[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Next Issue
Volume 13, April-1
Previous Issue
Volume 13, March-1
You seem to have javascript disabled. Please note that many of the page functionalities won't work as expected without javascript enabled.
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
5.3
2.5
Journals
Applied Sciences
Volume 13
Issue 6
applsci-logo

Journal Menu

► Journal Menu

Journal Browser

► Journal Browser
share announcement

Appl. Sci., Volume 13, Issue 6 (March-2 2023) – 699 articles

Cover Story (view full-size image): Various types of nanoparticles and compounds, including those belonging to the porphyrinoid group, have been researched in terms of future applications in technology and medicine, including photodynamic therapy—a non-invasive tumor treatment method. Among them, chlorins and their conjugates, combined with metallic nanoparticles, deserve special attention due to their enhanced photodynamic activity and the accompanied synergic photothermal effect. Many hybrid nanosystems reveal increased cellular uptake and improved stability and, therefore, can be applied in enhanced MRI imaging, as well as in targeting therapy. This review is focused on conjugates of metallic nanoparticles and chlorins, having in mind prospective applications as photosensitizers in multimodal neoplastic therapy, as well as tumor diagnosis. View this paper
  • Issues are regarded as officially published after their release is announced to the table of contents alert mailing list.
  • You may sign up for e-mail alerts to receive table of contents of newly released issues.
  • PDF is the official format for papers published in both, html and pdf forms. To view the papers in pdf format, click on the "PDF Full-text" link, and use the free Adobe Reader to open them.
Order results
Result details
Section
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
15 pages, 4177 KiB  
Article
Research on Truck Lane Management Strategies for Platooning Speed Optimization and Control on Multi-Lane Highways
by Yikang Rui, Shu Wang, Renfei Wu and Zhe Shen
Appl. Sci. 2023, 13(6), 4072; https://doi.org/10.3390/app13064072 - 22 Mar 2023
Cited by 1 | Viewed by 2208
Abstract
Automated truck platooning has become an increasingly popular research subject, and its applicability to highways is considered one of the earliest possible landing scenarios for automated driving. However, there is a lack of research regarding the combination of truck platooning technology and truck [...] Read more.
Automated truck platooning has become an increasingly popular research subject, and its applicability to highways is considered one of the earliest possible landing scenarios for automated driving. However, there is a lack of research regarding the combination of truck platooning technology and truck lane management strategy on multilane highways in the environment of a cooperative vehicle–infrastructure system (CVIS). For highway weaving sections under the CVIS environment, this paper proposes a truck platooning optimal speed control model based on multi-objective optimization. Through a combination of model predictive control and the cell transmission model, this approach considers the bottleneck cell traffic flow, overall vehicle travel time, and truck platooning fuel consumption as objectives. By conducting experiments on a mixed traffic flow simulation platform, the multi-lane management strategies and optimal speed control effect were evaluated through different scenarios. This study also determined the appropriate proportion of truck platooning for an exclusive lane and to increase truck lanes, thus providing effective lane management decision support for highway managers. Full article
(This article belongs to the Special Issue Novel Methods and Technologies for Intelligent Vehicles)
►▼ Show Figures

Figure 1

Figure 1
<p>Nanjing–Shanghai highway in bright cyan (<b>left</b>) and marked section map (<b>right</b>).</p> Full article ">Figure 2
<p>Schematic diagram of lane marking mode in weaving section (<b>left</b>) and simulation diagram of weaving section (<b>right</b>).</p> Full article ">Figure 3
<p>Cross-section traffic volume time varying diagram and section speed distribution diagram.</p> Full article ">Figure 4
<p>Delay and conflict rate of truck platooning in the queue phase in Scenario I.</p> Full article ">Figure 5
<p>Delay and conflict rate of truck platooning in the separation phase in Scenario I.</p> Full article ">Figure 6
<p>Separation phase with 4F32T strategy through the weaving section in Scenario II.</p> Full article ">Figure 7
<p>Comparison of different vehicle types under the 4F32T strategy.</p> Full article ">Figure 8
<p>Comparison of different lanes with the 4F32T strategy.</p> Full article ">
10 pages, 2249 KiB  
Article
Recovering Microscopic Images in Material Science Documents by Image Inpainting
by Taeyun Kim and Byung Chul Yeo
Appl. Sci. 2023, 13(6), 4071; https://doi.org/10.3390/app13064071 - 22 Mar 2023
Viewed by 2077
Abstract
Microscopic images in material science documents have increased in number due to the growth and common use of electron microscopy instruments. Through the use of data mining techniques, they are easily accessible and can be obtained from documents published online. As data-driven approaches [...] Read more.
Microscopic images in material science documents have increased in number due to the growth and common use of electron microscopy instruments. Through the use of data mining techniques, they are easily accessible and can be obtained from documents published online. As data-driven approaches are becoming increasingly common in the material science field, massively acquired experimental images through microscopy play important roles in terms of developing an artificial intelligence (AI) model for the purposes of automatically diagnosing crucial material structures. However, irrelevant objects (e.g., letters, scale bars, and arrows) that are often present inside original microscopic photos should be removed for the purposes of improving the AI models. To avoid the issue above, we applied four image inpainting algorithms (i.e., shift-net, global and local, contextual attention, and gated convolution) to a learning approach, with the aim of recovering microscopic images in journal papers. We estimated the structural similarity index measure (SSIM) and ℓ1/ℓ2 errors, which are often used as measures of image quality. Lastly, we observed that gated convolution possessed the best performance for inpainting the microscopic images. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Examples of microscopic images in material science documents: (<b>a</b>) represents the STEM image of Au nanoparticles. reprinted with permission from Ref. [<a href="#B12-applsci-13-04071" class="html-bibr">12</a>]. Copyright (2020) Elsevier; (<b>b</b>) represents the TEM image of Co<sub>8</sub>Fe<sub>2</sub>-MOF. reprinted with permission from Ref. [<a href="#B13-applsci-13-04071" class="html-bibr">13</a>]. Copyright (2021) Elsevier; and (<b>c</b>) represents the SEM image of S/PCMSs composites. reprinted with permission from Ref. [<a href="#B14-applsci-13-04071" class="html-bibr">14</a>]. Copyright (2019) Elsevier.</p> Full article ">Figure 2
<p>Architectures of deep learning methods (In the figures, input images are the example of the microscopic images in the reprinted with permission from Ref. [<a href="#B14-applsci-13-04071" class="html-bibr">14</a>]. Copyright (2019) Elsevier.): (<b>a</b>) represents the CNN-based method and (<b>b</b>) represents the GAN-based method.</p> Full article ">Figure 3
<p>Statistical results of the two additives: (<b>a</b>) represents the averaged positions (red dots) of the alphabet characters and scale bars regarding the example of the microscopic images (reprinted with permission from Ref. [<a href="#B12-applsci-13-04071" class="html-bibr">12</a>]. Copyright (2020) Elsevier) and (<b>b</b>) represents the cumulative probability distribution function, according to the length of the target area (pixel) of the target regions comprising the additives.</p> Full article ">Figure 4
<p>Processing sequence in regard to generating a threshold-based mask from the example of a microscopic image (reprinted with permission from Ref. [<a href="#B40-applsci-13-04071" class="html-bibr">40</a>]. Copyright (2021) Elsevier): (<b>a</b>) represents the input image; (<b>b</b>) represents the intermediate image in the context of threshold-based masking; and (<b>c</b>) represents the final image.</p> Full article ">Figure 5
<p>Comparison of the results, which was achieved by using a statistical mask and a threshold mask from the example of microscopic image (reprinted with permission from Ref. [<a href="#B41-applsci-13-04071" class="html-bibr">41</a>]. Copyright (2019) Elsevier): (<b>a</b>) represents input image; (<b>b</b>) represents output image using a statistical mask; and (<b>c</b>) represents the output image when using a threshold-based mask.</p> Full article ">Figure 6
<p>Comparison of the results of various image inpainting methods from the three examples of microscopic images (from top to bottom. reprinted with permission from Ref. [<a href="#B12-applsci-13-04071" class="html-bibr">12</a>]. Copyright (2020) Elsevier, from Ref. [<a href="#B13-applsci-13-04071" class="html-bibr">13</a>]. Copyright (2021) Elsevier, and from Ref. [<a href="#B14-applsci-13-04071" class="html-bibr">14</a>]. Copyright (2019) Elsevier): (<b>a</b>) represents the input image; (<b>b</b>) represents the output image of shift-net; (<b>c</b>) represents the output image of global and local; (<b>d</b>) represents the output image of contextual attention; (<b>e</b>) represents the output image of gated convolution; and (<b>f</b>) represents the ground truth.</p> Full article ">
58 pages, 2844 KiB  
Review
Fewer Dimensions for Higher Thermal Performance: A Review on 2D Nanofluids
by José Pereira, Ana Moita and António Moreira
Appl. Sci. 2023, 13(6), 4070; https://doi.org/10.3390/app13064070 - 22 Mar 2023
Cited by 5 | Viewed by 2496
Abstract
The current work aims to offer a specific overview of the homogeneous dispersions of 2D nanomaterials in heat transfer base fluids—so-called 2D nanofluids. This data compilation emerged from the critical overview of the findings of the published scientific articles regarding 2D nanofluids. The [...] Read more.
The current work aims to offer a specific overview of the homogeneous dispersions of 2D nanomaterials in heat transfer base fluids—so-called 2D nanofluids. This data compilation emerged from the critical overview of the findings of the published scientific articles regarding 2D nanofluids. The applicability of such fluids as promising alternatives to the conventional heat transfer and thermal energy storage fluids is comprehensively investigated. These are fluids that simultaneously possess superior thermophysical properties and can be processed according to innovative environmentally friendly methods and techniques. Furthermore, their very reduced dimensions are suitable for the decrease in the size of thermal management systems, and the devices have attracted a lot of attention from researchers in different fields. Some examples of 2D nanofluids are those which incorporate graphene, graphene oxide, hexagonal boron nitride, molybdenum disulfide nanoparticles, and hybrid formulations. Although the published results are not always consistent, it was found that this type of nanofluid can improve the thermal conductivity of traditional base fluids by more than 150%, achieving values of approximately 6500 W·m−1·K−1 and interface thermal conductance above 50 MW·m−2·K−1. Such beneficial features permit the attainment of increments above 60% in the overall efficiency of photovoltaic/thermal solar systems, a 70% reduction in the entropy generation in parabolic trough collectors and increases of approximately 200% in the convective heat transfer coefficient in heat exchangers and heat pipes. These findings identify those fluids as suitable heat transfer and thermal storage media. The current work intends to partially suppress the literature gap by gathering detailed information on 2D nanofluids in a single study. The thermophysical properties of 2D nanofluids and not of their traditional counterparts, as it is usually encountered in the literature, and the extended detailed sections dedicated to the potential applications of 2D nanofluids are features that may set this research apart from previously published works. Additionally, a major part of the included literature references consider exclusively 2D nanomaterials and the corresponding nanofluids, which also constitutes a major gathering of specific data regarding these types of materials. Upon its conclusion, this work will provide a general overview of 2D nanofluids. Full article
(This article belongs to the Section Materials Science and Engineering)
►▼ Show Figures

Figure 1

Figure 1
<p>Types of 0D, 1D, and 2D nanomaterials.</p> Full article ">Figure 2
<p>Schematic diagram of the qualitative comparison of the thermophysical properties of the main 2D nanofluids.</p> Full article ">Figure 3
<p>Main practical applications of 2D nanofluids.</p> Full article ">Figure 4
<p>Schematic diagram of the preparation method of the 2D h-BN/transformer oil nanofluids.</p> Full article ">Figure 5
<p>Main tungsten disulfide preparation methods.</p> Full article ">Figure 6
<p>Schematic diagram of a four-ball tribometer and effect of the concentration of nanosheets.</p> Full article ">Figure 7
<p>Schematic diagram of microwave-synthesized molybdenum disulfide nanoparticles. Adapted from [<a href="#B96-applsci-13-04070" class="html-bibr">96</a>].</p> Full article ">Figure 8
<p>Schematic diagram of the percolation mechanism of nanofluids.</p> Full article ">Figure 9
<p>Schematic diagram of the synthesis procedure of MXene nanosheets in a water-based nanofluid.</p> Full article ">Figure 10
<p>Main steps of the preparation of the MXene/soybean oil 2D nanofluid. Adapted from [<a href="#B103-applsci-13-04070" class="html-bibr">103</a>].</p> Full article ">Figure 11
<p>Schematic diagram of the rGO-Fe<sub>2</sub>O<sub>3</sub> nanocomposite. Adapted from [<a href="#B113-applsci-13-04070" class="html-bibr">113</a>].</p> Full article ">Figure 12
<p>Preparation method of the Gr-gold nanoparticles in ethylene glycol 2D nanofluid. Adapted from [<a href="#B114-applsci-13-04070" class="html-bibr">114</a>].</p> Full article ">Figure 13
<p>Configurations of typical PV/T systems with nanofluids: (<b>a</b>) as a coolant; (<b>b</b>) as a coolant and spectral filter with double-pass channel; (<b>c</b>) as a coolant and spectral filter with separate channels.</p> Full article ">Figure 14
<p>Schematic diagram of the Hummer method for the synthesis of GO.</p> Full article ">Figure 15
<p>Schematic diagram of the configuration of the PV/T system for four different conditions. Adapted from [<a href="#B137-applsci-13-04070" class="html-bibr">137</a>].</p> Full article ">
15 pages, 1313 KiB  
Article
Learning Methods and Predictive Modeling to Identify Failure by Human Factors in the Aviation Industry
by Rui P. R. Nogueira, Rui Melicio, Duarte Valério and Luís F. F. M. Santos
Appl. Sci. 2023, 13(6), 4069; https://doi.org/10.3390/app13064069 - 22 Mar 2023
Cited by 13 | Viewed by 3691
Abstract
This paper proposes a model capable of predicting fatal occurrences in aviation events such as accidents and incidents, using as inputs the human factors that contributed to each incident, together with information about the flight. This is important because aviation demands have increased [...] Read more.
This paper proposes a model capable of predicting fatal occurrences in aviation events such as accidents and incidents, using as inputs the human factors that contributed to each incident, together with information about the flight. This is important because aviation demands have increased over the years; while safety standards are very rigorous, managing risk and preventing failures due to human factors, thereby further increasing safety, requires models capable of predicting potential failures or risky situations. The database for this paper’s model was provided by the Aviation Safety Network (ASN). Correlations between leading causes of incident and the human element are proposed, using the Human Factors Analysis Classification System (HFACS). A classification model system is proposed, with the database preprocessed for the use of machine learning techniques. For modeling, two supervised learning algorithms, Random Forest (RF) and Artificial Neural Networks (ANN), and the semi-supervised Active Learning (AL) are considered. Their respective structures are optimized applying hyperparameter analysis to improve the model. The best predictive model, obtained with RF, was able to achieve an accuracy of 90%, macro F1 of 87%, and a recall of 86%, outperforming ANN models, with a lower ability to predict fatal accidents. These performances are expected to assist decision makers in planning actions to avoid human factors that may cause aviation incidents, and to direct efforts to the more important areas. Full article
(This article belongs to the Special Issue Research on Aviation Safety)
►▼ Show Figures

