Multi-layer perceptron-based data-driven multiscale modelling of granular materials with a novel Frobenius norm-based internal variable

One objective of developing machine learning(ML)-based material models is to integrate them with well-established numerical methods to solve boundary value problems(BVPs).In the family of ML models,recurrent neural networks(RNNs)have been extensively applied to capture history-dependent constitutive responses of granular materials,but these multiple-step-based neural networks are neither sufficiently efficient nor aligned with the standard finite element method(FEM).Single-step-based neural networks like the multi-layer perceptron(MLP)are an alternative to bypass the above issues but have to introduce some internal variables to encode complex loading histories.In this work,one novel Frobenius norm-based internal variable,together with the Fourier layer and residual architectureenhanced MLP model,is crafted to replicate the history-dependent constitutive features of representative volume element(RVE)for granular materials.The obtained ML models are then seamlessly embedded into the FEM to solve the BVP of a biaxial compression case and a rigid strip footing case.The obtained solutions are comparable to results from the FEM-DEM multiscale modelling but achieve significantly improved efficiency.The results demonstrate the applicability of the proposed internal variable in enabling MLP to capture highly nonlinear constitutive responses of granular materials.

Lower Bounds of Decay Rates for Solution to the Single-Layer Quasi-Geostrophic Model

In this paper, we study the long-time behavior of solutions of the single-layer quasi-geostrophic model arising from geophysical fluid dynamics. We obtain the lower bound of the decay estimate of the solution. Utilizing the Fourier splitting method, under suitable assumptions on the initial data, for any multi-index α, we show that the solution Ψ satisfies .

Recommendation System Based on Perceptron and Graph Convolution Network

The relationship between users and items,which cannot be recovered by traditional techniques,can be extracted by the recommendation algorithm based on the graph convolution network.The current simple linear combination of these algorithms may not be sufficient to extract the complex structure of user interaction data.This paper presents a new approach to address such issues,utilizing the graph convolution network to extract association relations.The proposed approach mainly includes three modules:Embedding layer,forward propagation layer,and score prediction layer.The embedding layer models users and items according to their interaction information and generates initial feature vectors as input for the forward propagation layer.The forward propagation layer designs two parallel graph convolution networks with self-connections,which extract higher-order association relevance from users and items separately by multi-layer graph convolution.Furthermore,the forward propagation layer integrates the attention factor to assign different weights among the hop neighbors of the graph convolution network fusion,capturing more comprehensive association relevance between users and items as input for the score prediction layer.The score prediction layer introduces MLP(multi-layer perceptron)to conduct non-linear feature interaction between users and items,respectively.Finally,the prediction score of users to items is obtained.The recall rate and normalized discounted cumulative gain were used as evaluation indexes.The proposed approach effectively integrates higher-order information in user entries,and experimental analysis demonstrates its superiority over the existing algorithms.

Dynamic Multi-Layer Perceptron for Fetal Health Classification Using Cardiotocography Data

Fetal health care is vital in ensuring the health of pregnant women and the fetus.Regular check-ups need to be taken by the mother to determine the status of the fetus’growth and identify any potential problems.To know the status of the fetus,doctors monitor blood reports,Ultrasounds,cardiotocography(CTG)data,etc.Still,in this research,we have considered CTG data,which provides information on heart rate and uterine contractions during pregnancy.Several researchers have proposed various methods for classifying the status of fetus growth.Manual processing of CTG data is time-consuming and unreliable.So,automated tools should be used to classify fetal health.This study proposes a novel neural network-based architecture,the Dynamic Multi-Layer Perceptron model,evaluated from a single layer to several layers to classify fetal health.Various strategies were applied,including pre-processing data using techniques like Balancing,Scaling,Normalization hyperparameter tuning,batch normalization,early stopping,etc.,to enhance the model’s performance.A comparative analysis of the proposed method is done against the traditional machine learning models to showcase its accuracy(97%).An ablation study without any pre-processing techniques is also illustrated.This study easily provides valuable interpretations for healthcare professionals in the decision-making process.

