Scaling Rules Research Articles

Blade Design and Aerodynamic Performance Analysis of a 20 MW Wind Turbine for LCoE Reduction

The aim of this study is to develop a blade mass model that incorporates a low-induction rotor (LIR) and a low-specific power concept to reduce aerodynamic loads and lower the Levelized Cost of Energy (LCoE). This blade mass model replaces the traditional simple scaling rule and incorporates the concept of LCoE reduction, presenting not only the mass distribution in the blade span direction but also the stiffness distribution. In order to achieve the desired reduction in LCoE, we developed a mathematical model that expresses blade mass as a function of the axial induction factor, which influences the aerodynamic load on the blade. We used this model to determine geometries of various low-induction rotors for 20 MW class horizontal axis wind turbine, and to identify the axial induction factor that correlates with the lowest blade mass. The chord length and twist angle in the spanwise direction of the blade were determined using PROPID’s reverse design process, based on the specified axial induction factor. Since the low-induction concept is not aerodynamically optimal, a low-specific power design approach was also adopted. This involved increasing the blade length and shifting the power curve to the left. By doing so, the AEP is increased, directly contributing to a reduction in the LCoE. Mass per unit length of the blade was presented, reflecting the distribution of airfoil type, blade geometry, and shapes of internal structures such as spars and webs.

CowClip: Reducing CTR Prediction Model Training Time from 12 Hours to 10 Minutes on 1 GPU

The click-through rate (CTR) prediction task is to predict whether a user will click on the recommended item. As mind-boggling amounts of data are produced online daily, accelerating CTR prediction model training is critical to ensuring an up-to-date model and reducing the training cost. One approach to increase the training speed is to apply large batch training. However, as shown in computer vision and natural language processing tasks, training with a large batch easily suffers from the loss of accuracy. Our experiments show that previous scaling rules fail in the training of CTR prediction neural networks. To tackle this problem, we first theoretically show that different frequencies of ids make it challenging to scale hyperparameters when scaling the batch size. To stabilize the training process in a large batch size setting, we develop the adaptive Column-wise Clipping (CowClip). It enables an easy and effective scaling rule for the embeddings, which keeps the learning rate unchanged and scales the L2 loss. We conduct extensive experiments with four CTR prediction networks on two real-world datasets and successfully scaled 128 times the original batch size without accuracy loss. In particular, for CTR prediction model DeepFM training on the Criteo dataset, our optimization framework enlarges the batch size from 1K to 128K with over 0.1% AUC improvement and reduces training time from 12 hours to 10 minutes on a single V100 GPU. Our code locates at github.com/bytedance/LargeBatchCTR.

Individual variation in field metabolic rates of wild living fish have phenotypic and ontogenetic underpinnings: insights from stable isotope compositions of otoliths

IntroductionIndividual metabolism has been identified as a key variable for predicting responses of individuals and populations to climate change, particularly for aquatic ectotherms such as fishes. Predictions of organism standard metabolic rate (SMR), and the thermal sensitivity of metabolic rate are typically based on allometric scaling rules and respirometry-based measures of respiratory potential under laboratory conditions. The relevance of laboratory-based measurement and theoretical allometric rules to predict performance of free-ranging animals in complex natural settings has been questioned, but determining time averaged metabolic rate in wild aquatic animals is challenging.MethodsHere we draw on stable isotope compositions of aragonite in fish otoliths to estimate time averaged experienced temperature and expressed field metabolic rate (FMR) simultaneously and retrospectively at an individual level. We apply the otolith FMR proxy to a population of European plaice (Pleuronectes platessa) from the North Sea during a period of rapid warming between the 1980s to the mid-2000s, sampling otolith tissue grown in both juvenile and adult stages.ResultsAmong-individual variations in realized mass-specific FMR were large and independent of temperature and scaled positively with body size in adult life stages, contradicting simplistic assumptions that FMR follows scaling relationships inferred for standard metabolic rates (SMR). In the same individuals, FMR in the first summer of life co-varied positively with temperature.DiscussionWe find strong evidence for the presence of consistent metabolic phenotypes within the sampled population, as FMR in the first year of life was the strongest single predictor for among individual variation in FMR at the point of sampling. Nonetheless, best fitting models explained only 20% of the observed variation, pointing to large among-individual variation in FMR that is unexplained by body mass, temperature or metabolic phenotype. Stable isotope-derived estimates of field metabolic rate have great potential to expand our understanding of ecophysiology in general and especially mechanisms underpinning the relationships between animal performance and changing environmental and ecological conditions.