Figure 1

Figure 1
<p>The evolution of the SMS through time.</p> Full article ">Figure 2
<p>Structure of a random forest with <span class="html-italic">N</span> trees.</p> Full article ">Figure 3
<p>Architectural graph of a MLP with four layers, of which two are hidden between the input and output layers.</p> Full article ">Figure 4
<p>Schema of the database modeling.</p> Full article ">Figure 5
<p>Confusion Matrixes for Model 1 (left) and Model 2 (right) with RF algorithm (top) and MLP algorithm (bottom). (<b>a</b>) Random Forest, Model 1. (<b>b</b>) Random Forest, Model 2. (<b>c</b>) Neural Network, Model 1. (<b>d</b>) Neural Network, Model 2.</p> Full article ">Figure 6
<p>Accuracy (<b>a</b>) and function loss (<b>b</b>) for Model 2 with MLP algorithm.</p> Full article ">Figure 7
<p>Results for Model 2 with AL after 50 queries performed with confusion matrix (<b>a</b>) and MLP algorithm (<b>b</b>).</p> Full article ">
44 pages, 4871 KiB  
Article
ShuffleDetect: Detecting Adversarial Images against Convolutional Neural Networks
by Raluca Chitic, Ali Osman Topal and Franck Leprévost
Appl. Sci. 2023, 13(6), 4068; https://doi.org/10.3390/app13064068 - 22 Mar 2023
Cited by 1 | Viewed by 1628
Abstract
Recently, convolutional neural networks (CNNs) have become the main drivers in many image recognition applications. However, they are vulnerable to adversarial attacks, which can lead to disastrous consequences. This paper introduces ShuffleDetect as a new and efficient unsupervised method for the detection of [...] Read more.
Recently, convolutional neural networks (CNNs) have become the main drivers in many image recognition applications. However, they are vulnerable to adversarial attacks, which can lead to disastrous consequences. This paper introduces ShuffleDetect as a new and efficient unsupervised method for the detection of adversarial images against trained convolutional neural networks. Its main feature is to split an input image into non-overlapping patches, then swap the patches according to permutations, and count the number of permutations for which the CNN classifies the unshuffled input image and the shuffled image into different categories. The image is declared adversarial if and only if the proportion of such permutations exceeds a certain threshold value. A series of 8 targeted or untargeted attacks was applied on 10 diverse and state-of-the-art ImageNet-trained CNNs, leading to 9500 relevant clean and adversarial images. We assessed the performance of ShuffleDetect intrinsically and compared it with another detector. Experiments show that ShuffleDetect is an easy-to-implement, very fast, and near memory-free detector that achieves high detection rates and low false positive rates. Full article
(This article belongs to the Section Computing and Artificial Intelligence)
►▼ Show Figures

Figure 1

Figure 1
<p>A <math display="inline"><semantics> <mrow> <mn>224</mn> <mo>×</mo> <mn>224</mn> </mrow> </semantics></math> image <math display="inline"><semantics> <mi mathvariant="script">I</mi> </semantics></math> is divided into 4 patches of size <math display="inline"><semantics> <mrow> <mn>112</mn> <mo>×</mo> <mn>112</mn> </mrow> </semantics></math> (top picture). The patches are shuffled around according to the permutation <math display="inline"><semantics> <mrow> <mi>σ</mi> <mo>=</mo> <mrow> <mo stretchy="false">(</mo> <mn>1</mn> <mo>,</mo> <mn>4</mn> <mo stretchy="false">)</mo> </mrow> <mrow> <mo stretchy="false">(</mo> <mn>2</mn> <mo>,</mo> <mn>3</mn> <mo stretchy="false">)</mo> </mrow> <mo>∈</mo> <msub> <mi mathvariant="fraktur">S</mi> <mn>4</mn> </msub> </mrow> </semantics></math>, leading to <math display="inline"><semantics> <mrow> <mi>s</mi> <msub> <mi>h</mi> <mi>σ</mi> </msub> <mrow> <mo stretchy="false">(</mo> <mi mathvariant="script">I</mi> <mo>,</mo> <mn>112</mn> <mo stretchy="false">)</mo> </mrow> </mrow> </semantics></math> (bottom picture). Both <math display="inline"><semantics> <mi mathvariant="script">I</mi> </semantics></math> and <math display="inline"><semantics> <mrow> <mi>s</mi> <msub> <mi>h</mi> <mi>σ</mi> </msub> <mrow> <mo stretchy="false">(</mo> <mi mathvariant="script">I</mi> <mo>,</mo> <mn>112</mn> <mo stretchy="false">)</mo> </mrow> </mrow> </semantics></math> are sent to the CNN to extract the output vector.</p> Full article ">Figure 2
<p>ShuffleDetect performed on 100 clean (ancestor) images with 100 permutations.</p> Full article ">Figure 3
<p>Average outcome over the 10 CNNs of ShuffleDetect performed with 100 permutations on the adversarial images created for the target scenario by EA, BIM, PGD Inf, and PGD L2.</p> Full article ">Figure 4
<p>Average outcome over all relevant CNNs of ShuffleDetect performed with 100 permutations on the adversarial images created for the untargeted scenario by EA, FGSM, BIM, PGD Inf, PGD L2, CW Inf, and DeepFool.</p> Full article ">Figure A1
<p>The 100 ancestor images <math display="inline"><semantics> <msubsup> <mi mathvariant="script">A</mi> <mi>q</mi> <mi>p</mi> </msubsup> </semantics></math> used in the experiments. <math display="inline"><semantics> <msubsup> <mi mathvariant="script">A</mi> <mi>q</mi> <mi>p</mi> </msubsup> </semantics></math> pictured in the <span class="html-italic">q</span>th row and <span class="html-italic">q</span>th column (<math display="inline"><semantics> <mrow> <mn>1</mn> <mo>≤</mo> <mi>p</mi> <mo>,</mo> <mi>q</mi> <mo>≤</mo> <mn>10</mn> </mrow> </semantics></math>) is randomly chosen from the ImageNet validation set of the ancestor category <math display="inline"><semantics> <msub> <mi>c</mi> <msub> <mi>a</mi> <mi>q</mi> </msub> </msub> </semantics></math> specified on the left of the <span class="html-italic">q</span>th row.</p> Full article ">Figure A2
<p>Shuffling test results of 100 clean (ancestor) images on <math display="inline"><semantics> <mrow> <mi mathvariant="script">C</mi> <mo>=</mo> <mi mathvariant="script">C</mi> </mrow> </semantics></math> for <math display="inline"><semantics> <mrow> <mn>1</mn> <mo>≤</mo> <mi>k</mi> <mo>≤</mo> <mn>10</mn> </mrow> </semantics></math> over 100 permutations.</p> Full article ">Figure A3
<p>ShuffleDetect results for adversarial images generated by the <b>EA-targeted</b> attack on <math display="inline"><semantics> <mrow> <mi mathvariant="script">C</mi> <mo>=</mo> <mi mathvariant="script">C</mi> </mrow> </semantics></math> for <math display="inline"><semantics> <mrow> <mn>1</mn> <mo>≤</mo> <mi>k</mi> <mo>≤</mo> <mn>10</mn> </mrow> </semantics></math> over 100 permutations.</p> Full article ">Figure A4
<p>ShuffleDetect results for adversarial images generated by the <b>BIM-targeted</b> attack on <math display="inline"><semantics> <mrow> <mi mathvariant="script">C</mi> <mo>=</mo> <mi mathvariant="script">C</mi> </mrow> </semantics></math> for <math display="inline"><semantics> <mrow> <mn>1</mn> <mo>≤</mo> <mi>k</mi> <mo>≤</mo> <mn>10</mn> </mrow> </semantics></math> over 100 permutations.</p> Full article ">Figure A5
<p>ShuffleDetect results for adversarial images generated by the <b>PGD Inf-targeted</b> attack on <math display="inline"><semantics> <mrow> <mi mathvariant="script">C</mi> <mo>=</mo> <mi mathvariant="script">C</mi> </mrow> </semantics></math> for <math display="inline"><semantics> <mrow> <mn>1</mn> <mo>≤</mo> <mi>k</mi> <mo>≤</mo> <mn>10</mn> </mrow> </semantics></math> over 100 permutations.</p> Full article ">Figure A6
<p>ShuffleDetect results for adversarial images generated by the <b>PGD L2-targeted</b> attack on <math display="inline"><semantics> <mrow> <mi mathvariant="script">C</mi> <mo>=</mo> <mi mathvariant="script">C</mi> </mrow> </semantics></math> for <math display="inline"><semantics> <mrow> <mn>1</mn> <mo>≤</mo> <mi>k</mi> <mo>≤</mo> <mn>10</mn> </mrow> </semantics></math> over 100 permutations.</p> Full article ">Figure A7
<p>ShuffleDetect results for adversarial images generated by the <b>EA-untargeted</b> attack on <math display="inline"><semantics> <mrow> <mi mathvariant="script">C</mi> <mo>=</mo> <mi mathvariant="script">C</mi> </mrow> </semantics></math> for <math display="inline"><semantics> <mrow> <mn>1</mn> <mo>≤</mo> <mi>k</mi> <mo>≤</mo> <mn>10</mn> </mrow> </semantics></math> over 100 permutations.</p> Full article ">Figure A8
<p>ShuffleDetect results for adversarial images generated by the <b>FGSM-untargeted</b> attack on <math display="inline"><semantics> <mrow> <mi mathvariant="script">C</mi> <mo>=</mo> <mi mathvariant="script">C</mi> </mrow> </semantics></math> for <math display="inline"><semantics> <mrow> <mn>2</mn> <mo>≤</mo> <mi>k</mi> <mo>≤</mo> <mn>10</mn> </mrow> </semantics></math> over 100 permutations.</p> Full article ">Figure A9
<p>ShuffleDetect results for adversarial images generated by the <b>BIM-untargeted</b> attack on <math display="inline"><semantics> <mrow> <mi mathvariant="script">C</mi> <mo>=</mo> <mi mathvariant="script">C</mi> </mrow> </semantics></math> for <math display="inline"><semantics> <mrow> <mn>1</mn> <mo>≤</mo> <mi>k</mi> <mo>≤</mo> <mn>10</mn> </mrow> </semantics></math> over 100 permutations.</p> Full article ">Figure A10
<p>ShuffleDetect results for adversarial images generated by the <b>PGD Inf-untargeted</b> attack on <math display="inline"><semantics> <mrow> <mi mathvariant="script">C</mi> <mo>=</mo> <mi mathvariant="script">C</mi> </mrow> </semantics></math> for <math display="inline"><semantics> <mrow> <mn>1</mn> <mo>≤</mo> <mi>k</mi> <mo>≤</mo> <mn>10</mn> </mrow> </semantics></math> over 100 permutations.</p> Full article ">Figure A11
<p>ShuffleDetect results for adversarial images generated by the <b>PGD L2-untargeted</b> attack on <math display="inline"><semantics> <mrow> <mi mathvariant="script">C</mi> <mo>=</mo> <mi mathvariant="script">C</mi> </mrow> </semantics></math> for <math display="inline"><semantics> <mrow> <mn>1</mn> <mo>≤</mo> <mi>k</mi> <mo>≤</mo> <mn>10</mn> </mrow> </semantics></math> over 100 permutations.</p> Full article ">Figure A12
<p>ShuffleDetect results for adversarial images generated by the <b>CW Inf-untargeted</b> attack on <math display="inline"><semantics> <mrow> <mi mathvariant="script">C</mi> <mo>=</mo> <mi mathvariant="script">C</mi> </mrow> </semantics></math> for <math display="inline"><semantics> <mrow> <mn>1</mn> <mo>≤</mo> <mi>k</mi> <mo>≤</mo> <mn>10</mn> </mrow> </semantics></math> over 100 permutations.</p> Full article ">Figure A13
<p>ShuffleDetect results for adversarial images generated by the <b>Deep Fool-untargeted</b> attack on <math display="inline"><semantics> <mrow> <mi mathvariant="script">C</mi> <mo>=</mo> <mi mathvariant="script">C</mi> </mrow> </semantics></math> for <math display="inline"><semantics> <mrow> <mn>1</mn> <mo>≤</mo> <mi>k</mi> <mo>≤</mo> <mn>10</mn> </mrow> </semantics></math> over 100 permutations.</p> Full article ">
14 pages, 7964 KiB  
Article
Failure Modelling of CP800 Using Acoustic Emission Analysis
by Eugen Stockburger, Hendrik Wester and Bernd-Arno Behrens
Appl. Sci. 2023, 13(6), 4067; https://doi.org/10.3390/app13064067 - 22 Mar 2023
Cited by 2 | Viewed by 1301
Abstract
Advanced high-strength steels (AHHS) are widely used in many production lines of car components. For efficient design of the forming processes, numerical methods are frequently applied in the automotive industry. To model the forming processes realistically, exact material data and analytical models are [...] Read more.
Advanced high-strength steels (AHHS) are widely used in many production lines of car components. For efficient design of the forming processes, numerical methods are frequently applied in the automotive industry. To model the forming processes realistically, exact material data and analytical models are required. With respect to failure modelling, the accurate determination of failure onset continues to be a challenge. In this article, the complex phase (CP) steel CP800 is characterised for its failure characteristics using tensile tests with butterfly specimens. The material failure was determined by three evaluation methods: mechanically by a sudden drop in the forming force, optically by a crack appearing on the specimen surface, and acoustically by burst signals. As to be expected, the mechanical evaluation method determined material failure the latest, while the optical and acoustical methods showed similar values. Numerical models of the butterfly tests were created using boundary conditions determined by each evaluation method. A comparison of the experiments, regarding the forming force and the distribution of the equivalent plastic strain, showed sufficient agreement. Based on the numerical models, the characteristic stress states of each test were evaluated, which showed similar values for the mechanical and optical evaluation method. The characteristic stress states derived from the acoustical evaluation method were shifted to higher triaxialities, compared to the other methods. Matching the point in time of material failure, the equivalent plastic strain at failure was highest for the mechanical evaluation method, with lower values for the other two methods. Furter, three Johnson–Cook (JC) failure models were parametrised and subsequently compared. The major difference was in the slope of the failure models, of which the optical evaluation method showed the lowest slope. The reasons for the differences are the different stress states and the different equivalent plastic strains due to different evaluation areas. Full article
(This article belongs to the Special Issue Advances in Acoustic Emission Technique: Tests Interpretation and Numerical Simulations)
►▼ Show Figures

