Search

The adaptive neurofuzzy inference system (ANFIS) is a structured multioutput learning machine that has been successfully adopted in learning problems without noise or outliers. However, it does not work well for learning problems with noise or outliers. High-accuracy real-time forecasting of traffic flow is extremely difficult due to the effect of noise or outliers from complex traffic conditions. In this study, a novel probabilistic learning system, probabilistic regularized extreme learning machine combined with ANFIS (probabilistic R-ELANFIS), is proposed to capture the correlations among traffic flow data and, thereby, improve the accuracy of traffic flow forecasting. The new learning system adopts a fantastic objective function that minimizes both the mean and the variance of the model bias. The results from an experiment based on real-world traffic flow data showed that, compared with some kernel-based approaches, neural network approaches, and conventional ANFIS learning systems, the proposed probabilistic R-ELANFIS achieves competitive performance in terms of forecasting ability and generalizability.

This study paper takes the Red Revolution architecture in Geyuan Town as an example. By discussing the value of Red Revolution architecture, so as to better recognize the current situation and problems of Red Revolution architecture in Jiyuan Town. Finally, it put forward the suggestions on the protection and inheritance of the Red Revolution architecture.

To enhance microwave loss abilities, constructing composites with one-dimensional (1D) structure is an excellent scheme. In this work, a high-efficiency microwave absorber of MnO nanograins decorated vanadium nitride/carbon nanofibers (MnO-VN/C NFs) was successfully prepared for the first time via co-electrospinning technology and subsequent nitriding treatment. Studying in detail the specific relationship between nitriding time and the morphology of the as-prepared NFs, the precipitations of MnO nanoparticles with tailored structures were attached on the surface of VN/C NFs to optimize their electromagnetic parameters. When the nitriding time was 2.0 h at 600 °C, the MnO-VN/C NFs displayed good microwave absorption performances: the minimum reflection loss (RL) value was -63.2 dB at 8.8 GHz, and the bandwidth of RL-10 dB was up to 6.4 GHz from 11.6 to 18 GHz at the thickness of 2.8 mm. Meanwhile, the absorption bandwidth (RL-10 dB) could cover the whole X and Ku band by adjusting the thickness, respectively. The outstanding performances could be attributed to the good impedance matching and various loss pathways including conductive loss and interfacial and dipole polarizations. In these regards, MnO-VN/C NFs are likely to be utilized as a high-efficiency microwave absorber. And the strategy in this work can provide great help to design other 1D structural microwave absorbers with a broader absorbing band.

Clustering by fast search and find of Density Peaks (DPC) has the advantages of being simple, efficient, and capable of detecting arbitrary shapes, etc. However, there are still some shortcomings: 1) the cutoff distance is specified in advance, and the selection of local density formula will affect the final clustering effect; 2) after the cluster centers are found, the assignment strategy of the remaining points may produce “Domino effect”, that is, once a point is misallocated, more points may be misallocated subsequently. To overcome these shortcomings, we propose a density peaks clustering algorithm based on natural nearest neighbor and multi-cluster mergers. In this algorithm, a weighted local density calculation method is designed by the natural nearest neighbor, which avoids the selection of cutoff distance and the selection of the local density formula. This algorithm uses a new two-stage assignment strategy to assign the remaining points to the most suitable clusters, thus reducing assignment errors. The experiment was carried out on some artificial and real-world datasets. The experimental results show that the clustering effect of this algorithm is better than those other related algorithms.