Enhancing Healthcare Data Security and Disease Detection Using Crossover-Based Multilayer Perceptron in Smart Healthcare Systems

The healthcare data requires accurate disease detection analysis,real-timemonitoring,and advancements to ensure proper treatment for patients.Consequently,Machine Learning methods are widely utilized in Smart Healthcare Systems(SHS)to extract valuable features fromheterogeneous and high-dimensional healthcare data for predicting various diseases and monitoring patient activities.These methods are employed across different domains that are susceptible to adversarial attacks,necessitating careful consideration.Hence,this paper proposes a crossover-based Multilayer Perceptron(CMLP)model.The collected samples are pre-processed and fed into the crossover-based multilayer perceptron neural network to detect adversarial attacks on themedical records of patients.Once an attack is detected,healthcare professionals are promptly alerted to prevent data leakage.The paper utilizes two datasets,namely the synthetic dataset and the University of Queensland Vital Signs(UQVS)dataset,from which numerous samples are collected.Experimental results are conducted to evaluate the performance of the proposed CMLP model,utilizing various performancemeasures such as Recall,Precision,Accuracy,and F1-score to predict patient activities.Comparing the proposed method with existing approaches,it achieves the highest accuracy,precision,recall,and F1-score.Specifically,the proposedmethod achieves a precision of 93%,an accuracy of 97%,an F1-score of 92%,and a recall of 92%.

A Multilayer Perceptron Artificial Neural Network Study of Fatal Road Traffic Crashes

This paper examines the relationship between fatal road traffic accidents and potential predictors using multilayer perceptron artificial neural network (MLANN) models. The initial analysis employed twelve potential predictors, including traffic volume, prevailing weather conditions, roadway characteristics and features, drivers’ age and gender, and number of lanes. Based on the output of the model and the variables’ importance factors, seven significant variables are identified and used for further analysis to improve the performance of models. The model is optimized by systematically changing the parameters, including the number of hidden layers and the activation function of both the hidden and output layers. The performances of the MLANN models are evaluated using the percentage of the achieved accuracy, R-squared, and Sum of Square Error (SSE) functions.

基于Perceptron建立某车型焊钉及白车身在线检测体系

以北京奔驰汽车有限公司206车型为基础,基于Perceptron就螺柱焊钉位置尺寸、车身尺寸和覆盖件尺寸在线自动检测以及相应的精度控制进行了研究,规划并建立焊钉位置尺寸在线测量点76个,普通测量特征1373个测量点;规划并实施焊钉在线测量方案以及后端、Z1和Z2.3在线测量方案;完成在线检测工位测量工装优化改进,达成了基于Perceptron的V206焊钉及白车身在线检测体系建设,并实现了工时和成本的节约。

RESEARCH ON IMPERFECTION SENSITIVE REGION OF SINGLE-LAYER LATTICED DOMES

The concept of the imperfection sensitive region is given. The advanced stochastic imperfection method is used to research the imperfection sensitive region of single-layer latticed domes. Taking a K6 single-layer latticed dome with a diameter of 50 m as an example, its imperfection sensitive region is made up of the first 12 kinds of joints. The influence of the imperfections of support joints on the stability of the K6 single-layer latticed dome is negligible. Influences of the joint imperfections of the main rib and the secondary rib on the structural stability are similar. The initial deviations of these joints all greatly lower the critical load of the dome. Results show that the method can analyze the structural imperfection sensitive region quantitatively and accurately.

Superconducting Single-Layer T-Graphene and Novel Synthesis Routes

Single-layer superconductors are ideal materials for fabricating superconducting nano devices.However,up to date,very few single-layer elemental superconductors have been predicted and especially no one has been successfully synthesized yet.Here,using crystal structure search techniques and ab initio calculations,we predict that a single-layer planar carbon sheet with 4-and 8-membered rings called T-graphene is a new intrinsic elemental superconductor with superconducting critical temperature(Tc)up to around 20.8 K.More importantly,we propose a synthesis route to obtain such a single-layer T-graphene,that is,a T-graphene potassium intercalation compound(C4 K with P4/mmm symmetry)is firstly synthesized at high pressure(>11.5 GPa)and then quenched to ambient condition;and finally,the single-layer T-graphene can be either exfoliated using the electrochemical method from the bulk C4 K,or peeled off from bulk T-graphite C4,where C4 can be obtained from C4 K by evaporating the K atoms.Interestingly,we find that the calculated Tc of C4 K is about 30.4 K at 0 GPa,which sets a new record for layered carbon-based superconductors.The present findings add a new class of carbon-based superconductors.In particular,once the single-layer T-graphene is synthesized,it can pave the way for fabricating superconducting devices together with other 2 D materials using the layer-by-layer growth techniques.