Figure 1

Figure 1
<p>Butterfly specimen with close-up of the investigation area as well as stochastic pattern (<b>A</b>), and schematic representation of the test setup for butterfly specimen with optical as well as acoustical measuring system (<b>B</b>).</p> Full article ">Figure 2
<p>Measured forming force—amplitude over time curve (<b>A</b>) and used evaluation methods for determining failure of the butterfly specimen: mechanical (<b>B</b>), optical (<b>C</b>), as well as acoustical (<b>D</b>).</p> Full article ">Figure 3
<p>Reduced simulation models of the butterfly tests with boundary conditions.</p> Full article ">Figure 4
<p>Evaluation areas of the butterfly simulation models: mechanical as well as optical (<b>A</b>) and acoustical evaluation method (<b>B</b>).</p> Full article ">Figure 5
<p>Microscopic image of CP800 steel (<b>A</b>) and close-up (<b>B</b>).</p> Full article ">Figure 6
<p>Experimental and extrapolated flow curves (<b>A</b>) and yield curves (<b>B</b>) for CP800.</p> Full article ">Figure 7
<p>Average failure displacement in x-direction <math display="inline"><semantics> <mrow> <msub> <mrow> <mover accent="true"> <mrow> <mi>u</mi> </mrow> <mo>−</mo> </mover> </mrow> <mrow> <mi mathvariant="normal">x</mi> </mrow> </msub> </mrow> </semantics></math> and in y-direction <math display="inline"><semantics> <mrow> <msub> <mrow> <mover accent="true"> <mrow> <mi>u</mi> </mrow> <mo>−</mo> </mover> </mrow> <mrow> <mi mathvariant="normal">y</mi> </mrow> </msub> </mrow> </semantics></math> of the butterfly tests for CP800.</p> Full article ">Figure 8
<p>Comparison of the experimental and numerical force—displacement curves for the three evaluation methods.</p> Full article ">Figure 9
<p>Comparison of the optically measured and numerically calculated equivalent plastic strain distributions of the butterfly tests for the optical evaluation method.</p> Full article ">Figure 10
<p>Equivalent plastic strain–triaxiality curves (<b>A</b>) and equivalent plastic-strain-normalised Lode angle curves (<b>B</b>) of the butterfly specimens for the three evaluation methods.</p> Full article ">Figure 11
<p>Stress state (<b>A</b>) and equivalent plastic strain (<b>B</b>) of the butterfly specimens for the three evaluation methods.</p> Full article ">Figure 12
<p>JC failure models of the three evaluation methods for CP800.</p> Full article ">
22 pages, 7790 KiB  
Article
Multicriteria Analysis of a Solar-Assisted Space Heating Unit with a High-Temperature Heat Pump for the Greek Climate Conditions
by Evangelos Bellos, Panagiotis Lykas, Dimitrios Tsimpoukis, Dimitrios N. Korres, Angeliki Kitsopoulou, Michail Gr. Vrachopoulos and Christos Tzivanidis
Appl. Sci. 2023, 13(6), 4066; https://doi.org/10.3390/app13064066 - 22 Mar 2023
Cited by 1 | Viewed by 1864
Abstract
The goal of this investigation is the thorough analysis and optimization of a solar-assisted heat pump heating unit for covering the space heating demand for a building in Athens, Greece. The novelty of the studied system is the use of a high-temperature heat [...] Read more.
The goal of this investigation is the thorough analysis and optimization of a solar-assisted heat pump heating unit for covering the space heating demand for a building in Athens, Greece. The novelty of the studied system is the use of a high-temperature heat pump that can operate with radiative terminal units, leading to high thermal comfort standards. The examined system includes flat-plate solar thermal collectors, an insulated thermal storage tank, auxiliary electrical thermal resistance in the tank and a high-temperature heat pump. The economic optimization indicates that the optimal design includes 35 m2 of solar thermal collectors connected with a storage tank of 2 m3 for facing the total heating demand of 6785 kWh. In this case, the life cycle cost was calculated at 22,694 EUR, the seasonal system coefficient of performance at 2.95 and the mean solar thermal efficiency at 31.60%. On the other hand, the multi-objective optimization indicates the optimum design is the selection of 50 m2 of solar field connected to a thermal tank of 3 m3. In this scenario, the life cycle cost was calculated at 24,084 EUR, the seasonal system coefficient of performance at 4.07 and the mean solar thermal efficiency at 25.33%. Full article
(This article belongs to the Special Issue Advances in Solar Collector: Techniques and Applications)
►▼ Show Figures

Figure 1

Figure 1
<p>The examined system with a solar-assisted heating unit which includes a solar field, a thermal storage tank with auxiliary electrical resistance and a high-temperature heat pump.</p> Full article ">Figure 2
<p>Performance of the heat pump for different operating conditions.</p> Full article ">Figure 3
<p>Depiction for the accuracy of the approximation formula for the heat pump COP.</p> Full article ">Figure 4
<p>Climate conditions of Athens in the winter period (1 October to 30 April).</p> Full article ">Figure 5
<p>Heating load and cumulative heating demand for the examined building (1 October to 30 April).</p> Full article ">Figure 6
<p>Mean water temperature in the storage tank for the case with 30 m<sup>2</sup> collecting area and 3 m<sup>3</sup> tank volume (1 October to 30 April).</p> Full article ">Figure 7
<p>Electricity demand for the compressor and the auxiliary electrical resistance in the tank for the case with 30 m<sup>2</sup> collecting area and 3 m<sup>3</sup> tank volume (1 October to 30 April).</p> Full article ">Figure 8
<p>COP for the system and the heat pump for the case with a 30 m<sup>2</sup> collecting area and 3 m<sup>3</sup> tank volume (1 October to 30 April).</p> Full article ">Figure 9
<p>Variation in energy quantities for different collecting areas with the storage tank volume at 3 m<sup>3</sup>.</p> Full article ">Figure 10
<p>Variation in energy quantities for different storage tank volumes with the collecting area at 30 m<sup>2</sup>.</p> Full article ">Figure 11
<p>Mean yearly solar thermal efficiency of the collectors for different combinations of (A<sub>col</sub>) and (V).</p> Full article ">Figure 12
<p>Total electricity demand (compressor and auxiliary electrical resistance) for different combinations of (A<sub>col</sub>) and (V).</p> Full article ">Figure 13
<p>Percentage of the auxiliary electricity compared to the total electricity demand for different combinations of (A<sub>col</sub>) and (V).</p> Full article ">Figure 14
<p>System seasonal COP for different combinations of (A<sub>col</sub>) and (V).</p> Full article ">Figure 15
<p>Life cycle cost of the heating system for different combinations of (A<sub>col</sub>) and (V).</p> Full article ">Figure 16
<p>Illustration of the multi-objective optimization process of the system aiming for the maximization of the system SCOP and the minimization of the LCC.</p> Full article ">
19 pages, 997 KiB  
Article
Vehicular Edge-Computing Framework for Making Use of Parking and Charging Electric Vehicles
by Qi Deng and Feng Zeng
Appl. Sci. 2023, 13(6), 4065; https://doi.org/10.3390/app13064065 - 22 Mar 2023
Cited by 4 | Viewed by 1909
Abstract
In big cities, there are more and more parking lots and charging piles for electric vehicles, and the resources of parking and charging vehicles can be aggregated to provide strong computing power for vehicular edge computing (VEC). In this paper, we propose a [...] Read more.
In big cities, there are more and more parking lots and charging piles for electric vehicles, and the resources of parking and charging vehicles can be aggregated to provide strong computing power for vehicular edge computing (VEC). In this paper, we propose a VEC framework that uses charging vehicles in parking lots to assist edge servers in processing computational tasks, and an edge crowdsourcing platform (ECP) is designed to manage and integrate the idle computation resources of electric vehicles in parking lots to provide computation services for requesting vehicles. Based on game theory, we first model the interactions among the edge server, the ECP and the requesting vehicles as a Stackelberg game, and theoretically prove the existence of a Nash equilibrium for this Stackelberg game. Then, a genetic algorithm-based game-strategy solving algorithm is proposed to find the optimal strategy for the edge server and ECP. The simulation results demonstrate that the performance of our proposed solution is better than other traditional solutions. Compared with the solution without ECP, our solution can increase the utilities of the edge server and the requesting vehicle by 13.3% and 10.99%, respectively. Full article
(This article belongs to the Special Issue Vehicular Edge Computing and Networking)
►▼ Show Figures

Figure 1

Figure 1
<p>The SDN-based VEC framework.</p> Full article ">Figure 2
<p>Task offloading steps in parking vehicles assisted edge computing.</p> Full article ">Figure 3
<p>Impact of the processing capability and service price on the performance.</p> Full article ">Figure 4
<p>Impact of the processing capability and service price on the utility of requesting vehicle.</p> Full article ">Figure 5
<p>Performance comparison for the utility of requesting vehicle.</p> Full article ">Figure 6
<p>Impact of <math display="inline"><semantics> <msub> <mi>C</mi> <mi>v</mi> </msub> </semantics></math> on the utility of requesting vehicle.</p> Full article ">Figure 7
<p>GSA algorithm performance.</p> Full article ">Figure 8
<p>The Utilities of the edge server, ECP, and requesting vehicle.</p> Full article ">
21 pages, 14593 KiB  
Article
Research on Fault Diagnosis Algorithm of Ship Electric Propulsion Motor
by Fengxin Ma, Liang Qi, Shuxia Ye, Yuting Chen, Han Xiao and Shankai Li
Appl. Sci. 2023, 13(6), 4064; https://doi.org/10.3390/app13064064 - 22 Mar 2023
Cited by 1 | Viewed by 1603
Abstract
The permanent magnet synchronous motor (PMSM) has been used in electric propulsion and other fields. However, it is prone to the stator winding inter-turn short-circuit, and if no effective measures are taken, the ship’s power system will be paralyzed. To realize intelligent diagnosis [...] Read more.
The permanent magnet synchronous motor (PMSM) has been used in electric propulsion and other fields. However, it is prone to the stator winding inter-turn short-circuit, and if no effective measures are taken, the ship’s power system will be paralyzed. To realize intelligent diagnosis of inter-turn short circuits, this paper proposes an intelligent fault diagnosis method based on improved variational mode decomposition (VMD), multi-scale principal component analysis (PCA) feature extraction, and improved Bi-LSTM. Firstly, the stator current simulation dataset is obtained by using the mathematic model of the inter-turn short-circuit of PMSM, and the parameters of VMD are optimized by the grey wolf algorithm. Then, the data is coarse-grained to obtain multi-scale features, and the main features are selected as the sample data for fault classification by PCA. Subsequently, the Bi-LSTM neural network is used for training and analyzing the data of the sample set and the test set. Finally, the learning rate and the number of hidden-layer nodes of the Bi-LSTM are optimized by the whale algorithm to increase the diagnosis accuracy. Experimental results show that the accuracy of the proposed method for inter-turn short-circuited fault diagnosis is as high as 100%, which confirms the effectiveness of the method. Full article
(This article belongs to the Special Issue Intelligent Fault Diagnosis and Health Detection of Machinery)
►▼ Show Figures