Spectral proper orthogonal decomposition (SPOD) is performed to identify the spatiotemporal coherent structures of trailing edge cutback film cooling flows. The analyzed data obtained from delayed-detached eddy simulation are validated against the experimental data first, followed by the extraction of SPOD modes and the reconstruction of flow fields using the SPOD modes. Results show that the flow topology of trailing edge cutback film cooling flows is mainly determined by the blowing ratio. At low blowing ratios, the vortex shedding phenomenon is captured by the first leading SPOD mode. Its mode shape corresponds to the bluff-body wake formation of the lip featured by spanwise coherent structures, which enhance the coolant–hot gas mixing process. At high blowing ratios, the vortex shedding phenomenon is represented by both the first and the second leading SPOD modes. In addition to the lip wake mode, the mode of the bluff-body wake formation of the coolant channel rib arises, whose mode shape is featured by transverse coherent structures. These structures enhance the mixing process between the coolant flows. As the blowing ratio increases, the dominating mode gradually transits from the lip wake mode to the rib wake mode. From the SPOD reconstruction study, the deterministic unsteadiness of vortex shedding is found to contribute to a substantial proportion of unsteady kinetic energy, especially at low blowing ratios. These findings help advance the understanding of trailing edge cutback film cooling flow.

Film cooling effect on trailing edge cutback with land extensions (i.e., landed case) was numerically investigated with the delayed-detached eddy simulation (DDES) method. At three rib geometries (i.e., inline rib arrays, six-row pin-fin arrays, and five-row pin-fin arrays) and five blowing ratios (i.e., M = 0.5, 0.8, 1.1, 1.5, and 2.0), film cooing effectiveness, coherent vortex structures, and discharge coefficients for the landed cases were analyzed and compared with baseline cases (i.e., cutback without land extensions). The results show that land extensions have significant influences on coherent flow structures, vortex energy levels, film cooling effectiveness, and discharge coefficients in cutback region. Different from the baseline cases, the dominant vortex structures in landed cases exhibit the “double helix” (for cutback with inline rib array) or “strip” pattern (for cutback with pin-fin arrays), and the thickness of mixing region in landed cases is decreased. Among three rib geometries, the trailing edge cutback with six-row pin-fin arrays has the worst cooling effect for both baseline and landed cases. Compared with the baseline cases, the discharge coefficients for the landed cases are decreased by about 21.4%. With land extensions, the overall film cooling effectiveness on cutback is decreased first and then increased with increasing blowing ratio. Among all investigated cases, cutback with five-row pin-fin arrays for the landed case performs the best film cooling effect at M = 0.5.

As gas turbine inlet temperature continuously increases, blade trailing edge suffers an extremely high thermal load due to the thin structure and constraint of internal convective cooling arrangements. To overcome these difficulties, pressure-side cutback, which is strengthened with multiple internal structures and land extensions, is widely used in blades to protect trailing edge from high thermal stresses. Due to the geometrical complexity and strong interactions between coolant flow and mainstream, sophisticated heat transfer and flow patterns exist in the cutback region, which presents a great challenge for the trailing edge cutback design. To understand the heat transfer and aerodynamic performance in blade with trailing edge cutback, CFD method has become an efficient tool which provides deep insights into the flow mechanisms and heat transfer characteristics in the detailed region. To accurately resolve the flow and heat transfer performance in a turbine blade with trailing edge cutback, structured grids are preferred because of higher resolution in flow/heat transfer prediction than unstructured grids, especially in boundary layers. However, for a blade with landed trailing edge cutback, few researchers tried to employ structured grids to predict aero-thermal performance due to the geometrical complexity. In this paper, the Background-Grid Based Mapping (BGBM) method proposed in Part I of this study was adopted to generate multi-block structured grids for a gas turbine vane with landed trailing edge cutback. With the coordinate transformation strategies, multi-block structured grids for the vane with landed trailing edge cutback were generated conveniently. With the generated structured grids, flow and heat transfer performance in vane were investigated using RANS (Reynolds-Averaged Navier-Stokes) equations solutions combined with transitional turbulence model. Effects of land extensions on the heat transfer and aerodynamic performance were analyzed, as well as the effects of inflow turbulence intensity, mainstream Reynolds number and ejection rate. The results show that heat transfer coefficients on vane surface, total pressure loss coefficient and energy loss coefficient in vane are all increased with the increase of inflow turbulence intensity. However, heat transfer coefficients on cutback and trailing edge surface are not sensitive to inflow turbulence intensity. At the same inflow turbulence intensity, the aerodynamic loss in vane is decreased with increasing the Reynolds number of mainstream. The increase of ejection rate significantly increases the heat transfer coefficients on cutback surface. Compared with the vane without land extensions, heat transfer coefficients and pressure coefficients on vane surface are reduced and the heat transfer coefficients on cutback surface are increased for the vane with land extensions. In the case of Re = 2.0 × 106, the area-averaged heat transfer coefficient on landed cutback is 14.46% higher than the cutback without lands. Compared with the experimental data, predictions with structured grids based on BGBM method are more agreeable than those with unstructured grids.

