure. It is convenient to implement the code due to the usage of discontinuous elements. Numerical results in two and three dimensions are presented to verify the error estimates. [ABSTRACT FROM AUTHOR]

Using the numerical simulation method and the drafting software of Surfer, we charted the temperature fields and remaining oil saturation fields of steam assisted gravity drainage (SAGD) and steam and gas push (SAGP) in oil drainage periods, including beginning, peak stage and last stage of oil drainage. An intuitive and reliable comparative research was done about the formation and expanding process of steam chambers, the shape of steam chambers and flow features in different oil drainage periods. The recovery effectiveness of the two production methods was compared. The results show that: in the SAGP production process, the lateral expansion of the steam chamber is faster than that in SAGD and the vertical expansion slower than that in SAGD. By adding non-condensate gas in the steam of SAGD, it can reduce the amount of steam injection, increase thermal efficiency, improve the ratio of oil to steam, and ultimately improve the development effectiveness of SAGD. The case study indicates that SAGP can slow the decline of oil rate and extend the development time of heavy oil reservoir and improve the development effectiveness. Key words: heavy oil reservoir, flow feature, SAGD, SAGP, numerical simulation

For direct formic acid fuel cells (DFAFC), palladium (Pd)-based alloy catalysts with competitive morphology and elemental composition are essential to boost the performance of the formic acid oxidation reaction (FAOR) in the anode zone. Herein, we design and synthesize RuPdx alloy nano-network structures (ANs) via the facile wet-chemical reduction of Pd-Ru cyanogel (Pdx [Ru(CN)6]y·aH2O) as an effective electrocatalyst for the FAOR. The formation of Pd-Ru cyanogel depends on the facile coordination of K2PdCl4 and K3 [Ru(CN)6]. The unique structure of cyanogel ensures the presentation of a three-dimensional mesoporous morphology and the homogeneity of the elemental components. The as-prepared RuPd3 ANs exhibit good electrocatalytic activity and stability for the FAOR. Notably, the RuPd3 ANs achieve a mass-specific activity of 2068.4 mA mg−1 in FAOR, which shows an improvement of approximately 16.9 times compared to Pd black. Such a competitive FAOR performance of RuPd3 ANs can be attributed to the advantages of structure and composition, which facilitate the exposure of more active sites, accelerate mass/electron transfer rates, and promote gas escape from the catalyst layer, as well as enhance chemical stability. [ABSTRACT FROM AUTHOR]

Ruthenium (Ru) has been theoretically considered a viable alkaline hydrogen evolution reaction electrocatalyst due to its fast water dissociation kinetics. However, its strong affinity to the adsorbed hydroxyl (OHad) blocks the active sites, resulting in unsatisfactory performance during the practical HER process. Here, we first reported a competitive adsorption strategy for the construction of SnO2 nanoparticles doped with Ru single‐atoms supported on carbon (Ru SAs‐SnO2/C) via atomic galvanic replacement. SnO2 was introduced to regulate the strong interaction between Ru and OHad by the competitive adsorption of OHad between Ru and SnO2, which alleviated the poisoning of Ru sites. As a consequence, the Ru SAs‐SnO2/C exhibited a low overpotential at 10 mA cm−2 (10 mV) and a low Tafel slope of 25 mV dec−1. This approach provides a new avenue to modulate the adsorption strength of active sites and intermediates, which paves the way for the development of highly active electrocatalysts. [ABSTRACT FROM AUTHOR]

To solve the problem of low accuracy in the traditional FunkSVD recommendation algorithm, an improved FunkSVD algorithm (RM-FS) is proposed. RM-FS is an improvement of the traditional FunkSVD algorithm, using RMSProp, a deep learning optimization algorithm. The RM-FS algorithm can not only solve the problem of reduced accuracy of the traditional FunkSVD algorithm because of iterative oscillations but also alleviate the impact of data sparseness on the accuracy of the algorithm, achieving the effect of improving the accuracy of the traditional algorithm. The experimental results show that the RM-FS algorithm proposed in this paper effectively improves the accuracy of the recommendation algorithm, which is better than the traditional FunkSVD recommendation algorithm and other improved FunkSVD algorithms. [ABSTRACT FROM AUTHOR]

We propose and analyze a discontinuous least squares finite element method for solving the indefinite time-harmonic Maxwell equations. The scheme is based on the |$L^2$| norm least squares functional with weak imposition of continuity across interior faces. We minimize the functional over the piecewise polynomial spaces to seek numerical solutions. The method is shown to be stable without any constraint on the mesh size. We prove the optimal convergence rate under the energy norm and sub-optimal convergence rate under the |$L^2$| norm. Numerical results in two and three dimensions are presented to verify the error estimates. [ABSTRACT FROM AUTHOR]

Surface functionalization is an effective way to improve the activity of electrocatalysts by adjusting the adsorption and desorption capacity of intermediate species on noble metal surfaces. Herein, density functional theory (DFT) calculations demonstrate that cyanogroup functionalization can effectively reduce the d-band centre (εd) of the Pt (111) plane, thus resulting in a hydrogen evolution reaction (HER) performance improvement. Cyanogroup functionalized Pt ultrafine nanonetworks (CN–Pt UNs, sub-2 nm) are synthesized via a facile one-pot heating method with high yield and abundant defects/grain boundaries as the highly active sites. Owing to the optimized Pt electronic properties and ultrathin network nanostructure, the overpotentials of CN–Pt UNs at 10 mA cm−2 are merely 6 mV and 8 mV in 1 M HClO4 and 1 M KOH, respectively, and are superior to those of commercial 20 wt% Pt/C in a wide pH range. This work offers a new perspective towards the rational design of noble metal catalysts for boosting electrocatalysis by surface cyanogroup functionalization. [ABSTRACT FROM AUTHOR]