Figure 1

Figure 1
<p>The diagnosis process of the proposed method.</p> Full article ">Figure 2
<p>PMSM equivalent model of a phase inter-turn short-circuited fault.</p> Full article ">Figure 3
<p>The structure of the LSTM model.</p> Full article ">Figure 4
<p>The structure of the Bi-LSTM network.</p> Full article ">Figure 5
<p>The time-frequency characteristic method at a scale of 3.</p> Full article ">Figure 6
<p>The flowchart of using the whale algorithm to optimize the hyperparameters of the Bi-LSTM diagnostic model.</p> Full article ">Figure 7
<p>The simulation diagram of the permanent magnet synchronous motor a phase stator winding inter-turn short-circuit.</p> Full article ">Figure 8
<p>(<b>a</b>) The waveform of the stator current signal. (<b>b</b>) The stator current. (<b>c</b>) The simulation of the stator current with a short-circuit turn ratio of 0.2.</p> Full article ">Figure 8 Cont.
<p>(<b>a</b>) The waveform of the stator current signal. (<b>b</b>) The stator current. (<b>c</b>) The simulation of the stator current with a short-circuit turn ratio of 0.2.</p> Full article ">Figure 9
<p>The accuracy in the training process. (<b>a</b>) The accuracy in the training process by unimproved VMD. (<b>b</b>) The accuracy in the training process by improved VMD.</p> Full article ">Figure 10
<p>The confusion matrix. (<b>a</b>) The confusion matrix in the training process by the unimproved VMD. (<b>b</b>) The confusion matrix in the training process by the improved VMD.</p> Full article ">Figure 11
<p>The 2D scatter plot. (<b>a</b>) The 2D scatter plot in the training process by the unimproved VMD. (<b>b</b>) The 2D scatter plot in the training process by the improved VMD.</p> Full article ">Figure 12
<p>The accuracy of the training set using the improved multiscale feature extraction.</p> Full article ">Figure 13
<p>The confusion matrix of using the improved multiscale feature extraction.</p> Full article ">Figure 14
<p>The 2D scatter plot using the improved multiscale feature extraction.</p> Full article ">Figure 15
<p>The accuracy of the training set using the improved Bi-LSTM network.</p> Full article ">Figure 16
<p>The confusion matrix using the improved Bi-LSTM network.</p> Full article ">Figure 17
<p>The 2D scatter plot using the improved Bi-LSTM network.</p> Full article ">
17 pages, 4699 KiB  
Article
Active Optics and Aberration Correction Technology for Sparse Aperture Segmented Mirrors
by Benlei Zhang, Fei Yang, Fuguo Wang and Baowei Lu
Appl. Sci. 2023, 13(6), 4063; https://doi.org/10.3390/app13064063 - 22 Mar 2023
Cited by 1 | Viewed by 1544
Abstract
Active optics and aberration correction techniques for sparse aperture segmented mirrors are studied. A finite element model of the sparse aperture segmented mirror was established, and a multi-aperture aberration polynomial was derived. According to the hard spot theorem, a co-phase maintenance method based [...] Read more.
Active optics and aberration correction techniques for sparse aperture segmented mirrors are studied. A finite element model of the sparse aperture segmented mirror was established, and a multi-aperture aberration polynomial was derived. According to the hard spot theorem, a co-phase maintenance method based on the change of the edge sensor position in the conventional mode is derived. And a co-phase maintenance method based on the change of the aberration of the segmented mirror surface without the participation of the edge sensor is proposed. The method can correct aberrations of the segmented mirror surface, which are caused by the rigid body displacement along the horizontal direction of the segments. This method can reduce the RMS of the segmented mirror surface to 2.2 nm. The correction principle of the Warping Harness (WH) technique is derived. For the problems of tedious steps and a small number of target aberrations, the correction method is proposed to directly target the aberrations of the segmented mirrors, which is simple and has a wider range of target aberrations. Using this method, the amplitude of each aberration of the stitched mirror is corrected to below 10−4nm. It is also verified that combining the generalized ridge estimation method and the differential evolution algorithm can effectively solve the correction quantity. Finally, it is verified that the SiC material can effectively improve the adaptability of the segmented mirror to gravity load by reducing the mirror’s weight. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Schematic diagram of the segment arrangement.</p> Full article ">Figure 2
<p>Schematic diagram of WH calibration.</p> Full article ">Figure 3
<p>Schematic diagram of the calibration process.</p> Full article ">Figure 4
<p>Mirror deformation under load, (<b>a</b>) Deformation of mirror surface in Y condition, (<b>b</b>) Deformation of mirror surface in Z condition, (<b>c</b>) Deformation of mirror surface in T condition.</p> Full article ">Figure 5
<p>Zerodur segment segmented mirror surface after correction by two schemes, (<b>a</b>) Scheme 1, deformation of mirror surface in Y condition, (<b>b</b>) Scheme 1, deformation of mirror surface in Z condition, (<b>c</b>) Scheme 2, deformation of mirror surface in Y condition, (<b>d</b>) Scheme 2, deformation of mirror surface in Z condition.</p> Full article ">Figure 5 Cont.
<p>Zerodur segment segmented mirror surface after correction by two schemes, (<b>a</b>) Scheme 1, deformation of mirror surface in Y condition, (<b>b</b>) Scheme 1, deformation of mirror surface in Z condition, (<b>c</b>) Scheme 2, deformation of mirror surface in Y condition, (<b>d</b>) Scheme 2, deformation of mirror surface in Z condition.</p> Full article ">Figure 6
<p>Residual deformation of the mirror surface after aberration correction for Scheme 1 (<b>top</b>) and Scheme 2 (<b>bottom</b>), (<b>a</b>) Deformation of mirror surface in Y condition, (<b>b</b>) Deformation of mirror surface in Z condition, and (<b>c</b>) Deformation of mirror surface in T condition.</p> Full article ">Figure 7
<p>Comparison of residual aberration before and after correction of two schemes, (<b>a</b>) Comparison of mirror aberration coefficients in the Y condition, (<b>b</b>) Comparison of mirror aberration coefficients in the Z condition, and (<b>c</b>) Comparison of mirror aberration coefficients in the T condition.</p> Full article ">Figure 8
<p>The mirror deformation cloud of the SiC segment after correction by the two schemes, (<b>a</b>) Scheme 1, deformation of mirror surface in Y condition, (<b>b</b>) Scheme 1, deformation of mirror surface in Z condition, (<b>c</b>) Scheme 2, deformation of mirror surface in Y condition, (<b>d</b>) Scheme 2, deformation of mirror surface in Z condition.</p> Full article ">Figure 8 Cont.
<p>The mirror deformation cloud of the SiC segment after correction by the two schemes, (<b>a</b>) Scheme 1, deformation of mirror surface in Y condition, (<b>b</b>) Scheme 1, deformation of mirror surface in Z condition, (<b>c</b>) Scheme 2, deformation of mirror surface in Y condition, (<b>d</b>) Scheme 2, deformation of mirror surface in Z condition.</p> Full article ">
20 pages, 1494 KiB  
Review
Gaucher Disease in Internal Medicine and Dentistry
by Michele Basilicata, Giulia Marrone, Manuela Di Lauro, Eleonora Sargentini, Vincenza Paolino, Redan Hassan, Giuseppe D’Amato, Patrizio Bollero and Annalisa Noce
Appl. Sci. 2023, 13(6), 4062; https://doi.org/10.3390/app13064062 - 22 Mar 2023
Viewed by 5913
Abstract
Gaucher disease (GD) is a lysosomal storage pathological condition, characterized by a genetic autosomal recessive transmission. The GD cause is the mutation of GBA1 gene, located on the chromosome 1 (1q21), that induces the deficiency of the lysosomal enzyme glucocerebrosidase with consequent abnormal [...] Read more.
Gaucher disease (GD) is a lysosomal storage pathological condition, characterized by a genetic autosomal recessive transmission. The GD cause is the mutation of GBA1 gene, located on the chromosome 1 (1q21), that induces the deficiency of the lysosomal enzyme glucocerebrosidase with consequent abnormal storage of its substrate (glucosylceramide), in macrophages. The GD incidence in the general population varies from 1:40,000 to 1:60,000 live births, but it is higher in the Ashkenazi Jewish ethnicity (1:800 live births). In the literature, five different types of GD are described: type 1, the most common clinical variant in Europe and USA (90%), affects the viscera; type 2, characterized by visceral damage and severe neurological disorders; type 3, in which the neurological manifestations are variable; cardiovascular type; and, finally, perinatal lethal type. The most affected tissues and organs are the hematopoietic system, liver, bone tissue, nervous system, lungs, cardiovascular system and kidneys. Another aspect of GD is represented by oral and dental manifestations. These can be asymptomatic or cause the spontaneous bleeding, the post oral surgery infections and the bone involvement of both arches through the Gaucher cells infiltration into the maxilla and mandibular regions. The pharmacological treatment of choice is the enzyme replacement therapy, but the new pharmacological frontiers are represented by oral substrate reduction therapy, chaperone therapy, allogeneic hematopoietic stem cell transplantation and gene therapy. Full article
(This article belongs to the Special Issue Oral Pathology and Medicine: Diagnosis and Therapy)
►▼ Show Figures

Figure 1

Figure 1
<p>Gaucher disease manifestations in internal medicine and dentistry, including pharmacological treatment.</p> Full article ">Figure 2
<p>Orthopantomografy of a 47-year-old Gaucher disease patient. The white arrows indicate the typical “soap-bubble” appearance in premolar-molar regions. The red arrows indicate the mandibular involvement.</p> Full article ">
17 pages, 2918 KiB  
Article
Illegal Domain Name Generation Algorithm Based on Character Similarity of Domain Name Structure
by Yuchen Liang, Yanan Cheng, Zhaoxin Zhang, Tingting Chai and Chao Li
Appl. Sci. 2023, 13(6), 4061; https://doi.org/10.3390/app13064061 - 22 Mar 2023
Cited by 2 | Viewed by 2013
Abstract
Detecting and controlling illegal websites (gambling and pornography sites) through illegal domain names has been an unsolved problem. Therefore, how to mine and discover potential illegal domain names in advance has become a current research hotspot. This paper studies a method of generating [...] Read more.
Detecting and controlling illegal websites (gambling and pornography sites) through illegal domain names has been an unsolved problem. Therefore, how to mine and discover potential illegal domain names in advance has become a current research hotspot. This paper studies a method of generating illegal domain names based on the character similarity of domain name structure. Firstly, the K-means algorithm classified illegal domain names with similar structures. Then, put the classified clusters into the adversarial generative network for training. Finally, through a specific result verification method, the experiment shows that the average concentration of the generation algorithm is 23.82%, the effective concentration is 63.54%, and the expansion rate is 7.5. By comparing the results with the enumeration algorithm, the generation algorithm has greatly improved in terms of generation efficiency and accuracy. Full article
(This article belongs to the Special Issue Advances in Computer Network Security and Information Retrieval Technology)
►▼ Show Figures

Figure 1

Figure 1
<p>Generation algorithms process.</p> Full article ">Figure 2
<p>SeqGAN structure.</p> Full article ">Figure 3
<p>Generator design.</p> Full article ">Figure 4
<p>Structure diagram of discriminator.</p> Full article ">Figure 5
<p>The sum of squared distances corresponding to different cluster numbers.</p> Full article ">Figure 6
<p>Classification accuracy of the discriminator.</p> Full article ">Figure 7
<p>The accuracy of the discriminator during the confrontation training process changes with the number of confrontations.</p> Full article ">Figure 8
<p>Histogram of batch number and concentration.</p> Full article ">Figure 9
<p>Histogram of batch number and effective concentration.</p> Full article ">Figure 10
<p>Changes in the number of new illegal domain names included when different numbers of domain names are generated.</p> Full article ">Figure 11
<p>Changes in the number of batches required to generate the same number of new domain names.</p> Full article ">
19 pages, 2161 KiB  
Article
An Efficient Boosting-Based Windows Malware Family Classification System Using Multi-Features Fusion
by Zhiguo Chen and Xuanyu Ren
Appl. Sci. 2023, 13(6), 4060; https://doi.org/10.3390/app13064060 - 22 Mar 2023
Cited by 8 | Viewed by 2963
Abstract
In previous years, cybercriminals have utilized various strategies to evade identification, including obfuscation, confusion, and polymorphism technology, resulting in an exponential increase in the amount of malware that poses a serious threat to computer security. The use of techniques such as code reuse, [...] Read more.
In previous years, cybercriminals have utilized various strategies to evade identification, including obfuscation, confusion, and polymorphism technology, resulting in an exponential increase in the amount of malware that poses a serious threat to computer security. The use of techniques such as code reuse, automation, etc., also makes it more arduous to identify variant software in malware families. To effectively detect the families to which malware belongs, this paper proposed and discussed a new malware fusion feature set and classification system based on the BIG2015 dataset. We used a forward feature stepwise selection technique to combine plausible binary and assembly malware features to produce new and efficient fused features. A number of machine-learning techniques, including extreme gradient boosting (XGBoost), random forest, support vector machine (SVM), K-nearest neighbors (KNN), and adaptive boosting (AdaBoost), are used to confirm the effectiveness of the fusion feature set and malware classification system. The experimental findings demonstrate that the XGBoost algorithm’s classification accuracy on the fusion feature set suggested in this paper can reach 99.87%. In addition, we applied tree-boosting-based LightGBM and CatBoost algorithms to the domain of malware classification for the first time. On our fusion feature set, the corresponding classification accuracy can reach 99.84% and 99.76%, respectively, and the F1-scores can achieve 99.66% and 99.28%, respectively. Full article
(This article belongs to the Special Issue Advances and Application of Intelligent Video Surveillance System)
►▼ Show Figures

Figure 1

Figure 1
<p>(<b>a</b>) XGBoost level-wise tree growth; (<b>b</b>) LightGBM leaf-wise tree growth.</p> Full article ">Figure 2
<p>Visualizing malware as an image.</p> Full article ">Figure 3
<p>Overview of our proposed architecture.</p> Full article ">Figure 4
<p>Dataset t-SNE 2D visualization.</p> Full article ">Figure 5
<p>Confusion matrix of fused features based on XGBoost classifier.</p> Full article ">Figure 6
<p>Our model is compared to recent research on the BIG2015 dataset [<a href="#B19-applsci-13-04060" class="html-bibr">19</a>,<a href="#B37-applsci-13-04060" class="html-bibr">37</a>,<a href="#B38-applsci-13-04060" class="html-bibr">38</a>,<a href="#B39-applsci-13-04060" class="html-bibr">39</a>,<a href="#B40-applsci-13-04060" class="html-bibr">40</a>,<a href="#B41-applsci-13-04060" class="html-bibr">41</a>].</p> Full article ">
16 pages, 1615 KiB  
Article
Comparison of Lipid Profile and Oxidative Stability of Vacuum-Packed and Longtime-Frozen Fallow Deer, Wild Boar, and Pig Meat
by Anna Reitznerová, Boris Semjon, Martin Bartkovský, Monika Šuleková, Jozef Nagy, Tatiana Klempová and Slavomír Marcinčák
Appl. Sci. 2023, 13(6), 4059; https://doi.org/10.3390/app13064059 - 22 Mar 2023
Cited by 5 | Viewed by 1536
Abstract
The present study aimed to evaluate the lipid content and oxidation of fallow deer (FD), wild boar (WB), and pig meat (PM) at −18 °C for a 360-day storage period. The lowest fat content was observed in thigh meat (TM) of FD (2.53%; [...] Read more.
The present study aimed to evaluate the lipid content and oxidation of fallow deer (FD), wild boar (WB), and pig meat (PM) at −18 °C for a 360-day storage period. The lowest fat content was observed in thigh meat (TM) of FD (2.53%; p ˂ 0.05). The ratio of polyunsaturated/saturated fatty acids (PUFA/SFA), n-6/n-3, hypocholesterolemic/hypercholesterolemic index (h/H), and the lipid nutritional quality indexes were calculated. The PUFA/SFA ratio of each meat sample was compared with the required value of more than 0.4 while the optimal n-6/n-3 ratio was determined only in shoulder meat (SM) of FD meat samples (3.94; p ˂ 0.001). An atherogenic index of lower than 1.0 was observed in each meat sample and a thrombogenic index of lower than 0.5 was observed only in TM of FD (0.53; p ˂ 0.001). During the storage period, the malondialdehyde (MDA) content of WB and PM samples showed a higher variability than the FD samples. On the initial day as well as on the 360th day of the storage period, the lowest MDA content in the loin of PM was measured. Long-term vacuum packaging resulted in lower lipid oxidation during meat storage (p ˂ 0.01); however, the duration of the storage period significantly affected the level of lipid oxidation (p ˂ 0.001). Full article
(This article belongs to the Special Issue The Influence of Oxidation Process on Shelf-Life of Meat and Meat Products)
►▼ Show Figures

Figure 1

Figure 1
<p>The combined graph of individuals with visualized meat cuts and meat type factors. LM: loin meat; SM: shoulder meat; TM: thigh meat samples.</p> Full article ">Figure 2
<p>The correlation circle of analyzed variables.</p> Full article ">
23 pages, 2889 KiB  
Article
Research on Transmission Task Static Allocation Based on Intelligence Algorithm
by Xinzhe Wang and Wenbin Yao
Appl. Sci. 2023, 13(6), 4058; https://doi.org/10.3390/app13064058 - 22 Mar 2023
Cited by 2 | Viewed by 1160
Abstract
Transmission task static allocation (TTSA) is one of the most important issues in the automatic management of radio and television stations. Different transmission tasks are allocated to the most suitable transmission equipment to achieve the overall optimal transmission effect. This study proposes a [...] Read more.
Transmission task static allocation (TTSA) is one of the most important issues in the automatic management of radio and television stations. Different transmission tasks are allocated to the most suitable transmission equipment to achieve the overall optimal transmission effect. This study proposes a TTSA mathematical model suitable for solving multiple intelligent algorithms, with the goal of achieving the highest comprehensive evaluation value, and conducts comparative testing of multiple intelligent algorithms. An improved crossover operator is proposed to solve the problem of chromosome conflicts. The operator is applied to improved genetic algorithm (IGA) and hybrid intelligent algorithms. A discrete particle swarm optimization (DPSO) algorithm is proposed, which redefines the particle position, particle movement direction, and particle movement speed for the problem itself. A particle movement update strategy based on a probability selection model is designed to ensure the search range of the DPSO, and random perturbation is designed to improve the diversity of the population. Based on simulation, comparative experiments were conducted on the proposed intelligent algorithms and the results of three aspects were compared: the success rate, convergence speed, and accuracy of the algorithm. The DPSO has the greatest advantage in solving TTSA. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Framework model for TTSA.</p> Full article ">Figure 2
<p>Process of ICO.</p> Full article ">Figure 3
<p>Process of IMO.</p> Full article ">Figure 4
<p>Selection operator and particle update operation.</p> Full article ">Figure 5
<p>Comparison chart of multi-algorithm comprehensive evaluation average.</p> Full article ">Figure 6
<p>Comparison chart of multi-algorithm success rate.</p> Full article ">Figure 7
<p>Comparison chart of multi-algorithm number of iterations.</p> Full article ">
16 pages, 5296 KiB  
Article
Acoustic Characterization and Quality Assessment of Cremona’s Ponchielli Theater
by Lamberto Tronchin, Antonella Bevilacqua and Ruoran Yan
Appl. Sci. 2023, 13(6), 4057; https://doi.org/10.3390/app13064057 - 22 Mar 2023
Cited by 19 | Viewed by 1732
Abstract
The Ponchielli theater of Cremona was built in 1808 after a fire destroyed the old wooden structure. The interior, the architecture and the shape of the plan layout are reminiscent of the Teatro alla Scala, Milan, a masterpiece by the architect Piermarini, albeit [...] Read more.
The Ponchielli theater of Cremona was built in 1808 after a fire destroyed the old wooden structure. The interior, the architecture and the shape of the plan layout are reminiscent of the Teatro alla Scala, Milan, a masterpiece by the architect Piermarini, albeit on a smaller scale. The four orders of balconies crowned by the top gallery are typical features of a 19th Century Italian Opera theater. Acoustic measurements have been undertaken across the stalls and in some selected boxes according to ISO 3382. The main acoustic parameters resulting from the measurements have been used for the acoustic calibration of a 3D model representing the Ponchielli theater. The calibration has been used to compare different scenarios involving the acoustic response of the main hall at 50% and 100% occupancy. The outcomes indicate that no significant change can be detected when the seats are provided with robust upholstery, which can be considered a positive result, especially for the actors who are not forced to change their effort between rehearsal and live performance. In order to contextualize the measured values in relation to the optimal ones, a comparison with other Italian Opera theaters provided with similar architectural characteristics has been carried out. Overall, the findings indicate that the acoustics of the Ponchielli theater are suitable for both music and speech in line with the other selected theaters, as these places were mainly created for multifunctional purposes in the 19th Century. Full article
(This article belongs to the Special Issue 5th Anniversary of Acoustics and Vibrations Section—Recent Advances in Acoustics and Vibration)
►▼ Show Figures