Mesoporous alumina modified by silica (MA-S) with good thermal stability was prepared by a self-assembly method. The morphology and microstructure of as-prepared samples were characterized by transmission electron microscopy (TEM), X-ray diffraction (XRD), and so on. Though adding 5 wt% SiO2 in Al2O3 framework, the MA-S exhibited high thermal stability and large specific surface area (87 m2 g−1) at 1200 °C. The results indicated that SiO2 can suppress the phase transformation from γ-Al2O3 to α-Al2O3 and postpone the crystallization temperature at same time. Therefore, the as-prepared materials possess a great potential for application at high temperature.

CO2-monitoring plays an important role in medicine, environmental sciences, and food industries. Here, a new CO2-responsive hydrogel film was fabricated from branched polyethyleneimine (BPEI) and partially oxidized dextran (PO-Dex) via layer-by-layer (LBL) assembly, based on the in situ Schiff base reaction between the two polymers. The swelling behaviours of the films were studied using Fabry-Perot fringes on their reflection spectra. Like ordinary hydrogels, the BPEI/PO-Dex films swell in water. In addition, they swell to a larger degree when CO2 is introduced, due to the reaction between CO2 and the amino groups in BPEI. The CO2-induced swelling can be reported by the shift of the Fabry-Perot fringes on their reflection spectra; therefore, the BPEI/PO-Dex film can be used as an optical sensor for dissolved CO2. In the new sensor, the BPEI/PO-Dex film acts as a CO2-sensing material and Fabry-Perot cavity simultaneously. The introduction of ordered structure is no longer required. The response of the sensor to CO2 is linear and reversible. Unlike other hydrogel-based sensors that suffer from a slow response, the new sensor can respond to CO2 quickly, making it applicable for real-time, continuous monitoring of CO2 levels in solution.

Deep learning (DL) techniques have been widely used in recommender systems for user modeling and matching function learning based on historical interaction matrix. However, existing DL-based recommendation methods usually perform static user preference modeling by using historical interacted items of the user. In this article, we present a time-aware deep CF framework which contains two stages: dynamic user preference modeling based on attention mechanism and matching score prediction based on DL. In the first stage, short-term user preferences are modeled by the time-aware attention mechanism that fully considered the predicted item, the recent interacted items and their interaction time. The resulting short-term preferences are combined with long-term preferences for dynamic user preference modeling. In the second stage, high-order user-item feature interactions are learned by two types of DL models, Deep Matrix Factorization (DMF) and Multiple-Layer Perception (MLP), and the feature interaction vectors of the two models are fused in the last layer of the model to predict the matching score. Extensive experiments on five datasets indicate that our method is superior to the existing time-aware and DL-based recommendation methods in top-k recommendations significantly and consistently.

