Search

Thermal management system of data center continuously face a lot of challenges, because data center industry has seen a boom growth in power density. In this paper we proposed the Tilted Active Tiles (TATs) to improve the local cold air supply and prevent the air flow blow over the rack. In traditional active tiles, fans are placed horizontally which cause the airflow blows over the rack, rather than into, the racks. To solve this issue, we adjusted the angle of the active tile to direct the airflow into the rack. We further introduced ANN based thermal models to predict the thermal performance of TATs. To train the ANN models, we adopted the data set obtained from a data center of Inner Mongolia Meteorology Information Center. The prediction accuracy of the model was extensively compared and analyzed, and the prediction accuracy and overhead of different neural network structures, i.e., BP and LSTM, were evaluated. Experimental results show that the rack with blanking panels has better thermal performance, and the temperature distribution at bottom, middle and top of the rack were same under smaller PWM. Thermal efficiency model was established by BP and LSTM, in this experiment single output model and multi output model were analyzed. The single output model can predict the temperature at different heights on the rack. In single output model the predicted effect of BP model is better than LSTM. The average prediction error is 0.57. The multi-output model can only predict the temperature at a fixed height of the rack. In multi output model LSTM model is better than BP. LSTM prediction error is less than BP. The average prediction error is 0.07.

This study deals with the mechanical behavior of a shear thickening fluid–sliding friction (STF–SF) isolator, and investigates the feasibility of its application in anti-seismic engineering from a theoretical perspective. The STF–SF isolator is composed of a conventional sliding friction (SF) material and a smart shear thickening fluid (STF) material. When the excitation energy is small, the SF material keeps the structure static or only has a small vibration. However, when the excitation energy is large, the STF material results a huge damping force and dissipation of the external energy, eventually reducing the relative motion between the base and the upper structure. To better understand this phenomenon, a mechanistic study of the STF–SF isolator is proceeded by an analytical model which is derived from flow momentum equations. After that, the seismic protective effect of the STF–SF isolator is evaluated at harmonic excitation with different accelerations through a single degree of freedom (SDOF) structure. Furthermore, the effects of layer gap, layer number and friction coefficient of the STF–SF isolator on the seismic protective effect are investigated. The results show that the STF–SF isolator can significantly decrease both the displacement response and the acceleration response of the SDOF structure. This theoretical study indicates the STF–SF isolator can be used in anti-seismic engineering.

Lightweight and thermal insulative cementitious materials with large porous structures for improving energy efficiency have rapid development currently. However, the design of cementitious materials for energy-efficient building system in terms of low thermal conductivity, excellent mechanical properties and lightweight remains a crucial challenge in construction & building fields. Recently, the polymer-based lattice-inspired structures fabricated by 3D printing technology (3DPT) show a rapid development that achieves a balance of thermal and mechanical properties due to their ordered and large-scale cellular architecture. In this work, a novel category of lattice-inspired cementitious lattice materials was first proposed for applying in the energy-efficient building systems by integrating the 3DPT and traditional casting technology. Taking face-centered cubic (FCC) lattice as an example, the advantages of the proposed fabrication technology and FCLMs are follows: (i) ultralight weight; (ii) enhanced ductility of brittle cementitious materials; (iii) low thermal conductivity. Through the structural and material design, the proposed fabrication technology could effectively overcome the current issues of 3DPT, such as the weakness of interlayer performance and low accuracy (Extrusion-based 3DPT) as well as limited feasible materials (Powder-bed-based 3DPT), in the civil engineering. Statement of Significance How to design cementitious materials for energy-efficient building system in terms of thermal conductivity, mechanical properties and lightweight requirement remains a crucial challenge in construction & building fields. Inspired by the low thermal conductivity and lightweight structure of metamaterials in the previous researches (Song et al., Materials & Design, 2019. 173: 107773-86; Nicholas et al. Nano letter, 2018. 18: 4755-61), here a novel category of lattice-inspired cementitious materials with ordered porous structures has been proposed for energy-efficient building materials with lightweight, low thermal conductivity and good mechanical properties characteristics. Additionally, recently there is increased interest to use 3D concrete printing technology (3DCPT) to automatically manufacture complicated and customized structure. However, currently, 3DCPT encounters many sticky issues, involving that (i) weak interlayer performance induced by cold joints between layers; (ii) deformation under self-weight due to the low yield stress of fresh mix; (iii) probable shrinkage under the low water/cement ratios typically used in printed mortars; (iv) geometrical instability owing to creating suspended structures. It is not a trivial to directly print cementitious porous materials, and therefore this work also achieved a breakthrough in developing cementitious porous materials combining advanced SLA 3D printing and traditional casting technology for the first time. It is worth mention that compared to Extrusion-based 3DCPT and Powder-bed-based 3DCPT, the advantages of the manufacture technology include that (i) automatically printing 3D complex structure; (ii) Optional cementitious materials; (iii) Good interlayer performances of prepared materials. We expect that the proposed design idea and novel manufacture strategy could serve as the guideline to develop material-structural-functional integrated energy-efficient building materials combining 3D printing technology, advanced structure design and high-performance cementitious materials.