Figure 1

Figure 1
<p>View of the main hall from stage of Ponchielli theater of Cremona [<a href="#B7-applsci-13-04057" class="html-bibr">7</a>].</p> Full article ">Figure 2
<p>Plan layout of the Ponchielli theater of Cremona.</p> Full article ">Figure 3
<p>Details of the boxes from the main hall.</p> Full article ">Figure 4
<p>Scheme of the equipment location during the acoustic survey.</p> Full article ">Figure 5
<p>Installation of receivers (<b>a</b>) and sound source (<b>b</b>) during the survey.</p> Full article ">Figure 6
<p>Measured results of the main acoustic parameters: EDT (<b>a</b>), T<sub>20</sub> (<b>b</b>), C<sub>50</sub> (<b>d</b>), C<sub>80</sub> (<b>e</b>), and D<sub>50</sub> (<b>f</b>). Correlation between reverberation time and volume size of the theater (<b>c</b>).</p> Full article ">Figure 7
<p>Spatial distribution of speech transmission index (STI) inside the Ponchielli theater of Cremona.</p> Full article ">Figure 8
<p>Digital model of the Ponchielli theater of Cremona: external view (<b>a</b>), perspectival view of the main hall from the stage (<b>b</b>).</p> Full article ">Figure 9
<p>Acoustic calibration based on main acoustic parameters: early decay time (<b>a</b>), reverberation time (<b>b</b>), music clarity index (<b>c</b>), and definition (<b>d</b>).</p> Full article ">Figure 9 Cont.
<p>Acoustic calibration based on main acoustic parameters: early decay time (<b>a</b>), reverberation time (<b>b</b>), music clarity index (<b>c</b>), and definition (<b>d</b>).</p> Full article ">Figure 10
<p>Spatial distribution of clarity index (<b>a</b>) and definition (<b>b</b>) at 1 kHz related to the Ponchielli theater of Cremona.</p> Full article ">Figure 11
<p>Simulated results of the acoustic parameters: EDT (<b>a</b>), T<sub>20</sub> (<b>b</b>) C<sub>80</sub> (<b>c</b>) and D<sub>50</sub> (<b>d</b>).</p> Full article ">Figure 12
<p>Spatial distribution of C<sub>80</sub> (<b>a</b>) and D<sub>50</sub> (<b>b</b>) simulated at 50% and 100% occupancy.</p> Full article ">Figure 13
<p>Comparison of measured results with other Italian Opera theaters: EDT (<b>a1</b>,<b>b1</b>), T<sub>20</sub> (<b>b1</b>,<b>b2</b>) C<sub>80</sub> (<b>c1</b>,<b>c2</b>) and D<sub>50</sub> (<b>d1</b>,<b>d2</b>). Type 1 stands for results measured in the stalls. Type 2 stands for results measured in boxes.</p> Full article ">Figure 13 Cont.
<p>Comparison of measured results with other Italian Opera theaters: EDT (<b>a1</b>,<b>b1</b>), T<sub>20</sub> (<b>b1</b>,<b>b2</b>) C<sub>80</sub> (<b>c1</b>,<b>c2</b>) and D<sub>50</sub> (<b>d1</b>,<b>d2</b>). Type 1 stands for results measured in the stalls. Type 2 stands for results measured in boxes.</p> Full article ">
18 pages, 4291 KiB  
Article
CWSXLNet: A Sentiment Analysis Model Based on Chinese Word Segmentation Information Enhancement
by Shiqian Guo, Yansun Huang, Baohua Huang, Linda Yang and Cong Zhou
Appl. Sci. 2023, 13(6), 4056; https://doi.org/10.3390/app13064056 - 22 Mar 2023
Cited by 4 | Viewed by 1884
Abstract
This paper proposed a method for improving the XLNet model to address the shortcomings of segmentation algorithm for processing Chinese language, such as long sub-word lengths, long word lists and incomplete word list coverage. To address these issues, we proposed the CWSXLNet (Chinese [...] Read more.
This paper proposed a method for improving the XLNet model to address the shortcomings of segmentation algorithm for processing Chinese language, such as long sub-word lengths, long word lists and incomplete word list coverage. To address these issues, we proposed the CWSXLNet (Chinese Word Segmentation XLNet) model based on Chinese word segmentation information enhancement. The model first pre-processed Chinese pretrained text by Chinese word segmentation tool, and proposed a Chinese word segmentation attention mask mechanism by combining PLM (Permuted Language Model) and two-stream self-attention mechanism of XLNet. While performing natural language processing at word granularity, it can reduce the degree of masking between masked and non-masked words for two words belonging to the same word. For the Chinese sentiment analysis task, proposed the CWSXLNet-BiGRU-Attention model, which introduces bi-directional GRU as well as self-attention mechanism in the downstream task. Experiments show that CWSXLNet has achieved 89.91% precision, 91.53% recall rate and 90.71% F1-score, and CWSXLNet-BiGRU-Attention has achieved 92.61% precision, 93.19% recall rate and 92.90% F1-score on ChnSentiCorp dataset, which indicates that CWSXLNet has better performance than other models in Chinese sentiment analysis. Full article
(This article belongs to the Special Issue Natural Language Processing (NLP) and Applications)
►▼ Show Figures

Figure 1

Figure 1
<p>XLNet training process. (<b>a</b>) The text pre-processing process; (<b>b</b>) The pre-training process.</p> Full article ">Figure 2
<p>Structure of “input”.</p> Full article ">Figure 3
<p>An example of perm_mask matrix.</p> Full article ">Figure 4
<p>Structure of GRU unit.</p> Full article ">Figure 5
<p>Network structure of BiGRU.</p> Full article ">Figure 6
<p>Word lists and weights in Chinese XLNet.</p> Full article ">Figure 7
<p>Original text after using LTP. (<b>a</b>) The original text; (<b>b</b>) After using LTP.</p> Full article ">Figure 8
<p>Generation of vector tok_id and is_TOK.</p> Full article ">Figure 9
<p>Generation of the tok_mask matrix.</p> Full article ">Figure 10
<p>The tok_mask matrix.</p> Full article ">Figure 11
<p>Masking of perm_mask.</p> Full article ">Figure 12
<p>The attn_mask matrix, where 1 (solid circle) and 0 (hollow circle) represent masked and non-masked cases.</p> Full article ">Figure 13
<p>ChnSentiCorp dataset experimental results.</p> Full article ">Figure 14
<p>Weibo_senti_100k dataset experimental results.</p> Full article ">
27 pages, 2544 KiB  
Article
An Application of Statistical Methods in Data Mining Techniques to Predict ICT Implementation of Enterprises
by Mihalj Bakator, Dragan Cockalo, Mila Kavalić, Edit Terek Stojanović and Verica Gluvakov
Appl. Sci. 2023, 13(6), 4055; https://doi.org/10.3390/app13064055 - 22 Mar 2023
Cited by 4 | Viewed by 2113
Abstract
Globalization, Industry 4.0, and the dynamics of the modern business environment caused by the pandemic have created immense challenges for enterprises across industries. Achieving and maintaining competitiveness requires enterprises to adapt to the new business paradigm that characterizes the framework of the global [...] Read more.
Globalization, Industry 4.0, and the dynamics of the modern business environment caused by the pandemic have created immense challenges for enterprises across industries. Achieving and maintaining competitiveness requires enterprises to adapt to the new business paradigm that characterizes the framework of the global economy. In this paper, the applications of various statistical methods in data mining are presented. The sample included data from 214 enterprises. The structured survey used for the collection of data included questions regarding ICT implementation intentions within enterprises. The main goal was to present the application of statistical methods that are used in data mining, ranging from simple/basic methods to algorithms that are more complex. First, linear regression, binary logistic regression, a multicollinearity test, and a heteroscedasticity test were conducted. Next, a classifier decision tree/QUEST (Quick, Unbiased, Efficient, Statistical Tree) algorithm and a support vector machine (SVM) were presented. Finally, to provide a contrast to these classification methods, a feed-forward neural network was trained on the same dataset. The obtained results are interesting, as they demonstrate how algorithms used for data mining can provide important insight into existing relationships that are present in large datasets. These findings are significant, and they expand the current body of literature. Full article
(This article belongs to the Section Computing and Artificial Intelligence)
►▼ Show Figures

Figure 1

Figure 1
<p>Illustration of the main SVM principle for a linear problem (modified) [<a href="#B73-applsci-13-04055" class="html-bibr">73</a>].</p> Full article ">Figure 2
<p>Illustration of the SVM principle for a non-linear problem (modified) [<a href="#B75-applsci-13-04055" class="html-bibr">75</a>].</p> Full article ">Figure 3
<p>Neural network architecture schematics.</p> Full article ">Figure 4
<p>QUEST algorithm decision tree for predicting intentions on ICT solution implementation.</p> Full article ">Figure 5
<p>ROC curve for the validation samples.</p> Full article ">
19 pages, 2935 KiB  
Article
Ghost-ResNeXt: An Effective Deep Learning Based on Mature and Immature WBC Classification
by Sai Sambasiva Rao Bairaboina and Srinivasa Rao Battula
Appl. Sci. 2023, 13(6), 4054; https://doi.org/10.3390/app13064054 - 22 Mar 2023
Cited by 12 | Viewed by 5007
Abstract
White blood cells (WBCs) must be evaluated to determine how well the human immune system performs. Abnormal WBC counts may indicate malignancy, tuberculosis, severe anemia, cancer, and other serious diseases. To get an early diagnosis and to check if WBCs are abnormal or [...] Read more.
White blood cells (WBCs) must be evaluated to determine how well the human immune system performs. Abnormal WBC counts may indicate malignancy, tuberculosis, severe anemia, cancer, and other serious diseases. To get an early diagnosis and to check if WBCs are abnormal or normal, one needs to examine the numbers and determine the shape of the WBCs. To address this problem, computer-aided procedures have been developed because hematologists perform this laborious, expensive, and time-consuming process manually. Resultantly, a powerful deep learning model was developed in the present study to categorize WBCs, including immature WBCs, from the images of peripheral blood smears. A network based on W-Net, a CNN-based method for WBC classification, was developed to execute the segmentation of leukocytes. Thereafter, significant feature maps were retrieved using a deep learning framework built on GhostNet. Then, they were categorized using a ResNeXt with a Wildebeest Herd Optimization (WHO)-based method. In addition, Deep Convolutional Generative Adversarial Network (DCGAN)-based data augmentation was implemented to handle the imbalanced data issue. To validate the model performance, the proposed technique was compared with the existing techniques and achieved 99.16%, 99.24%, and 98.61% accuracy levels for Leukocyte Images for Segmentation and Classification (LISC), Blood Cell Count and Detection (BCCD), and the single-cell morphological dataset, respectively. Thus, we can conclude that the proposed approach is valuable and adaptable for blood cell microscopic analysis in clinical settings. Full article
(This article belongs to the Special Issue Deep Learning Application in Medical Image Analysis)
►▼ Show Figures

Figure 1

Figure 1
<p>System architecture of the proposed framework.</p> Full article ">Figure 2
<p>Visualization result on BCCD dataset.</p> Full article ">Figure 3
<p>Segmentation results on the BCCD dataset.</p> Full article ">Figure 4
<p>Visualization result of five types of WBC on the LISC dataset.</p> Full article ">Figure 5
<p>Segmentation results on the LISC dataset.</p> Full article ">Figure 6
<p>Visualization result on single-cell morphology dataset.</p> Full article ">Figure 7
<p>Segmentation results of various WBC subtypes on the single-cell morphology dataset.</p> Full article ">
19 pages, 9929 KiB  
Article
Boundary–Inner Disentanglement Enhanced Learning for Point Cloud Semantic Segmentation
by Lixia He, Jiangfeng She, Qiang Zhao, Xiang Wen and Yuzheng Guan
Appl. Sci. 2023, 13(6), 4053; https://doi.org/10.3390/app13064053 - 22 Mar 2023
Cited by 2 | Viewed by 1671
Abstract
In a point cloud semantic segmentation task, misclassification usually appears on the semantic boundary. A few studies have taken the boundary into consideration, but they relied on complex modules for explicit boundary prediction, which greatly increased model complexity. It is challenging to improve [...] Read more.
In a point cloud semantic segmentation task, misclassification usually appears on the semantic boundary. A few studies have taken the boundary into consideration, but they relied on complex modules for explicit boundary prediction, which greatly increased model complexity. It is challenging to improve the segmentation accuracy of points on the boundary without dependence on additional modules. For every boundary point, this paper divides its neighboring points into different collections, and then measures its entanglement with each collection. A comparison of the measurement results before and after utilizing boundary information in the semantic segmentation network showed that the boundary could enhance the disentanglement between the boundary point and its neighboring points in inner areas, thereby greatly improving the overall accuracy. Therefore, to improve the semantic segmentation accuracy of boundary points, a Boundary–Inner Disentanglement Enhanced Learning (BIDEL) framework with no need for additional modules and learning parameters is proposed, which can maximize feature distinction between the boundary point and its neighboring points in inner areas through a newly defined boundary loss function. Experiments with two classic baselines across three challenging datasets demonstrate the benefits of BIDEL for the semantic boundary. As a general framework, BIDEL can be easily adopted in many existing semantic segmentation networks. Full article
(This article belongs to the Special Issue 3D Scene Understanding and Object Recognition)
►▼ Show Figures