To cool a high-pressure gas turbine blade, many rows of cooling holes with different arrangements and configurations are manufactured to achieve higher cooling effect and lower aerodynamic loss. To evaluate the heat transfer and film cooling effect in the full-cooled turbine blade, efficient numerical simulations are required in the design and performance optimization processes. From the view of numerical accuracy, the structured grids have to be employed because of higher resolution in flow and heat transfer than the unstructured grids. Because many splitting, attaching and merging manipulations are involved in meshing the cooling features and curved boundaries, it is very complex and time-consuming for a researcher to generate multi-block structured grids for a full-cooled gas turbine blade. As a result, in the industrial applications, almost all researchers preferred to generate unstructured grids instead of structured grids for the full-cooled blade. Unlike the previous research, the aim of this study is to apply the Background-Grid Based Mapping (BGBM) method proposed in Part I to generate multi-block structured grids for a full-cooled gas turbine vane. With the strategy of BGBM method, meshes were conveniently generated in the computational space with simple geometrical features and plain interfaces, and then were mapped back into physical space to obtain the multi-block structured grids which can be used for numerical simulations. With the experimental data, the present numerical methods and BGBM strategy were carefully validated. Then, the flow and film cooling performance in the full-cooled NASA GE-E3 nozzle guided vane were numerically investigated. The effects of coolant mass flow rate and land extensions on film cooling effectiveness were discussed. The results show that film cooling effectiveness near the stagnation point is the lowest and film cooling effectiveness on the pressure side is slightly higher than that on the suction side. When the coolant mass flow rate increases up to the value of 1.5 design flow, the relative outflow mass flow rates of cooling hole arrays and slots are no longer affected by the increase of the coolant flow rate. At half design flow, the outflow mass flow rates of No.5 hole-array to No.10 hole-array are almost zero, and the area-averaged film cooling effectiveness on vane surface is as low as 0.268. Compared with the cases of half design flow and double design flow, better film cooling performance is obtained in the cases of design flow and 1.5 design flow. Compared with the vane without lands, the area-average cooling effectiveness on vane surface is slightly higher for the vane with lands. Land extensions have a considerable influence on film cooling performance in the cutback region.

Film cooling technique is commonly adopted in modern gas turbine engines to protect high-temperature components from erosion and damage caused by thermal stress. To improve film cooling effectiveness, many efficient prediction tools have been developed and have shown promising results, which are helpful for turbine aero-thermal design. For film cooling, evidence has shown that it is strongly affected by the momentum and heat transport in the boundary layer when hot gas and coolant are mixed downstream of the ejection. From the view of resolution accuracy in the boundary layer, structured grids will be the primary choice in fluid domain. However, the high-pressure gas turbine blades usually have several hundreds of cooling holes with different configurations and arrangements. Numerical simulations often face a big challenge in multi-block structured-grid generations when a large number of cooling holes are involved on curved hole-to-mainstream interfaces. Conventional block-splitting and mesh-generation for all holes are quite time-consuming and cumbersome, because the copying, translating and rotating manipulations cannot be applied on curved hole-to-mainstream interfaces directly. To solve these difficulties, this paper presents a novel mesh-generation strategy, which is a background-grid based mapping (BGBM) method, to generate multi-block structured grids for film-cooled blade efficiently without modifying the existing meshing tools and solvers, which is convenient for CFD users. It consists of three main steps: At first, the correspondence between physical space and computational space is established by two sets of background grids. Then, the sectional curves of geometry features in physical space are projected to the computational space. With these treatments, the curved hole-to-mainstream interfaces are flattened in computational space, where grids can be quickly generated with block copying, translating, rotating and merging manipulations. Thereafter, meshes in computational space are mapped back to the physical space based on the correspondence between physical and computational spaces, and high-quality structured-meshes can be obtained for numerical simulations. To demonstrate the presented meshing strategy, several typical cases with film cooling are selected for testing, including single cooling hole on curved surface, multiple rows of cooling holes on curved surface and NASA C3X vane with multiple hole arrays. In these cases, different holes, including the cylindrical holes and shaped holes with different ejection angles and arrangements, on curved interfaces are taken into consideration. The quality of generated structured grids for each test case is illustrated, which is able to meet the requirement of CFD solver. With the generated meshes, conjugate heat transfer performance in the turbine vane with different cooling arrangements is investigated and also validated with the existing experimental data.