Due to the complexity of crosswind vibrations of rectangular cylinders, current research on crosswind vibrations of rectangular cylinders mainly relies on expensive wind tunnel tests and time-consuming numerical simulation techniques. In this study, in order to evaluate crosswind vibrations of rectangular cylinders, machine learning method was used to build an efficient and effective prediction model for supplementing the above two research tools. 5 machine learning models based on decision tree regression, k-nearest neighbor regression, random forest, gradient boosting regression tree (GBRT) and histogram gradient boosting regression tree algorithms were trained based on the existing high-quality and reliable wind tunnel test datasets of crosswind responses of rectangular cylinders. The hyper-parameters were optimized by using particle swarm optimization method. 4 types of crosswind vibration phenomena, including over-coupled, coupled, semi-coupled and decoupled, were predicted. It was found that the GBRT model is capable of predicting crosswind responses of rectangular cylinders at side ratios from 0.75 to 3 and Scruton numbers from 0 to 150 under wind flow with turbulence intensities from 0 to 16%. Evidently, GBRT model can be an effective and economical method to study crosswind vibrations of rectangular cylinders and hence supplement traditional wind tunnel tests and numerical simulation techniques.

Inflow conditions play a key role in the Computational Fluid Dynamics (CFD) simulation of wind environment. Taking the micro wind climate of Hong Kong Kowloon Bay costal town as a research object, two kinds of widely used inflow condition determination methods are adopted to test their performances. One is to fit the velocity profile into the empirical (logarithmic/exponential) law, hereafter referred to as the Fitted Empirical Profile (FEP) method. The other is to interpolate the outflow velocities and turbulence properties from a pre-simulation of the upstream region, hereafter referred to as the Interpolated Multiscale Profile (IMP) method. The GIS data of this mountainous/ urban area are digitalized and simplified into the CFD geometry model. Computational treatments for numerical algorithms, domain size, grid systems and boundary conditions are carefully configured according to the published CFD Best Practice Guidelines (BPGs). By validating with one year of real scale wind measurement data from two meteorological stations, it is found that these two inflow conditions lead to considerably different results. Having less consideration for the blockage effects of terrain/buildings, the FEP method tends to predict higher wind speed. As more thermal effects are removed by increasing wind speed thresholds, the results of IMP method demonstrate an incremental agreement with the measurement data. Finally, the validated simulation results are applied to the spatial representativeness assessment of two meteorological stations. Both the point-to-surface consistency indicator and point-centered semivariance are employed. The results show that the meteorological stations show a good representation within a range of 400 m.