Figure 1

Figure 1
<p>Visualization of boundary generated from ground-truth image. Each scene was selected from S3DIS [<a href="#B13-applsci-13-04053" class="html-bibr">13</a>]. Red outlines represent boundary areas.</p> Full article ">Figure 2
<p>From left to right, (<b>a</b>) the entanglement between the boundary point and its four neighboring collections in the control group, and (<b>b</b>) the entanglement between the boundary point and its four neighboring collections in the experimental group. Each point in the figure represents a boundary point selected from a batch of input.</p> Full article ">Figure 3
<p>Detailed illustration of the Boundary–Inner Disentanglement Enhanced Learning (BIDEL) framework.</p> Full article ">Figure 4
<p>Overall architecture of segmentation network embedded within the BIDEL framework.</p> Full article ">Figure 5
<p>Visualization results on S3DIS Area 5 after applying BIDEL to KPConv [<a href="#B31-applsci-13-04053" class="html-bibr">31</a>]. Images in the first column and third column represent the input point cloud overlaid with the boundaries. Images in the second column and last column represent improved areas (blue regions were misclassified by the baseline but identified accurately by BIDEL).</p> Full article ">Figure 6
<p>Visualization results on S3DIS Area 5 after applying BIDEL to RandLA-Net. The images from left to right are (<b>a</b>) the input point cloud overlaid with the boundaries, (<b>b</b>) the ground truth, (<b>c</b>) the baseline (RandLA-Net), (<b>d</b>) the baseline + BIDEL, and (<b>e</b>) the improved areas (blue regions were misclassified by the baseline but identified accurately by BIDEL).</p> Full article ">Figure 7
<p>Visualization results on Toronto-3D dataset [<a href="#B48-applsci-13-04053" class="html-bibr">48</a>], highlighting mislabeling. The images from top to bottom are the input point cloud and the ground truth. Yellow rectangles show regions where objects were labeled wrongly in the ground truth. (<b>a</b>) Subscene-1; (<b>b</b>) Subscene-2; (<b>c</b>) Subscene-3.</p> Full article ">Figure 8
<p>Qualitative results on Toronto-3D L002 dataset. The images from top to bottom are (<b>a</b>) the input point cloud overlaid with the boundaries (red points) generated from the ground truth, (<b>b</b>) the ground truth, (<b>c</b>) the baseline (RandLA-Net), (<b>d</b>) the baseline + BIDEL, and (<b>e</b>) the improved areas (blue regions were misclassified by the baseline but identified accurately by BIDEL).</p> Full article ">Figure 8 Cont.
<p>Qualitative results on Toronto-3D L002 dataset. The images from top to bottom are (<b>a</b>) the input point cloud overlaid with the boundaries (red points) generated from the ground truth, (<b>b</b>) the ground truth, (<b>c</b>) the baseline (RandLA-Net), (<b>d</b>) the baseline + BIDEL, and (<b>e</b>) the improved areas (blue regions were misclassified by the baseline but identified accurately by BIDEL).</p> Full article ">Figure 9
<p>Visualization results on the challenging Semantic3D reduced-8 dataset [<a href="#B49-applsci-13-04053" class="html-bibr">49</a>]. The images from left to right are (<b>a</b>) the input point cloud, (<b>b</b>) the baseline (RandLA-Net), and (<b>c</b>) the baseline + BIDEL. Objects in red rectangles were misclassified by the baseline but identified accurately by BIDEL. Note that, although the ground truth of the test set was not publicly provided, the class of objects in the red rectangles could be easily recognized by human eyes with the support of RGB attributes.</p> Full article ">
17 pages, 4108 KiB  
Article
Modeling and Analysis of Contactless Solar Evaporation for Scalable Application
by Siyang Zheng, Jie Yu and Zhenyuan Xu
Appl. Sci. 2023, 13(6), 4052; https://doi.org/10.3390/app13064052 - 22 Mar 2023
Cited by 2 | Viewed by 2088
Abstract
Zero-liquid discharge wastewater treatment driven by sunlight shows potential to minimize its environmental impact by producing solid-only waste from solar energy. To overcome the key barrier of solar absorber contamination, solar-driven contactless evaporation (SCE) has been proposed. However, only a small-scale laboratory device [...] Read more.
Zero-liquid discharge wastewater treatment driven by sunlight shows potential to minimize its environmental impact by producing solid-only waste from solar energy. To overcome the key barrier of solar absorber contamination, solar-driven contactless evaporation (SCE) has been proposed. However, only a small-scale laboratory device has been studied, which cannot support its scalable application. To analyze the potential of SCE, it is essential to understand the conjugated heat and mass transfer under a scalable application scenario. In this study, a comprehensive model of SCE is developed, which is validated by the laboratory evaporation test and applied to scalable evaporation scenario. Results showed that the scalable evaporation (0.313 kg·m−2·h−1) could obtain higher evaporation rate than the laboratory evaporation (0.139 kg·m−2·h−1) due to suppressed heat losses from the sidewalls. If the design parameters are finely tuned and thermal insulation are properly applied, the evaporation rate could be further enhanced to 0.797 kg·m−2·h−1, indicating a 473.3% performance enhancement than the laboratory SCE. The modelling framework and understanding are expected to pave a way for the further improvement and scalable application of SCE. Full article
(This article belongs to the Special Issue Feature Papers in Section 'Applied Thermal Engineering')
►▼ Show Figures

Figure 1

Figure 1
<p>Schematic of the SCE. The red region in bulk water represents the penetration depth of infrared thermal radiation (~10 μm).</p> Full article ">Figure 2
<p>Schematic diagrams of (<b>a</b>) the laboratory SCE device, (<b>b</b>) the scalable SCE device and (<b>c</b>) the detailed layout of absorber–emitter of both devices.</p> Full article ">Figure 3
<p>Independence study of mesh for scalable SCE.</p> Full article ">Figure 4
<p>Schematic diagram of the test rig for laboratory SCE.</p> Full article ">Figure 5
<p>Temperatures from experiment and simulation results.</p> Full article ">Figure 6
<p>Comparison between laboratory and scalable SCE. (<b>a</b>) Evaporation rates of SCE under different hole area ratios (0.7%, 14.4% and 36.9%). (<b>b</b>) Heat-balance analysis and key temperatures comparison of two SCE processes under a hole area ratio of 14.4%.</p> Full article ">Figure 7
<p>Mechanism diagram of the downward heat transfer and the vapor diffusion for SCE.</p> Full article ">Figure 8
<p><span class="html-italic">T</span><sub>ae</sub>, <span class="html-italic">T</span><sub>wat-t</sub> and evaporation rate under different hole area ratios, with the same hole diameter (<span class="html-italic">d</span> = 1.1 mm) and air layer thickness (<span class="html-italic">δ</span> = 2.0 mm).</p> Full article ">Figure 9
<p><span class="html-italic">T</span><sub>ae</sub>, <span class="html-italic">T</span><sub>wat-t</sub> and evaporation rate under different hole diameters, with the same hole area ratio (<span class="html-italic">φ</span> = 20%) and air layer thickness (<span class="html-italic">δ</span> = 2.0 mm).</p> Full article ">Figure 10
<p>(<b>a</b>) <span class="html-italic">T</span><sub>ae</sub>, <span class="html-italic">T</span><sub>wat-t</sub> and evaporation rate under different air layer thickness, with the same hole diameter (<span class="html-italic">d</span> = 1.1 mm) and hole area ratio (<span class="html-italic">φ</span> = 14%). (<b>b</b>) The temperature difference between the water surface and infrared emitter surface and the proportion of heat flow under different air layer thickness.</p> Full article ">Figure 11
<p>Energy flows distribution of absorber–emitter and water surface during heat energy transfer from the top to bottom under the condition that the thickness of air layer is 2 mm, the hole diameter is 1.1 mm, and the hole area ratio is 14%.</p> Full article ">Figure 12
<p>Potential of foam insulation to reduce conduction heat loss. (<b>a</b>) The structure of scalable SCE adding foam insulation and the cross-sectional temperature distribution of the absorber–emitter, vapor diffusion hole, air layer and 5 mm thick water layer under different insulation layer thicknesses of 0 mm, 10 mm, 35 mm and 45 mm. (<b>b</b>) The variation trends of energy distribution ratio and evaporation rate under different insulation layer thicknesses.</p> Full article ">Figure 13
<p>Potential of a convection cover to reduce convection heat loss. (<b>a</b>) The structure of scalable SCE adding convection cover and the cross-sectional temperature distribution of the absorber–emitter, vapor diffusion hole and air layer under different convection cover thicknesses of 0 mm, 2 mm, 4 mm and 6 mm. (<b>b</b>) The energy flow distribution and evaporation rate under different convection cover thicknesses.</p> Full article ">
19 pages, 2722 KiB  
Article
Accuracy of the Sentence-BERT Semantic Search System for a Japanese Database of Closed Medical Malpractice Claims
by Naofumi Fujishiro, Yasuhiro Otaki and Shoji Kawachi
Appl. Sci. 2023, 13(6), 4051; https://doi.org/10.3390/app13064051 - 22 Mar 2023
Cited by 4 | Viewed by 2989
Abstract
In this study, we developed a similar text retrieval system using Sentence-BERT (SBERT) for our database of closed medical malpractice claims and investigated its retrieval accuracy. We assigned each case in the database a short Japanese summary of the accident as well as [...] Read more.
In this study, we developed a similar text retrieval system using Sentence-BERT (SBERT) for our database of closed medical malpractice claims and investigated its retrieval accuracy. We assigned each case in the database a short Japanese summary of the accident as well as two labels: the category was classified as a hospital department mainly, and the process indicated a failed medical procedure. We evaluated the accuracies of a similar text retrieval system with the two labels using three different multilabel evaluation metrics. For the encoders of SBERT, we employed two pretrained BERT models, UTH-BERT and NICT-BERT, that were trained on huge Japanese corpora, and we performed iterative optimization to train the SBERTs. The accuracies of the similar text retrieval systems using the trained SBERTs were more than 15 points higher than those of the Okapi BM25 system and the pretrained SBERT system. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>SBERT architectures: (<b>a</b>) the bi-encoder model and (<b>b</b>) the cross-encoder model.</p> Full article ">Figure 2
<p>A histogram of the numbers of tokens in the short summary texts of medical accidents. The purple area is where the NICT-BERT tokenizer area shown in blue overlaps the UTH-BERT tokenizer area shown in red.</p> Full article ">Figure 3
<p>The (<b>a</b>) categories and (<b>b</b>) processes that occurred most often in the dataset according to their frequency of appearance.</p> Full article ">Figure 4
<p>The SBERT iterative optimization method. Documents of the same color indicate that the labels are an exact match.</p> Full article ">Figure 5
<p>Changes in the numbers of label-matched pairs and evaluation metrics with the optimization iterations: (<b>a</b>) number of label-matched pairs, (<b>b</b>) averaged exact match ratio, (<b>c</b>) averaged Hamming score, and (<b>d</b>) averaged Hamming loss. The dots and the error bars indicate the mean and the standard deviation of the results from 10 different random seeds, respectively. To distinguish between UTH-BERT and NICT-BERT symbols, the UTH-BERT iteration number is offset by +0.1, and the NICT-BERT iteration number is offset by −0.1. The zeroth iteration shows the results from the pretrained SBERT systems.</p> Full article ">
10 pages, 528 KiB  
Article
Detecting COVID-19 Effectively with Transformers and CNN-Based Deep Learning Mechanisms
by Afamefuna Promise Umejiaku, Prastab Dhakal and Victor S. Sheng
Appl. Sci. 2023, 13(6), 4050; https://doi.org/10.3390/app13064050 - 22 Mar 2023
Cited by 3 | Viewed by 2225
Abstract
The COVID-19 pandemic has been a major global concern in the field of respiratory diseases, with healthcare institutions and partners investing significant resources to improve the detection and severity assessment of the virus. In an effort to further enhance the detection of COVID-19, [...] Read more.
The COVID-19 pandemic has been a major global concern in the field of respiratory diseases, with healthcare institutions and partners investing significant resources to improve the detection and severity assessment of the virus. In an effort to further enhance the detection of COVID-19, researchers have investigated the performance of current detection methodologies and proposed new approaches that leverage deep learning techniques. In this article, the authors propose a two-step transformer model for the multi-class classification of COVID-19 images in a patient-aware manner. This model is implemented using transfer learning, which allows for the efficient use of pre-trained models to accelerate the training of the proposed model. The authors compare the performance of their proposed model to other CNN models commonly used in the detection of COVID-19. The experimental results of the study show that CNN-based deep learning networks obtained an accuracy in the range of 0.76–0.92. However, the proposed two-step transformer model implemented with transfer learning achieved a significantly higher accuracy of 0.9735 ± 0.0051. This result indicates that the proposed model is a promising approach to improving the detection of COVID-19. Overall, the findings of this study highlight the potential of deep learning techniques, particularly the use of transfer learning and transformer models, to enhance the detection of COVID-19. These approaches can help healthcare institutions and partners to reduce the time and difficulty in detecting the virus, ultimately leading to more effective and timely treatment for patients. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Examples of COVID-19 Infected, Normal, and Pneumonia Lung Images in the Large COVID-19 CT Scan Dataset.</p> Full article ">Figure 2
<p>Images Illustrating Various Processing Steps and Outputs.</p> Full article ">Figure 3
<p>One-step Transfer Learning approach.</p> Full article ">Figure 4
<p>Two-step Transfer Learning approach.</p> Full article ">
14 pages, 4346 KiB  
Article
Power Losses Investigation in Direct 3 × 5 Matrix Converter Using MATLAB Simulink
by Michal Praženica, Slavomír Kaščák and Patrik Resutík
Appl. Sci. 2023, 13(6), 4049; https://doi.org/10.3390/app13064049 - 22 Mar 2023
Cited by 4 | Viewed by 1921
Abstract
This article addressed the problem of matrix converters (MxC), specifically the investigation of power losses and matrix converter efficiency in a 3 × 5 arrangement. In today’s modern world, efficiency is very important; hence, power loss and efficiency analysis are important throughout the [...] Read more.
This article addressed the problem of matrix converters (MxC), specifically the investigation of power losses and matrix converter efficiency in a 3 × 5 arrangement. In today’s modern world, efficiency is very important; hence, power loss and efficiency analysis are important throughout the design process of modern semiconductor converters. The ability to evaluate power losses more quickly using the simulation approach can greatly reduce the amount of time necessary for the design, in comparison with numerical analysis. The described model employed contemporary SiC semiconductors, which offer substantial benefits over IGBT transistors. The 3 × 5 converter model was shown, along with a study of power losses in various elements of the converter, such as the power circuit, input filter, and so on. A summary of the simulated findings was offered at the end of the study, along with the benefits and drawbacks of employing SiC semiconductors in bidirectional switches for matrix converters. Full article
(This article belongs to the Topic Power Electronics Converters)
►▼ Show Figures