Dual-band polarization-independent high quality factor Fano resonances using a twisted tetrameric nanohole slab

The Fano profile inspired by bound states in the continuum (BICs) has emerged as an effective approach to obtain high quality factor (QF) resonances. However, achieving polarization-independent high QF Fano resonances through the excitation of BICs is still challenging. Herein, we demonstrate that dual-band polarization-independent high QF Fano resonances can be realized by using a twisted tetrameric nanohole slab (TTNS). By twisting the nanoholes of the tetramerized slab, two quasi-BICs can be transformed into dual-band high QF Fano resonances due to the Brillouin zone folding as well as the symmetry breaking from C4v to C2v of the structure. The variation of the twist angle (θ) significantly alters the location of the Fano resonance in the longer wavelength, but it has slight influence on the Fano resonance in the shorter wavelength. A larger QF corresponds to a larger average electric-field enhancement-factor (AEE) for both of the Fano resonances, but the QF of the Fano resonance at the longer wavelength is more robust to the variation of θ due to its larger scaling rule as QF∼|θ|−6. According to multipole decompositions, Fano resonances at the longer and shorter wavelength correspond to toroidal dipole (TD) mode and hybrid TD-magnetic quadrupole (MQ) mode, respectively. In addition, both the two Fano resonances survive even if the structural parameters are significantly altered, and they exhibit polarization-independent features because the rotational symmetry of the structure can be maintained as θ is varied.

Scaling of biological rates with body size as a backbone in the assembly of metacommunity biodiversity.

The dispersal-body mass association has been highlighted as a main determinant of biodiversity patterns in metacommunities. However, less attention has been devoted to other well-recognized determinants of metacommunity diversity: the scaling in density and regional richness with body size. Among active dispersers, the increase in movement with body size may enhance local richness and decrease β-diversity. Nevertheless, the reduction of population size and regional richness with body mass may determine a negative diversity-body size association. Consequently, metacommunity assembly probably emerges from a balance between the effect of these scalings. We formalize this hypothesis by relating the exponents of size-scaling rules with simulated trends in α-, β- and γ-diversity with body size. Our results highlight that the diversity-body size relationship in metacommunities may be driven by the combined effect of different scaling rules. Given their ubiquity in most terrestrial and aquatic biotas, these scaling rules may represent the basic determinants-backbone-of biodiversity, over which other mechanisms operate determining metacommunity assembly. Further studies are needed, aimed at explaining biodiversity patterns from functional relationships between biological rates and body size, as well as their association with environmental conditions and species interactions.

Regret theory-based multivariate fusion prediction system and its application to interest rate estimation in multi-scale information systems

Estimating interest rates is a typical multivariate prediction problem that has garnered considerable attention in the finance industry. However, the rising complexity of the prediction environment has presented significant obstacles to the methodology and application of traditional single-scale rational prediction models. Considering this challenge, the paper presents a new multivariate fusion prediction system that aims to tackle the following issues: (1) How to mine prediction information from multi-scale information systems (MSISs)? (2) How to describe the irrational behavior of decision-makers (DMs) in prediction processes? (3) How to improve the prediction performance and generalization capability of the models? Specifically, the score function of each feature under MSISs is primarily established by applying scale rules, regret theory (RT), and dominance relations to thoroughly explore prediction information across varying scales. Second, an adaptive S3WD model is developed for feature selection, which leverages the feature score distribution and the sequential three-way decision (S3WD) concept to avoid the problem of dimensional disasters, thereby improving the accuracy of selected features and reducing the learning cost. Furthermore, to enhance the effectiveness and rationality of the kernel extreme learning machine (KELM), an improved version known as IKELM is developed, which combines the regret-based comprehensive score of each sample. Finally, a multivariate fusion prediction system is constructed and abbreviated as an S3WD-IKELM-PSO system, where the particle swarm optimization (PSO) algorithm is employed to determine optimal parameters. It is worth noting that actual interest rate estimation datasets are utilized to verify the feasibility and validity of the proposed prediction system. According to the experimental results, the S3WD-IKELM-PSO system exhibits superior prediction performance and generalization ability compared to existing machine learning models.

