scholarly journals A Neural Network Inverse Optimization Procedure for Constitutive Parameter Identification and Failure Mode Estimation of Laterally Loaded Unreinforced Masonry Walls

CivilEng ◽  
2021 ◽  
Vol 2 (4) ◽  
pp. 943-968
Author(s):  
Qudama Albu-Jasim ◽  
George Papazafeiropoulos
Keyword(s):  
Neural Network ◽  
Parameter Identification ◽  
Large Scale ◽  
Optimization Procedure ◽  
Experimental Tests ◽  
Unreinforced Masonry ◽  
Inverse Optimization ◽  
Constitutive Parameter ◽  
Small Scale ◽  
Constitutive Parameters

A new Neural Network Optimization (NNO) algorithm for constitutive material parameter identification based on inverse analysis of experimental tests of small-scale masonry prisms under compressive loads is presented. The Concrete Damaged Plasticity (CDP) constitutive model is used for the brick and mortar of the Unreinforced Masonry (URM) walls. By comparisons with experimental data taken from laboratory tests, it is demonstrated that the constitutive parameters calibrated by application of the proposed inverse optimization procedure on the small-scale (prism) experimental results are sufficiently accurate to allow for the prediction of the mechanical response of large-scale URM walls subject to compressive and lateral loads. This eliminates the need for large-scale URM wall experimental tests for the identification of their material properties, making the calibration process more economic. After verifying the accuracy of the calibrated constitutive parameters based on the above comparisons, a numerical parametric study is performed for the investigation of the effect of material behavior and geometrical aspect ratios on the failure mechanisms of large-scale URM walls.

scholarly journals Attention Enhanced Serial Unet++ Network for Removing Unevenly Distributed Haze

Electronics ◽  
2021 ◽  
Vol 10 (22) ◽  
pp. 2868
Author(s):  
Wenxuan Zhao ◽  
Yaqin Zhao ◽  
Liqi Feng ◽  
Jiaxi Tang
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Real World ◽  
Large Scale ◽  
Learning Strategy ◽  
Contextual Information ◽  
Small Scale ◽  
Image Dehazing ◽  
Atmospheric Scattering ◽  
Real World Datasets

The purpose of image dehazing is the reduction of the image degradation caused by suspended particles for supporting high-level visual tasks. Besides the atmospheric scattering model, convolutional neural network (CNN) has been used for image dehazing. However, the existing image dehazing algorithms are limited in face of unevenly distributed haze and dense haze in real-world scenes. In this paper, we propose a novel end-to-end convolutional neural network called attention enhanced serial Unet++ dehazing network (AESUnet) for single image dehazing. We attempt to build a serial Unet++ structure that adopts a serial strategy of two pruned Unet++ blocks based on residual connection. Compared with the simple Encoder–Decoder structure, the serial Unet++ module can better use the features extracted by encoders and promote contextual information fusion in different resolutions. In addition, we take some improvement measures to the Unet++ module, such as pruning, introducing the convolutional module with ResNet structure, and a residual learning strategy. Thus, the serial Unet++ module can generate more realistic images with less color distortion. Furthermore, following the serial Unet++ blocks, an attention mechanism is introduced to pay different attention to haze regions with different concentrations by learning weights in the spatial domain and channel domain. Experiments are conducted on two representative datasets: the large-scale synthetic dataset RESIDE and the small-scale real-world datasets I-HAZY and O-HAZY. The experimental results show that the proposed dehazing network is not only comparable to state-of-the-art methods for the RESIDE synthetic datasets, but also surpasses them by a very large margin for the I-HAZY and O-HAZY real-world dataset.

scholarly journals Extensive deep neural networks for transferring small scale learning to large scale systems

2019 ◽  
Vol 10 (15) ◽  
pp. 4129-4140 ◽  
Author(s):  
Kyle Mills ◽  
Kevin Ryczko ◽  
Iryna Luchak ◽  
Adam Domurad ◽  
Chris Beeler ◽  
...  
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Large Scale ◽  
Deep Neural Network ◽  
Deep Neural Networks ◽  
Small Scale ◽  
Large Scale Systems ◽  
Energy Entropy ◽  
Large Systems ◽  
Number Of Particles

We present a physically-motivated topology of a deep neural network that can efficiently infer extensive parameters (such as energy, entropy, or number of particles) of arbitrarily large systems, doing so with scaling.