Dynamic behavior of single-layer latticed cylindrical shells subjected to seismic loading

The single-layer latticed cylindrical shell is one of the most widely adopted space-fl'amed structures.In this paper,free vibration properties and dynamic response to horizontal and vertical seismic waves of single-layer latticed cylindrical shells are analyzed by the finite element method using ANSYS software.In the numerical study,where hundreds of cases were analyzed,the parameters considered included rise-span ratio,length-span ratio,surface load and member section size.Moreover,to better define the actual behavior of single-layer latticed shells,the study is focused on the dynamic stress response to both axial forces and bending moments.Based on the numerical results,the effects of the parameters considered on the stresses are discussed and a modified seismic force coefficient method is suggested.In addition,some advice based on these research results is presented to help in the future design of such structures.

Ultra-thin circularly polarized lens antenna based on single-layered transparent metasurface

Circularly polarized （CP） lens antenna has been applied to numerous wireless communication systems based on its unique advantages such as high antenna gain, low manufacturing cost, especially stable data transmission between the transmitter and the receiver. Unfortunately, current available CP lens antennas mostly suffer from high profile, low aperture efficiency as well as complex design. In this paper, we propose an ultra-thin CP lens antenna based on the designed single- layered Pancharatnam-Berry （PB） transparent metasurface with focusing property. The PB metasurface exhibits a high transmissivity, which ensures a high efficiency of the focusing property. Launched the metasurface with a CP patch antenna at its focal point, a low-profile lens antenna is simulated and measured. The experimental results show that our lens antenna exhibits a series of advantages including high radiation gain of 20.7 dB, aperture efficiency better than 41.3%, and also narrow half power beam width （HPBW） of 13°at about 14GHz. Our finding opens a door to realize ultra-thin transparent metasurface with other functionalities or at other working frequencies.

Single-layer graphene as a highly selective barrier for vanadium crossover with high proton selectivity

We report near-zero crossover for vanadium cross-permeation through single-layer graphene immobilized at the interface of two Nafion?polymer electrolyte membranes.Vanadium ion diffusion and migration,including proton mobility through membrane composites,were studied with and without graphene under diffusion and migration conditions.Single-layer graphene was found to effectively inhibit vanadium ion diffusion and migration under specific conditions.The single-layer graphene composites also enabled remarkable ion transmission selectivity improvements over pure Nafion membranes,with proton transport being four orders of magnitude faster than vanadium ion transport.Resistivity values of 0.02±0.005Ωcm^(2) for proton and 223±4Ωcm^(2) for vanadium ion through single atomic layer graphene are reported.This high selectivity may have significant impact on flow battery applications or for other electrochemical devices where proton conductivity is required,and transport of other species is detrimental.Our results emphasize that crossover may be essentially completely eliminated in some cases,enabling for greatly improved operational viability.

Damage mechanism of single-layer reticulated domes under severe earthquakes

To study the damage mechanism of single-layer reticulated domes subject to severe earthquakes, three limit states of single-layer reticulated domes under earthquakes are defined firstly in this paper. Then, two failure modes are presented by analyzing damage behaviors, and their characteristics are pointed out respectively. Furthermore, the damage process is analyzed and the causes of structural damage in different levels are studied. Finally, by comparing deformation and vibration status of domes with different failure modes, the principles of different failures are revealed and an integrated frame of damage mechanism is set up.