An adaptive auto-scaling framework for cloud resource provisioning

Cloud computing emerged as a technology that offers scalable access to computing resources in conjunction with low maintenance costs. In this domain, cloud users utilize virtualized resources to benefit from on-demand and long-term pricing strategies. Although the latter consists of a more cost-efficient solution, it requires accurate estimations of future workload demands, which is a challenging task. Furthermore, clouds offer threshold-based auto-scaling rules that need to be manually controlled by the users according to application requirements. Still, tuning scaling parameters is not trivial, since it is mainly based on static scaling rules that may lead to unreasonable costs and quality of service violations. In this work we introduce ADA-RP, an adaptive auto-scaling framework for reliable resource provisioning in the cloud. ADA-RP uses historical time series data for training K-means and convolutional neural networks (CNN) to categorize future workload demands as High, Medium or Low based on CPU utilization. We auto-scale cloud resources in real-time based on the predicted workload demand to reduce costs and improve application performance. The experimental analysis is based on TPC-C runs on MySQL containers deployed on the Google Cloud Platform. Experimental results are prosperous, demonstrating the ability of ADA-RP (i) to reduce MySQL deployment costs by 48% in a single-tenant environment, and (ii) to double the executed queries per second in a multi-tenant environment considering user’s budget requirements.

Single-neuron analysis of dendrites and axons reveals the network organization in mouse prefrontal cortex.

The structures of dendrites and axons form the basis for the connectivity of neural network, but their precise relationship at single-neuron level remains unclear. Here we report the complete dendrite and axon morphology of nearly 2,000 neurons in mouse prefrontal cortex (PFC). We identified morphological variations of somata, dendrites and axons across laminar layers and PFC subregions and the general rules of somatodendritic scaling with cytoarchitecture. We uncovered 24 morphologically distinguishable dendrite subtypes in 1,515 pyramidal projection neurons and 405 atypical pyramidal projection neurons and spiny stellate neurons with unique axon projection patterns. Furthermore, correspondence analysis among dendrites, local axons and long-range axons revealed coherent morphological changes associated with electrophysiological phenotypes. Finally, integrative dendrite-axon analysis uncovered the organization of potential intra-column, inter-hemispheric and inter-column connectivity among projection neuron types in PFC. Together, our study provides a comprehensive structural repertoire for the reconstruction and analysis of PFC neural network.

Effect of thickness on the superconducting properties of the thin films of FeTe0.55Se0.45 deposited on YSZ substrates using Pulsed laser deposition (PLD) technique

We report the successful fabrication of thin films of thickness ∼26 nm, 59 nm, 76 nm, and 158 nm of the composition FeTe0.55Se0.45 on the (100)-oriented Yttria-stabilized zirconia (YSZ) substrate using the pulsed laser deposition (PLD) technique. The effect of thickness on the structure and superconducting properties of the thin films has been investigated. X-ray diffraction study reveals the growth of thin films along the c-axis. The temperature-dependent resistivity measurements confirm superconductivity in all the fabricated thin films, and the highest ∼ 16.2 K has been observed for the thin film of thickness ∼59 nm, which is higher than the corresponding value of the polycrystalline target sample. The superconducting parameters such as upper critical magnetic field coherence length thermally activated energy (TAE) U0, and vortex phase state have been measured by magnetotransport measurements near in the different applied magnetic fields up to 12 T. The TAE values determined from the conventional and modified thermally activated flux flow (TAFF) models are discussed and compared. A crossover from 2-dimensional (2D) vortex to 3-dimensional (3D) vortex behaviour has been observed. The vortex phase diagrams are used to observe the transition from the vortex liquid to the vortex glass state in the grown thin films. Moreover, magnetoresistance (MR) measurements show non-saturating linear magnetoresistance (LMR) in the high field region, indicating the presence of the possible topological state in the grown thin films. The values of magnetic field dependent transport critical current density, have been evaluated from the current-voltage (I-V) curves. Furthermore, the normalised pinning force density has been used to characterize the pinning processes using the Dew Hughes’ scaling rule which reveals that the grown thin films contain δl-surface pinning centers.