scholarly journals Thermodynamic Neural Network

Entropy ◽  
2020 ◽  
Vol 22 (3) ◽  
pp. 256
Author(s):  
Todd Hylton
Keyword(s):  
Neural Network ◽  
Large Scale ◽  
Open Systems ◽  
Self Organization ◽  
Small Scale ◽  
Edge States ◽  
Fluctuation Dissipation ◽  
Multiscale Dynamics ◽  
Physical Quantities ◽  
Open Thermodynamic Systems

A thermodynamically motivated neural network model is described that self-organizes to transport charge associated with internal and external potentials while in contact with a thermal reservoir. The model integrates techniques for rapid, large-scale, reversible, conservative equilibration of node states and slow, small-scale, irreversible, dissipative adaptation of the edge states as a means to create multiscale order. All interactions in the network are local and the network structures can be generic and recurrent. Isolated networks show multiscale dynamics, and externally driven networks evolve to efficiently connect external positive and negative potentials. The model integrates concepts of conservation, potentiation, fluctuation, dissipation, adaptation, equilibration and causation to illustrate the thermodynamic evolution of organization in open systems. A key conclusion of the work is that the transport and dissipation of conserved physical quantities drives the self-organization of open thermodynamic systems.

scholarly journals Investigation of a new methodology for the prediction of drawing force in deep drawing process with respect to dimensionless analysis

2019 ◽  
Vol 14 (1) ◽  
Author(s):  
Saeed Hajiahmadi ◽  
Majid Elyasi ◽  
Mohsen Shakeri
Keyword(s):  
Deep Drawing ◽  
Large Scale ◽  
Model Comparison ◽  
Experimental Tests ◽  
Geometric Parameters ◽  
Relational Model ◽  
Small Scale ◽  
Geometrical Parameters ◽  
Drawing Force ◽  
Force Estimation

AbstractIn this research, geometric parameters were given in dimensionless form by the Buckingham pi dimensional analysis method, and a series of dimensionless groups were found for deep drawing of the round cup. To find the best group of dimensionless geometric parameters, three scales are evaluated by commercial FE software. After analyzing all effective geometric parameters, a fittest relational model of dimensionless parameters is found. St12 sheet metals were used for experimental validation, which were formed at room temperature. In addition, results and response parameters were compared in the simulation process, experimental tests, and proposed dimensionless models. By looking at the results, it very well may be inferred that geometric qualities of a large scale can be predicted with a small scale by utilizing the proposed dimensionless model. Comparison of the outcomes for dimensionless models and experimental tests shows that the proposed dimensionless models have fine precision in determining geometrical parameters and drawing force estimation. Moreover, generalizing proposed dimensionless model was applied to ensure the estimating precision of geometric values in larger scales by smaller scales.

scholarly journals ELM-Based AFL–SLFN Modeling and Multiscale Model-Modification Strategy for Online Prediction

Processes ◽  
2019 ◽  
Vol 7 (12) ◽  
pp. 893
Author(s):  
Xiaoli Wang ◽  
He Zhang ◽  
Yalin Wang ◽  
Shaoming Yang
Keyword(s):  
Neural Network ◽  
Hybrid Model ◽  
Large Scale ◽  
Multiscale Model ◽  
Small Scale ◽  
Activation Functions ◽  
Model Accuracy ◽  
Process Dynamics ◽  
Model Modification ◽  
Online Prediction

Online prediction of key parameters (e.g., process indices) is essential in many industrial processes because online measurement is not available. Data-based modeling is widely used for parameter prediction. However, model mismatch usually occurs owing to the variation of the feed properties, which changes the process dynamics. The current neural network online prediction models usually use fixed activation functions, and it is not easy to perform dynamic modification. Therefore, a few methods are proposed here. Firstly, an extreme learning machine (ELM)-based single-layer feedforward neural network with activation-function learning (AFL–SLFN) is proposed. The activation functions of the ELM are adjusted to enhance the ELM network structure and accuracy. Then, a hybrid model with adaptive weights is established by using the AFL–SLFN as a sub-model, which improves the prediction accuracy. To track the process dynamics and maintain the generalization ability of the model, a multiscale model-modification strategy is proposed. Here, small-, medium-, and large-scale modification is performed in accordance with the degree and the causes of the decrease in model accuracy. In the small-scale modification, an improved just-in-time local modeling method is used to update the parameters of the hybrid model. In the medium-scale modification, an improved elementary effect (EE)-based Morris pruning method is proposed for optimizing the sub-model structure. Remodeling is adopted in the large-scale modification. Finally, a simulation using industrial process data for tailings grade prediction in a flotation process reveals that the proposed method has better performance than some state-of-the-art methods. The proposed method can achieve rapid online training and allows optimization of the model parameters and structure for improving the model accuracy.