Figure 1

Figure 1
<p>Current flow in IGBT-based bidirectional switch (<b>a</b>) Positive current flow, (<b>b</b>) Negative current flow.</p> Full article ">Figure 2
<p>Current flow in SiC-based bidirectional switch.</p> Full article ">Figure 3
<p>(<b>a</b>) Direct matrix converter, (<b>b</b>) Indirect matric converter.</p> Full article ">Figure 4
<p>Distribution of power losses.</p> Full article ">Figure 5
<p>Instantaneous power losses in bidirectional switch.</p> Full article ">Figure 6
<p>Distribution of power losses.</p> Full article ">Figure 7
<p>Efficiency comparison of IGBT simulations.</p> Full article ">Figure 8
<p>Distribution of power losses in SiC-based MxC.</p> Full article ">Figure 9
<p>Output powers of the simulation models.</p> Full article ">Figure 10
<p>Simulated efficiency of the SiC-based MxC.</p> Full article ">Figure 11
<p>Prototype of the 3 × 5 matrix converter.</p> Full article ">Figure 12
<p>Measured five phase output of the MxC prototype (phase currents—top and phase voltages—bottom).</p> Full article ">Figure 13
<p>Measured five phase output of the MxC prototype (phase currents—top and phase voltages—bottom)—Software Filtered.</p> Full article ">Figure 14
<p>Waveforms at the output of the 3 × 5 matrix converter (output phase current—blue, phase to neutral voltage—cyan, phase to phase voltage—purple).</p> Full article ">Figure 15
<p>Compared efficiency (simulation vs. measurement).</p> Full article ">Figure 16
<p>Compared input waveforms.</p> Full article ">
11 pages, 3966 KiB  
Article
Global Mechanical Response Sensing of Corrugated Compensators Based on Digital Twins
by Run Zhou, Jingyan Jiang, Jianhua Qin, Ning Du, Haoran Shi and Ying Wang
Appl. Sci. 2023, 13(6), 4048; https://doi.org/10.3390/app13064048 - 22 Mar 2023
Viewed by 1431
Abstract
The corrugated compensators are important components in the piping system, absorbing mechanical deformation flexibly. To reduce the risk of the piping system with corrugated compensators and improve the safety and stability of industrial equipment, condition monitoring and fault diagnosis of bellows is necessary. [...] Read more.
The corrugated compensators are important components in the piping system, absorbing mechanical deformation flexibly. To reduce the risk of the piping system with corrugated compensators and improve the safety and stability of industrial equipment, condition monitoring and fault diagnosis of bellows is necessary. However, the stress monitoring method of corrugated compensators with limited localized sensors lack real-time and full-domain sensing. Therefore, this paper proposes a digital twin construction method for global mechanical response sensing of corrugated compensators, combining Gaussian process regression in machine learning and finite element analysis. The sensing data of three types of displacements are used as the associated information of a finite element model with 19,800 elements and its digital twin. The results show that the values of performance metrics correlation of determination R2 and standardized average leave-one-out cross-validation CVavg of the digital twin satisfy the recommended threshold, which indicates that the digital twin has excellent predictive performance. The single prediction time of the digital twin is 0.76% of the time spent on finite element analysis, and the prediction result has good consistency with the true response under dynamic input, indicating that the digital twin can achieve fast and accurate stress field prediction. The important state information hidden in the multi-source data obtained by limited sensors is effectively mined to achieve the real-time prediction of the stress field. This paper provides a new approach for intelligent sensing and feedback of corrugated compensators in the piping system. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Movement capabilities of bellows.</p> Full article ">Figure 2
<p>FE model of the corrugated bellow.</p> Full article ">Figure 3
<p>The construction process of a digital twin for global mechanical response sensing.</p> Full article ">Figure 4
<p>Sampling results of the first simulation.</p> Full article ">Figure 5
<p>Calculated results of the performance criterion for the simulations.</p> Full article ">Figure 6
<p>Dynamic monitoring data and FEA loading curves.</p> Full article ">Figure 7
<p>Comparison of Mises stress values of DT and FEA outputs at observation points A, B, C and D.</p> Full article ">
15 pages, 1706 KiB  
Article
Integrated Analysis of Polycyclic Aromatic Hydrocarbons and Polychlorinated Biphenyls: A Comparison of the Effectiveness of Selected Methods on Dried Fruit Matrices
by Artur Ciemniak, Agata Witczak and Kamila Pokorska-Niewiada
Appl. Sci. 2023, 13(6), 4047; https://doi.org/10.3390/app13064047 - 22 Mar 2023
Viewed by 1763
Abstract
Polycyclic aromatic hydrocarbons (PAHs) and polychlorinated biphenyls (PCBs) are groups of chemical substances commonly found in the environment. Because of large differences in the concentrations of PAHs and PCBs in the materials tested, separate analytical methods specific to each group of compounds are [...] Read more.
Polycyclic aromatic hydrocarbons (PAHs) and polychlorinated biphenyls (PCBs) are groups of chemical substances commonly found in the environment. Because of large differences in the concentrations of PAHs and PCBs in the materials tested, separate analytical methods specific to each group of compounds are usually used. The aim of this study was to compare methods for the determination of PAHs and PCBs that permit the simultaneous determination of these compounds from one solvent extract. The analysis of the content of 15 PCB congeners and 16 PAHs was conducted using dried fruits. The analyses were performed with gas chromatography coupled with mass spectrometry. PAHs and PCBs were determined separately in each fruit sample using specific extraction and cleanup procedures for the respective groups of compounds. Analyses were also performed with two methods that permitted the simultaneous analysis of PAHs and PCBs in one solvent extract. The integrated methods did not provide adequate extract cleanup of interfering substances; consequently, the results of determinations of PAHs and PCBs using these methods were significantly different from the values obtained with proven determination methods for PAHs and PCBs. Full article
(This article belongs to the Special Issue Toxicity of Chemicals: Evaluation, Analysis and Impact)
►▼ Show Figures

Figure 1

Figure 1
<p>Procedures for determinations of PAHs and PCBs, according to Vives and Grimalt [<a href="#B32-applsci-13-04047" class="html-bibr">32</a>] ((<b>a</b>), method 3) and Jaouen-Madulet et al. [<a href="#B33-applsci-13-04047" class="html-bibr">33</a>] ((<b>b</b>), method 4).</p> Full article ">Figure 2
<p>Comparison of PCB contents in dried fruits analyzed with methods 1, 3, and 4. (<b>a</b>)—apricot, (<b>b</b>)—pear, (<b>c</b>)—apple; 1—determined with the specific method for PCBs; 3—determined according to the methods of Vives and Grimalt [<a href="#B32-applsci-13-04047" class="html-bibr">32</a>]; 4—determined according to the methods of Jaouen-Madoulet et al. [<a href="#B33-applsci-13-04047" class="html-bibr">33</a>].</p> Full article ">Figure 3
<p>Recovery of PCBs congeners (<b>a</b>) and PAHs (<b>b</b>) (method 3—according to Vives and Grimalt [<a href="#B32-applsci-13-04047" class="html-bibr">32</a>]; method 4—according to Jaouen-Madoulet et al. [<a href="#B33-applsci-13-04047" class="html-bibr">33</a>]).</p> Full article ">Figure 4
<p>Comparison of PHA contents in dried fruits analyzed with methods 2, 3, and 4. (<b>a</b>)—apricot, (<b>b</b>)—pear, (<b>c</b>)—apple; 2—determined with the specific method for PAHs; 3—determined according to the methods of Vives and Grimalt [<a href="#B32-applsci-13-04047" class="html-bibr">32</a>]; 4—determined according to the methods of Jaouen-Madoulet et al. [<a href="#B33-applsci-13-04047" class="html-bibr">33</a>].</p> Full article ">
20 pages, 5896 KiB  
Article
Numerical Investigation of the Relationship between Anastomosis Angle and Hemodynamics in Ridged Spiral Flow Bypass Grafts
by Jhon Jasper Apan, Lemmuel Tayo and Jaime Honra
Appl. Sci. 2023, 13(6), 4046; https://doi.org/10.3390/app13064046 - 22 Mar 2023
Cited by 3 | Viewed by 2397
Abstract
Bypass graft failures are linked to hemodynamic disturbances resulting from poor design. Several studies have tried to improve graft patency by modifying conventional graft designs. One strategy being employed is to induce spiral flow in bypass grafts using an internal ridge which has [...] Read more.
Bypass graft failures are linked to hemodynamic disturbances resulting from poor design. Several studies have tried to improve graft patency by modifying conventional graft designs. One strategy being employed is to induce spiral flow in bypass grafts using an internal ridge which has been proposed to optimize blood flow. However, there is still no study focusing on how the anastomosis angle can affect the hemodynamics of such a design despite its huge influence on local flow fields. To fill this gap, we aimed to understand and optimize the relationship between anastomosis angle and ridged spiral flow bypass graft hemodynamics to minimize disturbances and prolong graft patency. Steady-state, non-Newtonian computational fluid dynamics (CFD) analysis of a distal, end-to-side anastomosis between a ridged graft and idealized femoral artery was used to determine the anastomosis angle that would yield the least hemodynamic disturbances. Transient, pulsatile, non-Newtonian CFD analysis between a conventional and ridged graft at the optimal angle was performed to determine if such a design has an advantage over conventional designs. The results revealed that smaller anastomosis angles tend to optimize graft performance by the reduction in the pressure drop, recirculation, and areas in the host artery affected by abnormally high shear stresses. It was also confirmed that the modified design outperformed conventional bypass grafts due to the increased shear stress generated which is said to have atheroprotective benefits. The findings of the study may be taken into consideration in the design of bypass grafts to prevent their failure due to hemodynamic disturbances associated with conventional designs and highlight the importance of understanding and optimizing the relationship among different geometric properties in designing long-lasting bypass grafts. Full article
(This article belongs to the Section Fluid Science and Technology)
►▼ Show Figures

Figure 1

Figure 1
<p>Conceptualized geometric model for CFD simulation.</p> Full article ">Figure 2
<p>Geometric model used for CFD simulation. (<b>a</b>) The cross-sectional profile of the ridged bypass graft; (<b>b</b>) model dimensions based on anastomosis angle.</p> Full article ">Figure 3
<p>Hybrid mesh generated for the discretization of the computational model.</p> Full article ">Figure 4
<p>Pulsatile velocity waveform of the femoral artery for transient CFD analysis.</p> Full article ">Figure 5
<p>Pressure contours across the longitudinal section of the models.</p> Full article ">Figure 6
<p>Axial velocity contours across the longitudinal section of the models.</p> Full article ">Figure 7
<p>Secondary velocity contours and cross-flow streamlines 50 mm from the anastomosis toe.</p> Full article ">Figure 8
<p>Spatial average of WSS over the whole host artery walls.</p> Full article ">Figure 9
<p>Areas on the host artery (unrolled) affected by abnormal WSS. Only pertinent regions where abnormal WSS was observed were shown instead of the whole arterial length.</p> Full article ">Figure 10
<p>Extent of the area affected by abnormal WSS. The 30-degree anastomosis angle resulted in the total elimination of abnormally high WSS, so it is not present in the chart on the right.</p> Full article ">Figure 11
<p>Areas affected by recirculation 1 mm and 5 mm away from the anastomosis toe.</p> Full article ">Figure 12
<p>Qualitative validation of the simulation results by comparison of secondary velocity contours on the cross-section of the artery 5 mm from the distal anastomosis toe of a 60° ridged graft. (<b>a</b>) CFD result; (<b>b</b>) Doppler imaging result from Kokkalis et al. [<a href="#B50-applsci-13-04046" class="html-bibr">50</a>].</p> Full article ">Figure 13
<p>Comparison of hemodynamic performance resulting from bypass grafts of varying anastomosis angles.</p> Full article ">Figure 14
<p>Time-averaged wall shear stress contours of the artery (unrolled) using a 30° anastomosis angle with non-spiral and spiral graft configurations.</p> Full article ">Figure 15
<p>Frequency distribution of TAWSS values in the artery.</p> Full article ">
14 pages, 479 KiB  
Article
A Constrained Louvain Algorithm with a Novel Modularity
by Bibao Yao, Junfang Zhu, Peijie Ma, Kun Gao and Xuezao Ren
Appl. Sci. 2023, 13(6), 4045; https://doi.org/10.3390/app13064045 - 22 Mar 2023
Cited by 4 | Viewed by 2797
Abstract
Community detection is a significant and challenging task in network research. Nowadays, many community detection methods have been developed. Among them, the classical Louvain algorithm is an excellent method aiming at optimizing an objective function. In this paper, we propose a modularity function [...] Read more.
Community detection is a significant and challenging task in network research. Nowadays, many community detection methods have been developed. Among them, the classical Louvain algorithm is an excellent method aiming at optimizing an objective function. In this paper, we propose a modularity function F2 as a new objective function. Our modularity function F2 overcomes certain disadvantages of the modularity functions raised in previous literature, such as the resolution limit problem. It is desired as a competitive objective function. Then, the constrained Louvain algorithm is proposed by adding some constraints to the classical Louvain algorithm. Finally, through the comparison, we have found that the constrained Louvain algorithm with F2 is better than the constrained Louvain algorithm with other objective functions on most considered networks. Moreover, the constrained Louvain algorithm with F2 is superior to the classical Louvain algorithm and the Newman’s fast method. Full article
(This article belongs to the Special Issue Recent Advances in Big Data Analytics)
►▼ Show Figures