A benchmarking study of different DEM coarse graining strategies

DEM-CFD is computationally very demanding and currently restricted to lab-scale systems with limited particle numbers. To address this issue, coarse graining approaches are used, summarizing particles of the original system into grains. In this study, hydrodynamic force scaling models are compared in a fixed bed, to eliminate the influence of contact force. Additionally, common contact force scaling models are benchmarked in a mechanically agitated bed to eliminate the influence of hydrodynamic forces. For both optimal scaling rules are identified based on the comparison of physical parameters. In a second step, fluidized beds are analyzed whether the best scaling models from the preceding studies focusing solely on contact or fluid forces are equally applicable in systems governed by both contact and hydrodynamic forces. In doing so, this work aims to give a reasonable recommendation, which coarse graining models should be used for the future simulations of industrial scale particle systems.

Scaling effects on the seismic response of a closed-end pipe pile embedded in dry and saturated coarse grain soils

Foundations can be subjected to dynamic or seismic loads depending on their applications and the site being constructed in. The researchers concentrated their works on investigating the reasons of the significant damage of piles during seismic excitation. Based on the findings of laboratory experiments and other numerical analyses, such failures were referred to as the kinematic impact of the earthquake on piles since they were associated with discontinuities in the subsoil because of sudden changes in soil stiffness. The current work investigates the seismic response of closed-end (CE) pipe pile using three-dimensional finite element analysis, including the impact of the scaling-up model, acceleration-time history of the ground motion, and ground conditions. The numerical model is developed using a variety of scaling rules and the outputs of the available laboratory tests. The current results showed that the saturated sand models have larger pile deformation factors than dry sand models. Pile frictional resistance was evaluated numerically, and the entire findings were evaluated against the earlier work. Mainly, the frictional resistance around the pile shaft was lower than that at the pile tip, and the frictional resistance factor on the soil surface of dry soil models was larger than that of saturated soil models. Owing to the acceleration amplifications, the pile and soil suffered cycles of compression and tension stresses. A hysteresis loop is broader and flatter on the x-axis as the shear strain increases serve to identify the shear stress–strain plane behavior. The main outputs of the scaled models were normalized to provide a deep insight of model to prototype scaling effects.

Use of miniaturized tensile specimens to evaluate the ductility and formability of dual phased steels for Rapid Alloy Prototyping

This work aims to investigate feasibility of using non-standard miniaturized tensile specimens (MTS) to characterize total elongation and formability behaviour of dual phase steels: to establish the scaling rules for MTS to predict a range of mechanical properties of steels obtained through rapid alloy prototyping (RAP) in the laboratory. Accurate measurement of these properties using miniature specimens allows the formability behaviour of materials produced using RAP, a process which can produce lab scale samples of < 100 grams. This allows the effects of multiple compositions and thermomechanical processing parameters on formability to be evaluated. For this study, the ability of miniature tensile specimens to characterize ductility, forming limits and r-values has been compared to properties measured using standard sized tensile specimens for the dual phase steels, DP600 and DP800. 5 different tensile specimens were investigated, with gauge lengths varying from 80 mm to 5 mm and samples were taken at different orientations to the rolling direction. Total elongation was evaluated against slimness ratio for each tensile sample using the Bertelle–Oliver method and good correlations were obtained, despite specimen thicknesses being below the critical value, and the tensile strength of DP800 exceeding the limit stated for the method. Forming limit curves (FLC) were predicted using the Keeler–Brazier method, based on tensile data obtained using the standard and non-standard tensile specimens These curves compared favourably to experimentally obtained FLCs, however the curve predicted using the miniaturized tensile specimen over-predicted the major strain. The anisotropy of the rolled sheet steels was characterized from r-values obtained from tensile tests performed at different orientations to the rolling direction. Values obtained from the miniature tensile test were comparable to those obtained from standard tests within the range of scatter. The results obtained through this study provide confidence in using the miniature tensile specimens to predict the formability of heterogeneous alloys such as DP steels, manufactured using the RAP process.