scholarly journals Detection of Pulmonary Tuberculosis Manifestation in Chest X-Rays using Different Convolutional Neural Network (CNN) Models

2019 ◽  
Vol 9 (1) ◽  
pp. 2270-2275 ◽  
Keyword(s):  
Neural Network ◽  
Infectious Disease ◽  
Large Scale ◽  
Rapid Diagnosis ◽  
Small Scale ◽  
X Rays ◽  
Automated Systems ◽  
Medical Field ◽  
Huge Impact ◽  
Digital Formats

Tuberculosis (TB) is airborne infectious disease which has claimed many lives than any other infectious disease. Chest X-rays (CXRs) are often used in recognizing TB manifestation site in chest. Lately, CXRs are taken in digital formats, which has made a huge impact in rapid diagnosis using automated systems in medical field. In our current work, four simple Convolutional Neural Networks (CNN) models such as VGG-16, VGG-19, RestNet50, and GoogLenet are implemented in identification of TB manifested CXRs. Two public TB image datasets were utilized to conduct this research. This study was carried out to explore the limit of accuracies and AUCs acquired by simple and small-scale CNN with complex and large-scale CNN models. The results achieved from this work are compared with results of two previous studies. The results indicate that our proposed VGG-16 model has gained highest score overall compared to the models from other two previous studies.

scholarly journals Progressive Teaching Improvement For Small Scale Learning: A Case Study in China

2020 ◽  
Vol 12 (8) ◽  
pp. 137
Author(s):  
Bo Jiang ◽  
Yanbai He ◽  
Rui Chen ◽  
Chuanyan Hao ◽  
Sijiang Liu ◽  
...  
Keyword(s):  
Neural Network ◽  
Data Analysis ◽  
Network Model ◽  
Neural Network Model ◽  
Teaching And Learning ◽  
Large Scale ◽  
Small Scale ◽  
Teaching Improvement ◽  
Data Feedback ◽  
Learning Data

Learning data feedback and analysis have been widely investigated in all aspects of education, especially for large scale remote learning scenario like Massive Open Online Courses (MOOCs) data analysis. On-site teaching and learning still remains the mainstream form for most teachers and students, and learning data analysis for such small scale scenario is rarely studied. In this work, we first develop a novel user interface to progressively collect students’ feedback after each class of a course with WeChat mini program inspired by the evaluation mechanism of most popular shopping website. Collected data are then visualized to teachers and pre-processed. We also propose a novel artificial neural network model to conduct a progressive study performance prediction. These prediction results are reported to teachers for next-class and further teaching improvement. Experimental results show that the proposed neural network model outperforms other state-of-the-art machine learning methods and reaches a precision value of 74.05% on a 3-class classifying task at the end of the term.

scholarly journals Automatic sleep stage classification with deep residual networks in a mixed-cohort setting

SLEEP ◽  
2020 ◽  
Author(s):  
Alexander Neergaard Olesen ◽  
Poul Jørgen Jennum ◽  
Emmanuel Mignot ◽  
Helge Bjarup Dissing Sorensen
Keyword(s):  
Neural Network ◽  
Large Scale ◽  
Deep Neural Network ◽  
Sleep Stage ◽  
Scoring Systems ◽  
Training Data ◽  
Future Research ◽  
Small Scale ◽  
Sleep Stage Scoring ◽  
Single Cohort