It is a difficult and long-term task to develop highly active and stable electrocatalysts to realize efficient hydrogen evolution reaction (HER) in electrochemical water splitting. Herein, we report a facile oil-bath method to fabricate hierarchical Rh2Ir alloy nanodendrites (Rh2Ir NDs) with excellent HER performance in a broad pH range. Due to the unique branched nanostructure and rational Rh2Ir metal proportion, the developed Rh2Ir NDs exhibit superior activity and enhanced stability towards HER in a broad range of pH conditions. To deliver a current density of 10 mA cm−2 in 1 M HClO4 and 1 M KOH, Rh2Ir NDs require overpotentials of only 12 and 9 mV, respectively. This work not only offers a high-performance electrocatalyst for electrochemical water splitting but also provides a new perspective towards the rational synthesis of hierarchical noble-metal nanodendrites. [ABSTRACT FROM AUTHOR]

In order to decrease the agglomeration of carbon nanotubes (CNTs) and enhance the interaction between CNTs and epoxy asphalt (EA), CNTs were modified by epoxy resin (ER) after treated with mixture acid, and then CNTs-EA was synthesized via mechanical stirring method. Experimental results confirmed that ER was decorated on the surface of CNTs. Moreover, the softening point and anti-ultraviolet aging of matrix asphalt were significantly improved by introducing modified CNTs, which increased about 48.7% for softening point and the variety ratio of UV aging (UAI) value was close to 1. The enhancement is mainly attributed to the excellent property of CNTs and the effective synergistic effect between CNTs and ER. [ABSTRACT FROM AUTHOR]

A series of MIL-101-SO3H/Nafion composite membranes was synthesized. They show an improved proton conductivity, due to the abundance of SO3H groups, which fosters proton conduction by binding the water molecules and enabling a larger number of conducting sites. Gas (including water vapor, hydrogen, and oxygen) permeability, crystallinity, and free volumes of the MIL-101-SO3H/Nafion composite membranes were investigated, as well as their correlation. By increasing the MIL-101-SO3H content, the gas permeability of the membranes significantly decreases, since the crystalline region is larger and the water-bearing MIL-101-SO3H particles are efficient barriers for the gas molecules. The gas permeation in the composite membranes is a very complex process and the results indicate no simple linear relation between the gas permeability and the free volume size (VFV), or between the gas permeability and the crystallinity. Moreover, it is very interesting to observe that the influence of VFV on the gas permeability is closely related to the size of the particular gas molecules: the larger the size of the gas molecules, the larger the free volume needed to achieve their rapid diffusion in the membrane. The results suggest the presence of a threshold value for VFV, which depends on the size of the gas molecules: when VFV is lower than this value, the gas molecules cannot easily jump through neighboring free volumes to a neighboring site, and, as a result, the permeability drops quickly. [ABSTRACT FROM AUTHOR]

Abstract: Nanoassemblies have attracted considerable attention because of their unique structural advantages, such as the high porosity, abundant surface defects, and low‐coordinated atomic steps, which are of importance to improve electrocatalytic activity and stability. However, developing effective and general synthetic methods for preparation of well‐defined Pd‐based nanoassemblies remains a challenge. Herein, a facile, effective, and general synthetic strategy for fabrication of high‐quality Pd3M (M = Co and Ni) nanoassemblies is developed. It is identified that arginine is pivotal for control of the nucleation and growth process of nanocrystals, where there is strong self‐assembly capacity between arginine molecules as well as effective coordination interaction between guanidine group of arginine with metal ions, endowing its unique dual‐function for the structure‐controllable synthesis of Pd3M alloy. The electrocatalytic performances of Pd3M nanoassemblies for the oxygen reduction reaction are studied through the rotating ring‐disk electrode technique. As expected, both Pd3Co and Pd3Ni nanoassemblies exhibit the more positive onset reduced‐overpotentials and the better catalytic stabilities compared to their counterparts. This work opens a new and general guideline for designing PdM alloy nanoassemblies with exclusive active sites for boosting energy electrocatalysis. [ABSTRACT FROM AUTHOR]

Perovskite BaLaFeO was synthesized via a glycine-nitrate process as the cathode material for intermediate-temperature solid oxide fuel cells. Ultra-fine pure phase BaLaFeO powder was obtained after the self-propagating combustion of a complex xerogel precursor. The BaLaFeO cathode showed good compatibility and an enhanced phase interface with a samarium-doped ceria electrolyte. Electrochemical measurements showed that the BaLaFeO cathode achieved a low polarization resistance of 0.58 Ω cm at 800 °C in an air atmosphere. The porous nanostructure guaranteed the catalytic activity for the oxygen reduction reaction, even if the cathode material possessed a cobalt-free and low rare earth composition. Graphical abstract: [ABSTRACT FROM AUTHOR]

Graphene/ZnO-Bi2O3nanoplates (GZB) were prepared by means of the low temperature hydrothermal method, and their photocatalytic activity for degradating rhodamine B (RhB) was studied. Experimental results show that the ZnO-Bi2O3nanoparticles were supported on graphene sheet under the assist of cetyltrimethylammonium bromide (CTAB). Furthermore, the resultant GZB nanoplates exhibited very strong adsorption capacity and high photocatalytic activity, in which the first absorbed ability comes up to about 74.62% within 30 min. Moreover, 90% RhB in solution can be degraded within 100 min under visible light after the sample reach the maximum adsorption. [ABSTRACT FROM PUBLISHER]