Hardening of cosmic proton- and nuclei-spectra is explained by using the gluon condensation (GC) model, which states that a large amount of gluons in proton may condense near the high energy threshold. The results present the GC-effect as common origin of a series of anomalous astrophysical phenomena including the broken power-law in gamma-ray spectra, the excess in positron- and electron-spectra and hardening of proton- and nuclei-spectra., 25 pages, 9 figures; accepted for publication in JCAP

Modern data centers coordinate hundreds of thousands of heterogeneous tasks aiming at providing highly reliable cloud computing services. Failure prediction is of vital importance in the analysis of cloud reliability. Recently, a novel kernel learning method called relevance vector machine (RVM) has been widely applied to solve nonlinear predicting problems and has been verified to perform well in many situations. However, it remains a great challenge for existing approaches to acquire the optimal RVM parameters. In this research, an artificial immune system is introduced into a Quantum-inspired Binary Gravitational Search Algorithm (QBGSA) in order to improve the convergence rate of standard QBGSA. In addition, a hybrid model of RVM with improved QBGSA called IQBGSA-RVM is proposed that aims to predict the failure time of cloud services. To evaluate the effectiveness of IQBGSA-RVM in failure prediction, its predicting performance is compared with that of the following algorithms, all of which employs RVM: chaotic genetic algorithms, binary gravitational search algorithms, binary particle swarm optimization, quantum-inspired binary particle swarm optimization and standard QBGSA. The experimental results show that the IQBGSA-RVM model is either comparable to the other models or it outperforms them, to say the least.

This paper presents a novel reconfigurable spherical (rS) joint resulting from the line dependence of screws and their embodiment geometry. Subsequently, the structure and phase change of the rS joint are elaborated. With this new invention, two novel metamorphic mechanisms, that are, the Schatz-derived metamorphic mechanism and the R(rS)R-R(rS)R metamorphic mechanism, are obtained by inserting the rS joint. The mechanisms entail a reconfiguration process that occurs in two manners: the phase change of the rS joint and the bifurcation of the mechanism. The Schatz-derived metamorphic mechanism entails reconfiguration by phase change of the rS joint and the bifurcation, the R(rS)R-R(rS)R metamorphic mechanism reconfigures the configuration based on phase change of the two rS joints. Remarkably, the former realizes the switch between two assembly configurations without resorting to disassembly and reassembly, and maintains the characteristics of the original Schatz mechanism while the rS joint situates in phase R. The latter changes the mobility and bridges gap between a spatial single-loop mechanism and a structure. With this new reconfigurable joint, various reconfigurable mechanisms could be generated by inserting the reconfigurable joint into the classical ones.

Unsaturated metal species (UMS) confined in nanomaterials play important roles for electron transfer in a wide range of catalytic reactions. However, the limited fabrication methods of UMS restrict their wider catalytic applications. Here, we report on the synergy of unsaturated Zn and Cu dopants confined in carbon dots (ZnCu-CDs) to produce enhanced electron transfer and photooxidation processes in the doped CDs. The Zn/Cu species chelate with the carbon matrix mainly through Cu–O(N)–Zn–O(N)–Cu complexes. Within this structure, Cu2+ acts as a mild oxidizer that facilely increases the unsaturated Zn content and also precisely tunes the unsaturated Zn valence state to Znd+, where d is between 1 and 2, instead of Zn0. With the help of UMS, electron-transfer pathways are produced, enhancing both the electron-donating (7.0 times) and -accepting (5.3 times) abilities relative to conventional CDs. Because of these synergistic effects, the photocatalytic efficiency of CDs in photooxidation reactions is shown to ...

In this paper, we propose a deep collaborative filtering recommendation model, which consists of an attention-based representation learning component and a multi-input matching function learning component. This model takes interaction matrix based on implicit feedback as data source to construct representations of long-term user preferences and item latent features. In the representation learning, a time-aware attention network is used, which uses the embedding vectors of the predicted item, recent historical interaction items, and the interaction time of recent historical interaction items to estimate the weights of different historical interaction items to short-term user preferences modeling. Then, the dynamic user preference representation can be obtained by combining short-term preferences with long-term preferences. In the matching function learning, a multi-input deep learning model is used. Its input includes not only the dynamic user preference representation and the item latent feature representation, but also the linear interaction between the two representations, so that the model has more powerful feature interactions learning ability. Experimental results on four datasets from different domains show that our method is largely superior to the state-of-the-art collaborative filtering methods, and the novel techniques we propose in this paper are effective in improving recommendation performance.