Abstract Study Objectives Sleep stage scoring is performed manually by sleep experts and is prone to subjective interpretation of scoring rules with low intra- and interscorer reliability. Many automatic systems rely on few small-scale databases for developing models, and generalizability to new datasets is thus unknown. We investigated a novel deep neural network to assess the generalizability of several large-scale cohorts. Methods A deep neural network model was developed using 15,684 polysomnography studies from five different cohorts. We applied four different scenarios: (1) impact of varying timescales in the model; (2) performance of a single cohort on other cohorts of smaller, greater, or equal size relative to the performance of other cohorts on a single cohort; (3) varying the fraction of mixed-cohort training data compared with using single-origin data; and (4) comparing models trained on combinations of data from 2, 3, and 4 cohorts. Results Overall classification accuracy improved with increasing fractions of training data (0.25%: 0.782 ± 0.097, 95% CI [0.777–0.787]; 100%: 0.869 ± 0.064, 95% CI [0.864–0.872]), and with increasing number of data sources (2: 0.788 ± 0.102, 95% CI [0.787–0.790]; 3: 0.808 ± 0.092, 95% CI [0.807–0.810]; 4: 0.821 ± 0.085, 95% CI [0.819–0.823]). Different cohorts show varying levels of generalization to other cohorts. Conclusions Automatic sleep stage scoring systems based on deep learning algorithms should consider as much data as possible from as many sources available to ensure proper generalization. Public datasets for benchmarking should be made available for future research.

Application of RBF-STARMA Model in Shipping Flow Forecasting

2010 ◽  
Vol 108-111 ◽  
pp. 893-897 ◽  
Author(s):  
Hong Qiong Huang ◽  
Shu Lan Lin ◽  
Tian Hao Tang ◽  
Ji Fang Li
Keyword(s):  
Neural Network ◽  
Hybrid Model ◽  
Spatial Data ◽  
Large Scale ◽  
Linear Trend ◽  
Moving Average ◽  
Original Data ◽  
Spatial And Temporal Variation ◽  
Small Scale ◽  
Temporal And Spatial

Based on the idea of the neural network, intelligent computing methods are used to analyze temporal and spatial data. We present the temporal and spatial autocorrelation moving average (STARMA) model based on the in-depth systematic study on time sequence of hybrid model. Firstly this paper uses radial basis function neural network to extract the temporal and spatial sequence which is non-stationary caused by large-scale non-linear trend, secondly this paper presents STARMA modeling of small-scale random spatial and temporal variation. Comparative analysis between the original data and the forecasting data shows that proposed hybrid model has better performance of fitting and generalization.

scholarly journals Estimating Early-Winter Antarctic Sea Ice Thickness From Deformed Ice Morphology

2019 ◽  
Author(s):  
M. Jeffrey Mei ◽  
Ted Maksym ◽  
Hanumant Singh
Keyword(s):  
Neural Network ◽  
Sea Ice ◽  
Relative Error ◽  
Snow Depth ◽  
Large Scale ◽  
Small Scale ◽  
Ice Thickness ◽  
Sea Ice Thickness ◽  
Mean Relative Error ◽  
Antarctic Sea Ice

Abstract. Satellites have documented variability in sea ice areal extent for decades, but there are significant challenges in obtaining analogous measurements for sea ice thickness data in the Antarctic, primarily due to difficulties in estimating snow cover on sea ice. Sea ice thickness can be estimated from surface elevation measurements, such as those from airborne/satellite LiDAR, by assuming some snow depth distribution or empirically fitting with limited data from drilled transects from various field studies. Current estimates for large-scale Antarctic sea ice thickness have errors as high as ~ 50 %, and simple statistical models of small-scale mean thickness have similarly high errors. Averaging measurements over hundreds of meters can improve the model fits to existing data, though these results do not necessarily generalize to other floes. At present, we do not have algorithms that accurately estimate sea ice thickness at high resolutions. We use a convolutional neural network with laser altimetry profiles of sea ice surfaces at 0.2 m resolution to show that it is possible to estimate sea ice thickness at 20 m resolution with better accuracy and generalization than current methods (mean relative errors ~ 15 %). Moreover, the neural network does not require specifying snow depth/density, which increases its potential applications to other LiDAR datasets. The learned features appear to correspond to basic morphological features, and these features appear to be common to other floes with the same climatology. This suggests that there is a relationship between the surface morphology and the ice thickness. The model has a mean relative error of 20 % when applied to a new floe from the region and season, which is much lower than the mean relative error for a linear fit (errors up to 47 %). This method may be extended to lower-resolution, larger-footprint data such as such as IceBridge, and suggests a possible avenue to reduce errors in satellite estimates of Antarctic sea ice thickness from ICESat-2 over current methods, especially at smaller scale.

Sign in / Sign up

Export Citation Format

Share Document