Gap opening and tuning in single-layer graphene with combined electric and magnetic field modulation

The energy band structure of single-layer graphene under one-dimensional electric and magnetic field modulation is theoretically investigated. The criterion for bandgap opening at the Dirac point is analytically derived with a two-fold degeneracy second-order perturbation method. It is shown that a direct or an indirect bandgap semiconductor could be realized in a single-layer graphene under some specific configurations of the electric and magnetic field arrangement. Due to the bandgap generated in the single-layer graphene, the Klein tunneling observed in pristine graphene is completely suppressed.

Preliminary analysis of mode truncation of single-layered reticulated domes under earthquakes

In anti-seismic calculation, the mode truncation is a significant problem to engineers if the mode-superposition response spectrum method is used, which has not been completely solved yet in some large and complex structures such as reticulated domes. In this case, some useful advices, concentrating on the problem above, are expected through a careful and comprehensive investigation of this paper. During the investigation, the authors first point out shortcomings of former researches. Then frequency-spectrum characteristics of single-layered reticulated domes were studied from the perspective of structural responses. During this process, some important results such as the existence of the main resonant section, and the fact that the relative sensitivity of these domes under horizontal and vertical impulse varies with the different R/S ratios were achieved. Furthermore, based on the study of frequency-spectrum characteristics, as well as that of earthquake input, reasonable numbers of mode truncation in single layered reticulated domes with different R/S ratio were presented. Results of case studies prove the mode truncation number proposed is valid.

Nanoscale mass sensing based on vibration of single-layered graphene sheet in thermal environments

Based on vibration analysis, single-layered graphene sheet （SLGS） with multiple attached nanoparticles is developed as nanoscale mass sensor in thermal environments. Graphene sensors are assumed to be in simplysupported configuration. Based on the nonlocal plate the- ory which incorporates size effects into the classical theory, closed-form expressions lot the frequencies and relative fre- quency shills of SLGS-based mass sensor are derived using the Galerkin method. The suggested model is justified by a good agreement between the results given by the present model and available data in literature. The effects of tem- perature difference, nonlocal parameter, the location of the nanoparticle and the number of nanoparticles on the relative frequency shift of the mass sensor are also elucidated. The obtained results show that the sensitivity of the SLGS- based mass sensor increases with increasing temperature difference.

A Hybrid Learning Method for Multilayer Perceptrons

A Newton learning method for a neural network of multilayer perceptrons is proposed in this paper. Furthermore, a hybrid learning method id legitimately developed in combination of the backpropagation method proposed by Rumelhart et al with the Newton learning method. Finally, the hybrid learning algorithm is compared with the backpropagation algorithm by some illustrations, and the results show that this hybrid leaming algorithm bas the characteristics of rapid convergence.

ERROR RESPONSE AND ROBUSTNESS OF A CLASS OF MULTILAYERED PERCEPTRONS WITH THRESHOLD FUNCTIONS

In this paper, based on a stochastic mode! for inputs and weights, and in view of the disturbance of correlative and large input and weight errors, a general algorithm to obtain the output error characteristics of a class of multilayered perceptrons with threshold functions is proposed by using statistical approach. Furthermore, the formula to calculate the robustness of the networks is also given. The result of computer simulation indicates the correctness of the algorithm.

Building up Multi－Layered Perceptrons as Classifier System for Decision Support

This paper focuses on some application issues in m.multi-layered perceptrons researches. The following problem areas are discussed: (1) the classification capability of multi-layered perceptrons; (2) theself-configuration algorithm for facilitating the design of the neural nets' structure;and,finally (3) the application of the fast BP algorithm to speed up the learning procedure. Some experimental results with respect to the application of multi-layered perceptrons as classifier systems in the comprehensive evaluation of Chinese large cities are presented.

Theoretical Proposal for a Planar Single-Layer Carbon That Shows a Potential in Superconductivity

Single-layer superconductors[1]have been the subject of considerable interests as they are ideal systems for the fundamental understanding of two-dimensional(2D)physics and for device applications.A few singlelayer superconductors are experimentally achieved(e.g.,FeSe,MoS2,and NbSe2[2-4]in the field where either charge doping or tensile strain is often required to promote the superconductivity.