After 30 years of development, 3D printing technology (3DPT) is becoming increasingly popular in civil engineering field. However, weak interlayer performance and lack of reinforcement significantly limit its application. Nowadays, structural-functional integrated energy-efficient building materials, theoretically requiring a balance of mechanical properties and thermal resistance, are an emerging interdisciplinary research area in civil engineering field. This work proposed a novel digital cement-based manufacture technology by integrating 3D printing technology with casting technology. By incorporating tension-resistant polymeric truss network (PTN), a category of 3D face-centered cubic-inspired cement-based hybrid composites (FCHCs) is proposed. Results show the printed PTN can effectively improve the tension-resistant capacity of cement mortar. The FCHCs not only exhibit a multiple cracking damage characteristic but also a low thermal conductivity compared to traditional cement-based materials. A building energy simulation program was performed to evaluate the effectiveness of FCHCs on the energy savings. It is found that 34% heating energy could be saved for FCHCs building.

Atmospheric boundary layer (ABL) is generated due to frictions induced by obstacles on the Earth’s surface, such as terrains, vegetation, buildings, etc … Inspired by this physical mechanism, horizontally homogeneous ABL flow in computational fluid dynamics is achieved by proposing an enriched canopy drag (ECD) model, which is more flexible in reproducing the effects of various physical roughness ground, not limited to vegetation types. The multi-objective evolutionary algorithm is employed to determine the appropriate near-ground drag and the acting height of the proposed ECD model. Furthermore, the evolutionary algorithm is adopted to minimize the discrepancies of mean wind velocity and turbulence kinetic energy (TKE) between the inflow condition and the flow at a target downstream location. A set of simulations using logarithmic- and power-law inlet flows over rural, suburban, and urban terrains based on the shear-stress transport k-ω turbulence model are conducted to demonstrate the effectiveness of the ECD model. The results show that the proposed ECD model is capable of reproducing the Earth’s roughness numerically and improving the homogeneities of mean velocity and TKE in streamwise direction and applicable to different mesh resolutions, near-wall treatment methods, standard k-e turbulence model and simulation dimension.

In this paper, an Ytterbium-doped V-shape photonic crystal fiber (Yb-VPCF) with low dispersion and high nonlinearity is designed and fabricated in our laboratory. Through coupling femtosecond pulses into the fundamental mode of Yb-VPCF, the tunable anti-Stokes signals at the visible wavelength are efficiently generated based on the phase-matching four-wave mixing. When the pump wavelength is changed from 810, to 820, and to 830 nm and the input average power is increased from 0.4, to 0.5, and to 0.6 W, respectively, the anti-Stokes signals are generated within the wavelength range of 562–477 nm. The wavelength-tunable range is over 100 nm, and the maximum power ratio of anti-Stokes signal at 477 nm and the residual pump at 830 nm can be up to 23.9:1. The anti-Stokes signals generated can be used as the ultrashort pulse sources for ultrafast optoelectronics and spectroscopy.

This study evaluated the performance of wind energy harvesters with multiple shape-optimized circular cylinders in tandem via computational fluid dynamics simulations. The circular cylinders were optimized by attaching triangular protrusions on their surface. The circumferential location of the protrusion plays a crucial role in the efficiency of this kind of wind energy harvester. The protrusions at the circumferential angles of α = 60°and 90°significantly extend the wind velocity range with remarkable energy generation. When the reduced wind velocity is lower than 10, the harvester with three plain cylinders in tandem generates the most power. However, when the speed is higher than 10, the most power is generated by the harvester having three cylinders in tandem with protrusions at α = 60°. Therefore, in a low wind velocity environment, the harvester with three plain circular cylinders in tandem is superior to other tested configurations, whereas in a high wind velocity environment, the harvester with three circular cylinders with protrusions at α = 60°in tandem outstands from other tested configurations. The associated flow mechanisms are detailed as well.