Figure 1

Figure 1
<p>Performances of different objective functions on several representative networks. (<b>a</b>) Several 3-cliques connected to a ring through single edges (Reprinted/adapted with permission from Ref. [<a href="#B36-applsci-13-04045" class="html-bibr">36</a>], Copyright (2007) National Academy of Sciences, U.S.A.). On this network, two candidate community structures are generated. One is to identify each <span class="html-italic">q</span>-clique as a single community. The other is to merge each pair of adjacent cliques into one community. The values of modularity functions <span class="html-italic">Q</span>, <span class="html-italic">M</span> and <math display="inline"><semantics> <mrow> <mi>F</mi> <mn>2</mn> </mrow> </semantics></math> for these two structures are listed as ‘single’ and ‘merge’ when the network contains 10 3-cliques in the inserted table. (<b>b</b>) A network consisting of two 6-cliques and two 3-cliques (reprinted/adapted with permission from Ref. [<a href="#B36-applsci-13-04045" class="html-bibr">36</a>], Copyright (2007) National Academy of Sciences, U.S.A.). ‘Single’ in the inserted table refers to identifying each clique as an individual community, while ‘merge’ refers to the integration of the two 3-cliques. (<b>c</b>) A well-connected network which is more compact than networks of <a href="#applsci-13-04045-f001" class="html-fig">Figure 1</a> in Ref. [<a href="#B40-applsci-13-04045" class="html-bibr">40</a>]. “single” splits the network into two communities, while “merge” identifies the whole network as one community.</p> Full article ">Figure 2
<p>For the <math display="inline"><semantics> <mrow> <mi>L</mi> <mi>F</mi> <mi>R</mi> </mrow> </semantics></math> networks, the relationship between the values of NMI and the values of each parameter. (<b>a</b>) <math display="inline"><semantics> <mi>μ</mi> </semantics></math> is changed when <math display="inline"><semantics> <mrow> <mi>γ</mi> <mo>=</mo> <mn>2.5</mn> </mrow> </semantics></math>, <math display="inline"><semantics> <mrow> <mi>β</mi> <mo>=</mo> <mn>1.5</mn> </mrow> </semantics></math> and <math display="inline"><semantics> <mrow> <mo>〈</mo> <mi>k</mi> <mo>〉</mo> <mo>=</mo> <mn>20</mn> </mrow> </semantics></math>. (<b>b</b>) <math display="inline"><semantics> <mi>γ</mi> </semantics></math> is changed when <math display="inline"><semantics> <mrow> <mi>μ</mi> <mo>=</mo> <mn>0.3</mn> </mrow> </semantics></math>, <math display="inline"><semantics> <mrow> <mi>β</mi> <mo>=</mo> <mn>1.5</mn> </mrow> </semantics></math> and <math display="inline"><semantics> <mrow> <mo>〈</mo> <mi>k</mi> <mo>〉</mo> <mo>=</mo> <mn>20</mn> </mrow> </semantics></math>. (<b>c</b>) <math display="inline"><semantics> <mi>β</mi> </semantics></math> is changed when <math display="inline"><semantics> <mrow> <mi>μ</mi> <mo>=</mo> <mn>0.3</mn> </mrow> </semantics></math>, <math display="inline"><semantics> <mrow> <mi>γ</mi> <mo>=</mo> <mn>2.5</mn> </mrow> </semantics></math> and <math display="inline"><semantics> <mrow> <mo>〈</mo> <mi>k</mi> <mo>〉</mo> <mo>=</mo> <mn>20</mn> </mrow> </semantics></math>. (<b>d</b>) <math display="inline"><semantics> <mrow> <mo>〈</mo> <mi>k</mi> <mo>〉</mo> </mrow> </semantics></math> is changed when <math display="inline"><semantics> <mrow> <mi>μ</mi> <mo>=</mo> <mn>0.3</mn> </mrow> </semantics></math>, <math display="inline"><semantics> <mrow> <mi>γ</mi> <mo>=</mo> <mn>2.5</mn> </mrow> </semantics></math> and <math display="inline"><semantics> <mrow> <mi>β</mi> <mo>=</mo> <mn>1.5</mn> </mrow> </semantics></math>. The results are averaged for 100 networks with <math display="inline"><semantics> <mrow> <mi>N</mi> <mo>=</mo> <mn>1000</mn> </mrow> </semantics></math>.</p> Full article ">Figure 3
<p>The <math display="inline"><semantics> <mrow> <mi>N</mi> <mi>M</mi> <mi>I</mi> </mrow> </semantics></math> values of the constrained Louvain algorithm with <math display="inline"><semantics> <mrow> <mi>F</mi> <mn>2</mn> </mrow> </semantics></math>, the classical Louvain algorithm with <math display="inline"><semantics> <mrow> <mi>F</mi> <mn>2</mn> </mrow> </semantics></math> and the Newman’s fast algorithm averaged on the 100 <math display="inline"><semantics> <mrow> <mi>L</mi> <mi>F</mi> <mi>R</mi> </mrow> </semantics></math> networks with <math display="inline"><semantics> <mrow> <mi>N</mi> <mo>=</mo> <mn>1000</mn> </mrow> </semantics></math>. (<b>a</b>) <math display="inline"><semantics> <mi>μ</mi> </semantics></math> is changed when <math display="inline"><semantics> <mrow> <mi>γ</mi> <mo>=</mo> <mn>2.5</mn> </mrow> </semantics></math>, <math display="inline"><semantics> <mrow> <mi>β</mi> <mo>=</mo> <mn>1.5</mn> </mrow> </semantics></math> and <math display="inline"><semantics> <mrow> <mo>〈</mo> <mi>k</mi> <mo>〉</mo> <mo>=</mo> <mn>20</mn> </mrow> </semantics></math>. (<b>b</b>) <math display="inline"><semantics> <mi>γ</mi> </semantics></math> is changed when <math display="inline"><semantics> <mrow> <mi>μ</mi> <mo>=</mo> <mn>0.3</mn> </mrow> </semantics></math>, <math display="inline"><semantics> <mrow> <mi>β</mi> <mo>=</mo> <mn>1.5</mn> </mrow> </semantics></math> and <math display="inline"><semantics> <mrow> <mo>〈</mo> <mi>k</mi> <mo>〉</mo> <mo>=</mo> <mn>20</mn> </mrow> </semantics></math>. (<b>c</b>) <math display="inline"><semantics> <mi>β</mi> </semantics></math> is changed when <math display="inline"><semantics> <mrow> <mi>μ</mi> <mo>=</mo> <mn>0.3</mn> </mrow> </semantics></math>, <math display="inline"><semantics> <mrow> <mi>γ</mi> <mo>=</mo> <mn>2.5</mn> </mrow> </semantics></math> and <math display="inline"><semantics> <mrow> <mo>〈</mo> <mi>k</mi> <mo>〉</mo> <mo>=</mo> <mn>20</mn> </mrow> </semantics></math>. (<b>d</b>) <math display="inline"><semantics> <mrow> <mo>〈</mo> <mi>k</mi> <mo>〉</mo> </mrow> </semantics></math> is changed when <math display="inline"><semantics> <mrow> <mi>μ</mi> <mo>=</mo> <mn>0.3</mn> </mrow> </semantics></math>, <math display="inline"><semantics> <mrow> <mi>γ</mi> <mo>=</mo> <mn>2.5</mn> </mrow> </semantics></math> and <math display="inline"><semantics> <mrow> <mi>β</mi> <mo>=</mo> <mn>1.5</mn> </mrow> </semantics></math>.</p> Full article ">
32 pages, 2763 KiB  
Review
Children’s Safety on YouTube: A Systematic Review
by Saeed Ibrahim Alqahtani, Wael M. S. Yafooz, Abdullah Alsaeedi, Liyakathunisa Syed and Reyadh Alluhaibi
Appl. Sci. 2023, 13(6), 4044; https://doi.org/10.3390/app13064044 - 22 Mar 2023
Cited by 8 | Viewed by 12216
Abstract
Background: With digital transformation and growing social media usage, kids spend considerable time on the web, especially watching videos on YouTube. YouTube is a source of education and entertainment media that has a significant impact on the skill improvement, knowledge, and attitudes [...] Read more.
Background: With digital transformation and growing social media usage, kids spend considerable time on the web, especially watching videos on YouTube. YouTube is a source of education and entertainment media that has a significant impact on the skill improvement, knowledge, and attitudes of children. Simultaneously, harmful and inappropriate video content has a negative impact. Recently, researchers have given much attention to these issues, which are considered important for individuals and society. The proposed methods and approaches are to limit or prevent such threats that may negatively influence kids. These can be categorized into five main directions. They are video rating, parental control applications, analysis meta-data of videos, video or audio content, and analysis of user accounts. Objective: The purpose of this study is to conduct a systematic review of the existing methods, techniques, tools, and approaches that are used to protect kids and prevent them from accessing inappropriate content on YouTube videos. Methods: This study conducts a systematic review of research papers that were published between January 2016 and December 2022 in international journals and international conferences, especially in IEEE Xplore Digital Library, ACM Digital Library, Web of Science, Google Scholar, Springer database, and ScienceDirect database. Results: The total number of collected articles was 435. The selection and filtration process reduced this to 72 research articles that were appropriate and related to the objective. In addition, the outcome answers three main identified research questions. Significance: This can be beneficial to data mining, cybersecurity researchers, and peoples’ concerns about children’s cybersecurity and safety. Full article
(This article belongs to the Special Issue Intelligent Digital Forensics and Cyber Security)
►▼ Show Figures

Figure 1

Figure 1
<p>SR methodology framework.</p> Full article ">Figure 2
<p>Number of research articles from database sources.</p> Full article ">Figure 3
<p>Research articles per country.</p> Full article ">Figure 4
<p>An overview of existing methods for preventing inappropriate content on YouTube video content.</p> Full article ">Figure 5
<p>A general view of methods used for detecting inappropriate YouTube Kids videos using machine/deep learning approaches.</p> Full article ">Figure 6
<p>Kid’s video with user comments.</p> Full article ">Figure 7
<p>Example of YouTube video meta-data.</p> Full article ">
15 pages, 2805 KiB  
Article
Development of Infrared Reflective Textiles and Simulation of Their Effect in Cold-Protection Garments
by Irina Cherunova, Nikolai Kornev, Guobin Jia, Klaus Richter and Jonathan Plentz
Appl. Sci. 2023, 13(6), 4043; https://doi.org/10.3390/app13064043 - 22 Mar 2023
Cited by 4 | Viewed by 3187
Abstract
Two ways of to enhance the heat insulation of cold-protecting garments are studied using the mathematical model, which describes the coupled transport of temperature, humidity, and bound and condensed water. The model is developed in a one-dimensional formulation. The thermal radiation transport is [...] Read more.
Two ways of to enhance the heat insulation of cold-protecting garments are studied using the mathematical model, which describes the coupled transport of temperature, humidity, and bound and condensed water. The model is developed in a one-dimensional formulation. The thermal radiation transport is explicitly considered by the subdivision of the heat flux into radiative and conduction parts. The model is utilized to study the improvement of heat-insulating properties of cold protective garments using aerogel materials and thin infrared reflective textile layers. Special attention is paid to the technological aspects of manufacturing such reflective textiles. The numerical investigations show that the use of infrared reflective textiles is the most effective of the two studied methods. Due to the reflection of the radiant heat flow coming from the human body, the skin temperature rises and the thermal insulation of clothing is significantly improved. Full article
(This article belongs to the Special Issue Artificial Vision Systems for Industrial and Textile Control)
►▼ Show Figures

Figure 1

Figure 1
<p>Micrograph images of copper-plated textiles NL-VL-S-016 (<b>A</b>), NL-WE-S-056 (<b>B</b>) and NL-WE-S-045 (<b>C</b>) and the corresponding textiles with <math display="inline"><semantics> <mrow> <mi>N</mi> <mi>i</mi> </mrow> </semantics></math> plating in (<b>a</b>–<b>c</b>).</p> Full article ">Figure 2
<p>SEM images of copper-plated textiles consisting of 55% of viscose and 45% polyester (<b>A</b>), 100% polyestersulfone (<b>B</b>), a taffeta lined and calendered polyamide textile (<b>C</b>) and the corresponding fabrics (A-a, B-b and C-c) with Ni plating in (<b>a</b>–<b>c</b>). Scale bar for all the images: 100 <math display="inline"><semantics> <mi mathvariant="sans-serif">μ</mi> </semantics></math>m.</p> Full article ">Figure 3
<p>High magnification SEM images at the cross-section prepared by focused ion beam (FIB) for the determination of the coated layer thickness. Fabrics produced by 55% of viscose and 45% polyester (<b>A</b>), by 100% polyestersulfone (<b>B</b>), a taffeta lined and calendered polyamide textile (<b>C</b>) coated with copper, respectively. The corresponding fabrics coated with nickel are presented in (<b>a</b>–<b>c</b>), respectively. Scale bar for all samples: 2 <math display="inline"><semantics> <mi mathvariant="sans-serif">μ</mi> </semantics></math>m.</p> Full article ">Figure 4
<p>EDX measurements on the <math display="inline"><semantics> <mrow> <mi>C</mi> <mi>u</mi> </mrow> </semantics></math> (<b>left</b>) and <math display="inline"><semantics> <mrow> <mi>N</mi> <mi>i</mi> </mrow> </semantics></math> (<b>right</b>) layer deposited on textiles. Measurements are carried out at an electron beam energy of 10 keV.</p> Full article ">Figure 5
<p>Sketch of the garment geometry. <math display="inline"><semantics> <msub> <mi>ζ</mi> <mi>i</mi> </msub> </semantics></math> are emissivities. Circles are nodes of the computational grid.</p> Full article ">Figure 6
<p>Simulation of the temperature and comparison with measurements [<a href="#B20-applsci-13-04043" class="html-bibr">20</a>].</p> Full article ">Figure 7
<p>Simulation of the temperature and comparison with measurements [<a href="#B21-applsci-13-04043" class="html-bibr">21</a>].</p> Full article ">Figure 8
<p>Influence of the IR on the temperature at the boundary between the skin and the undershirt after 120 min. Emissivity of the IR <math display="inline"><semantics> <msub> <mi>ζ</mi> <mn>2</mn> </msub> </semantics></math> varies between <math display="inline"><semantics> <mrow> <mn>0.1</mn> </mrow> </semantics></math> and <math display="inline"><semantics> <mrow> <mn>0.9</mn> </mrow> </semantics></math>.</p> Full article ">Figure 9
<p>Influence of the IR on the temperature (<b>left</b>) and the vapor concentration (<b>right</b>) distribution across the garment after 120 min. Polyester batting thickness is <math display="inline"><semantics> <mrow> <mn>2.0</mn> </mrow> </semantics></math> cm.</p> Full article ">Figure 10
<p>Influence of the IR on the bound water content in fibers.</p> Full article ">Figure 11
<p>Influence of the volume fraction of aerogel <math display="inline"><semantics> <msub> <mi>ε</mi> <mrow> <mi>d</mi> <mi>s</mi> </mrow> </msub> </semantics></math> on the effective density and coefficient of thermal conductivity in the center of the insulating layer (batting) of clothing.</p> Full article ">Figure 12
<p>Influence of the volume fraction of aerogel <math display="inline"><semantics> <msub> <mi>ε</mi> <mrow> <mi>d</mi> <mi>s</mi> </mrow> </msub> </semantics></math> on the temperature on the human skin at different thicknesses of the insulating layer, taking into account (solid lines) and without taking into account (dotted lines) the transport of humidity inside clothing.</p> Full article ">Figure 13
<p>Influence of the volume fraction of aerogel <math display="inline"><semantics> <msub> <mi>ε</mi> <mrow> <mi>d</mi> <mi>s</mi> </mrow> </msub> </semantics></math> on the skin temperature at different <math display="inline"><semantics> <msub> <mi>ζ</mi> <mn>2</mn> </msub> </semantics></math> and insulation layer thickness.</p> Full article ">
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-699
Previous Issue
Next Issue
Back to TopTop