Numerical Investigation into Single and Double Gurney Flaps for Improving Airfoil Performance

The impact of single and double Gurney flaps of heights within the range of is determined for a typical symmetrical airfoil (NACA 0015) by numerical simulations using an unsteady Reynolds-averaged Navier–Stokes approach. The optimal scaling of the Gurney flap to maximize the lift-to-drag ratio of the airfoil is revisited and confirmed to depend on the boundary-layer thickness in the region where it is deployed. Appropriate scaling rules for both single and double Gurney flap are proposed. Studying the Gurney flap’s pressure distribution at a finely resolved level highlights its effect on the overall foil’s pressure distribution. Finally, the unsteady simulations carried out are harnessed to determine the shedding frequencies generated by the flaps and the flow structures associated with them, hinting at the physical mechanisms by which Gurney flaps may increase lift at a low drag cost, thus improving the lift-to-drag ratio of the airfoil.

Q-Factor Bounds for Microstrip Patch Antennas

Antenna bounds are a useful tool in assessing the feasibility or performance of an antenna design. Microstrip patch antennas are often limited by their relatively narrow bandwidth, and therefore Q-factor is an important design parameter, as it is related to the inverse of the fractional bandwidth. This article presents the first tight lower Q-factor bounds on microstrip patch antennas supported by an infinite dielectric substrate. The derived lower Q-factor bounds are orders of magnitude tighter than the Chu limit and introduce a new scaling rule. These bounds consider all possible geometries on the predefined design region. Moreover, it is shown that well-known patch antennas have Q-factors near the bounds and have thus a near-optimal bandwidth. The computation of the bounds is done using a method of moments (MoM) formulation. However, an approximation to these bounds using commonly available simulation tools is provided.

Delamination from an adhesive sphere: Curvature-induced dewetting versus buckling

Everyday experience confirms the tendency of adhesive films to detach from spheroidal regions of rigid substrates-what is a petty frustration when placing a sticky band aid onto a knee is a more serious matter in the coating and painting industries. Irrespective of their resistance to bending, a key driver of such phenomena is Gauss' Theorema Egregium, which implies that naturally flat sheets cannot conform to doubly curved surfaces without developing a strain whose magnitude grows sharply with the curved area. Previous attempts to characterize the onset of curvature-induced delamination, and the complex patterns it gives rise to, assumed a dewetting-like mechanism in which the propensity of two materials to form contact through interfacial energy is modified by an elastic energy penalty. We show that this approach may characterize moderately bendable sheets but fails qualitatively to describe the curvature-induced delamination of ultrathin films, whose mechanics is governed by their propensity to buckle and delaminate partially, under minute levels of compression. Combining mechanical and geometrical considerations, we introduce a minimal model for curvature-induced delamination accounting for the two buckling motifs that underlie partial delamination: shallow "rucks" and localized "folds". We predict nontrivial scaling rules for the onset of curvature-induced delamination and various features of the emerging patterns, which compare well with experiments. Beyond gaining control on the use of ultrathin adhesives in cutting-edge technologies such as stretchable electronics, our analysis is a significant step toward quantifying the multiscale morphology that emerges upon imposing geometrical and mechanical constraints on highly bendable solid objects.

Spatio-Temporal Coupling Evolution of Urbanisation and Carbon Emission in the Yangtze River Economic Belt.

The distribution characteristics of urbanisation level and per capita carbon emissions from 2006 to 2019 were investigated by the ranking scale rule, using 108 cities in the Yangtze River Economic Belt of China. A coupling coordination model was established to analyse the relative development relationship between the two, and exploratory spatial-temporal data analysis (ESTDA) was applied to reveal the spatial interaction characteristics and temporal evolution pattern of the coupling coordination degree. The results demonstrate that: (1) The urbanisation level and per capita carbon emissions of the Yangtze River Economic Belt show a stable spatial structure of 'high in the east and low in the west'. (2) The coupling and coordination degree of urbanisation level and carbon emissions show a trend of 'decreasing and then increasing', with a spatial distribution of 'high in the east and low in the west'. (3) The spatial structure exhibits strong stability, dependence, and integration. The stability is enhanced from west to east, the coupling coordination degree has strong transfer inertia, and the spatial pattern's path dependence and locking characteristics show a trend of weak fluctuation. Therefore, the coupling and coordination analysis is required for the coordinated development of urbanisation and carbon emission reduction.