Optimization method has been widely acknowledged as an effective approach to design engineering structures, and yet few studies adopt this method to design cellular materials. Here, we firstly adopted a Kriging assisted Multi-objective Genetic Algorithm to guideline the design of octet-truss (OCT) cellular materials with the maximum specific modulus. Subsequently, additional struts were artificially introduced into the optimized OCT to further mechanically reinforce performances. All the cellular materials were precisely fabricated using a Stereolithography 3D printing technique. By reasonably optimizing the sizes of OCT, the optimized OCT with a 1.112 mm diameter and 8.282 mm cell length was achieved, which displays a superior modulus-to-mass ratio. The highest modulus and strength of as-designed cellular materials achieved 34.12 MPa and 2.64 MPa, reinforced by ~3.11 and 4.81 times, respectively. Additionally, the absorbed energy efficiencies of them improved from 74.75% to 90.80%, which are significantly higher than other cellular materials. By in situ tests and fracture analyses, the high recoverability is attributed to the comprehensive effect of net-shaped architecture and elastic-plastic deformation. Keywords: Cellular materials, Optimization design, 3D printing technique, In situ tests, Deformation mechanism, Recoverability

Wearable triboelectric nanogenerators (TENGs) have attracted interest in recent years, which demand highly flexible, scalable, and low-cost features. Here, we report an ultra-flexible, large-scale and textile-based TENG (T-TENG) for scavenging human motion energy. The triboelectric layer was derived from the polydimethylsiloxane (PDMS) film with a cost-effective paper-induced rough surface via a facile doctor-blending technology. Ag-coated chinlon fabric (ACF) with ultra-flexible, large-scale and conductive characteristics was used as the electrode. The as-fabricated PDMS-based ACF (PACF) composites possess a 240 &times, 300 mm2 superficial area and remain highly flexible under mechanical squeezing, folding and even tearing deformation. The maximum output charge of ~21 &mu, C and voltage of 80.40 V were therefore achieved to directly power 100 LEDs based on the high surface area of 762.73 mm2 which was rationally replicated from the sandpaper of the T-TENG. Moreover, the output voltage signal can be also used as a trigger signal of a movement sensor. Importantly, the explicit theoretical model corresponding to T-TENG was quantitatively investigated under different applied force, frequency and effective surface factor.

The ultraviolet to visible continuum generations are experimentally demonstrated by coupling femtosecond pump pulses around 800 nm into the normal dispersion region close to the zero-dispersion wavelength in the fundamental mode of a silica photonic crystal fiber (PCF). The ultraviolet to visible continuum is generated through a combination of nonlinear effects including degenerate four-wave mixing, cross-phase modulation, and stimulated Raman scattering. When pump pulses with the wavelength of 800 nm and the input average power of 500 mW are used, more than 40% of the incident pump power is converted to the ultraviolet to visible spectral region. The evolution of the output spectra for different PCF lengths is studied to understand the nonlinear propagation dynamics during the process of the continuum generation. The ultraviolet to visible continuum generation will find important applications in the ultraviolet and visible-based biophotonics and spectroscopy.

In this paper, a dual-core polymer-filled photonic crystal fiber (PCF) is designed for broadband dispersion compensation. Large negative dispersion value from −255 to −823 ps/(nm•km) over the entire C band is achieved. Furthermore, the effective refractive index of the polymer-filled outer core can be well tuned by changing the temperature. This PCF can be used for the dynamical dispersion compensation in wavelength-division-multiplexing (WDM) systems.

A technical bottleneck to the broadening of applications of glass nanofluidic chips is bonding, due to the strict conditions, especially the extremely high temperatures (~1000 °C) and the high vacuum required in the current glass-to-glass fusion bonding method. Herein, we report a strong, nanostructure-friendly, and high pressure-resistant bonding method, performed at room temperature (RT, ~25 °C) for glass nanofluidic chips, using a one-step surface activation process with an O(2)/CF(4) gas mixture plasma treatment. The developed RT bonding method is believed to be able to conquer the technical bottleneck in bonding in nanofluidic fields.

A highly birefringent and nonlinear photonic crystal fiber with a V-shape simplified structure is proposed and analyzed by the fully vectorial finite element method. It will have important applications in the nonlinear optics and biophotonics.