Herein, N-doped carbon coated vanadium nitride nanowires (VN@NC NWs) with inner porous structure were synthesized by coating a thin layer of a carbon precursor of polydopamine (PDA) on ultralong V2O5 NWs followed by nitridation. The amorphous carbon coating could ensure the one-dimensional (1D) structure, as well as produce the inner pores during nitriding progress. The formational 1D porous architecture, high-crystallized VN nanoparticles and homogeneous N-doped carbon coating synergistically improved the materials' impedance match and thus enhanced their microwave absorption performances. The minimum reflection loss (RLmin) of the optimal samples/paraffin hybrid with 20 wt% was −60.2 dB with an effective absorption bandwidth (EAB, RL

Origami engineering as an interdisciplinary subject brings thriving progress of mechanisms innovation. In this paper, a construction approach from origami to multiple spherical-integrated mechanisms is proposed and subtly sets up the bridge between origami and Bennett linkage, Bricard linkage. Subsequently, the attention is drawn to a special origami-inspired integrated 8R kinematotropic metamorphic mechanism. The paper applies screw theory to illustrating the reconfigurability of singular configurations with distinct parametric constraints in the 8R mechanism. Two overconstrained linkages, a 4R linkage and a 6R linkage, are evolved under peculiar geometrical constraints respectively. Furthermore, the kaleidocycle and deployable properties of the 8R mechanism are revealed. The paper hence not only presents an interesting work extracted from origami but also sheds light on the promising investigation about the intersection of distinct types of reconfigurable mechanisms. Finally, by using the kinematotropic metamorphic 8R mechanism as the reconfigurable trunk, a novel quadruped robot is built and its characteristics will be investigated in future research.

We studied the effects of centrality fluctuation and deuteron formation on the cumulant ( ) and correlation functions ( ) of protons up to the sixth order in the most central ( < 3 fm) Au+Au collisions at = 3 GeV in a microscopic transport model (JAM). The results are presented as a function of rapidity acceptance within the transverse momentum 0.4 < p T < 2 GeV/ c. We compared the results obtained by the centrality bin width correction (CBWC) using charged reference particle multiplicities with the CBWC using impact parameter bins. It was found that, at low energies, the centrality resolution for determining the collision centrality using charged particle multiplicities is not sufficient to reduce the initial volume fluctuation effect for higher-order cumulant analysis. New methods need to be developed to classify events with high centrality resolution for heavy-ion collisions at low energies. Finally, we observed that the formation of deuterons suppresses the higher-order cumulants and correlation functions of protons and found it to be similar to the efficiency effect. This work can serve as a noncritical baseline for the QCD critical point search in the high baryon density region.

The three-dimensional graphene microspheres with well-defined shape and size were prepared successfully by a novel approach, i.e., a combination of microfluidic device and freeze-drying. In the synthetic process, a pragmatic and facile device has been designed for the controllable generation of uniform-sized droplets. Followed by freeze-drying, highly porous spherical graphene microballs were obtained for further research. It is worth noting that the size of the droplets can also be controlled by adjusting the flow rate ratio of each phase. Afterward, the morphology, structure and character of the three-dimensional graphene particles are characterized by scanning electron microscope, X-ray diffraction, Fourier transform infrared spectroscopy and X-ray photoelectron spectroscopy. Due to their highly porous structure, the as-prepared 3D graphene microspheres were employed as adsorbent. The adsorption capacity of acid orange 7 and Pb(II) was about 71.9 and 147.7 mg g−1, respectively. The results showed that the porous 3D graphene microspheres as adsorbent possessed excellent adsorption ability for organic and inorganic contaminants in this work, showing their potential applications in environmental pollutant treatment.