Exploring the parameter space of MagLIF implosions using similarity scaling. II. Current scaling

Magnetized liner inertial fusion (MagLIF) is a magneto-inertial-fusion (MIF) concept, which is presently being studied on the Z pulsed power facility. The MagLIF platform has achieved interesting plasma conditions at stagnation and produced significant fusion yields in the laboratory. Given the relative success of MagLIF, there is a strong interest to scale the platform to higher peak currents. However, scaling MagLIF is not entirely straightforward due to the large dimensionality of the experimental input parameter space and the numerous physical processes involved in MIF implosions. In this work, we propose a novel method to scale MagLIF loads to higher currents. Our method is based on similarity (or similitude) scaling and attempts to preserve much of the physics regimes already known or being studied on today's Z pulsed-power driver. By avoiding significant deviations into unexplored and/or less well-understood regimes, the risk of unexpected outcomes on future scaled-up experiments is reduced. Using arguments based on similarity scaling, we derive the scaling rules for the experimental input parameters characterizing a MagLIF load (as functions of the characteristic current driving the implosion). We then test the estimated scaling laws for various metrics measuring performance against results of 2D radiation–magneto-hydrodynamic hydra simulations. Agreement is found between the scaling theory and the simulation results.

Mechanical stability parameters of chalcogenides and pnictides based optoelectronic materials

A study of experimental data reveals that the bulk modulus of chalcogenides and pnictides based chalcopyrites (AIIBIVC2 V and AI BIIIC2 VI) can be explained by a simple scaling rule that rely only on the crystal ionicity, ionic charge product, and the melting temperature. PVV theory of crystal ionicity, temperature dependence of elasticity and product of ionic charge theory are taken into account for the study. Based on this result, a simple microhardnessbulk modulus relation is applied to evaluate the microhardness of the complex compounds; which correspond well with the experimental data and other published results. The proposed findings support in the modeling of emerging semiconductor materials and even understanding of their mechanical properties for optoelectronics, photovoltaic, electromagnetic (EM) screening, and spintronic applications. PACS: 62.20.-x; 62.20.Qp

Galactic scaling rules in a modified dynamical model

Schulz (2017) galactic scaling rules , which include baryonic Tully–Fisher relation, have been surveyed in this work within the context of a modified dynamical model. These scaling relations are derived by employing the virial theorem and applying equilibrium and stability conditions. The scaling rules are also obtained by dimensional analysis of an integral relation between surface density and circular velocity of disk galaxies. To check the validity of the scaling relations based on observational data, we have defined, based on the properties of the model, the proper equilibrium size Req and equilibrium velocity Veq of systems. By employing these measures of length and velocity, SPARC data (Lelli et al., 2016a) is used to analyze the results. The viability of the scaling relations is tested and it is shown that, compared to some other measures of length and velocity, Req and Veq provide the closest fits to the theoretical predictions. We have compared our results with prior works and have concluded that the set of baryonic Tully–Fisher relation plus mass–size relation ( or mass–velocity) provides an appropriate description of the general characteristics of the systems. Lastly, it is shown that these scaling relations predict a certain evolution of galactic properties with redshift. This behavior provides a chance to examine the cosmic evolution of the present modified dynamical model in future works.

Shape prediction on the basis of spectrum using neural networks

We have developed a deep neural network that reconstructs the shape of a polygonal domain given the first hundred of its Dirichlet Laplacian eigenvalues. Having an encoder-decoder structure, the network maps input spectra to a latent space and then predicts the discretized image of the domain on a square grid. Tested on randomly generated pentagons, the predictions prove to be highly accurate for both concave and convex pentagons. Our analysis indicates that the network has discovered fundamental properties of the Laplacian operator, the scaling rule, and the continuous rotational symmetry. Additionally, the latent variables are strongly correlated with Weyl's parameters (area, perimeter, and a certain function of the angles) of the test polygons.