[go: up one dir, main page]
 
 
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
Special Issues
Applications of Advanced Deep Learning Technology in Control and Intelligent...
applsci-logo
Submit to Special Issue Submit Abstract to Special Issue Review for Applied Sciences Propose a Special Issue

Journal Menu

► Journal Menu

Journal Browser

► Journal Browser
share announcement

Applications of Advanced Deep Learning Technology in Control and Intelligent Systems

A special issue of Applied Sciences (ISSN 2076-3417). This special issue belongs to the section "Computing and Artificial Intelligence".

Deadline for manuscript submissions: 30 April 2025 | Viewed by 7130

Special Issue Editors

Dr. Xuefeng Zhang

E-Mail Website
Guest Editor
College of Sciences, Northeastern University, Shenyang 110819, China
Interests: singular systems; fractional systems; robust control
Special Issues, Collections and Topics in MDPI journals
Prof. Dr. Jing Zhao

E-Mail Website
Guest Editor
School of Science, Northeastern University, Shenyang 110819, China
Interests: intelligent control; dynamics and control; mechanism and machine theory; autonoumous system; fault tolerant control; artificial intelligence with engineering applications; machine learning methods; signal processing; intelligent transportation; system modeling and identification
Dr. Jin-Xi Zhang

E-Mail
Guest Editor
State Key Laboratory of Synthetical Automation for Process Industries, Northeastern University, Shenyang 110819, China
Interests: fractional-order systems; nonlinear systems; multi-agent systems; prescribed performance control; nonlinear control
Special Issues, Collections and Topics in MDPI journals
Prof. Dr. Driss Boutat

E-Mail
Guest Editor
Institut National des Sciences Appliquées Centre Val de Loire, Bourges, France
Interests: non-asymptotic state estimation; non-asymptotic fractional order differentiator
Special Issues, Collections and Topics in MDPI journals
Dr. Da-Yan Liu

E-Mail
Guest Editor
INSA Centre Val de Loire, Université d’Orléans, PRISME EA 4229, CEDEX, 18022 Bourges, France
Interests: estimation and control for fractional order systems; numerical solutions for fractional order differential equations
Special Issues, Collections and Topics in MDPI journals

Special Issue Information

Dear Colleagues,

In recent years, numerous authors from diverse science and engineering fields have explored dynamical systems using advanced deep learning algorithms and fractional differential operators, leading to the proposal of many computational fractional intelligence systems and their reasoning applications. This Special Issue aims to provide an international platform for researchers to contribute original research focusing on the integration of mathematical ideas with optimal neural network algorithms and fractional operators. Interdisciplinary studies encompass theoretical frameworks, computational algorithm development, and applications in mechatronic systems and artificial intelligence. Fractional-order systems, which extend classical integer-order systems, accurately describe real-world physical phenomena. Constructing computational neuronal network models is essential for conducting experiments, either on computers or silicon chips, particularly in exploring virtual brain scenarios. Control systems derive significant benefits from artificial neural networks, facilitating the creation of intelligent interfaces and the storage of imprecise linguistic information. This intersects closely with computational intelligence, including neural networks and genetic and evolutionary algorithms. The exploration of advanced learner models and training approaches has demonstrated growing potential for industrial applications such as data modeling and predictive analytics. Additionally, the combination of powerful fractional operators and optimal algorithms exhibits promise for effectively analyzing and designing nonlinear and complex control systems, thus advancing control engineering. The interdisciplinary topics covered include control theory, fractional calculus, and the diverse applications of neural networks in intelligence systems.

Dr. Xuefeng Zhang
Prof. Dr. Jing Zhao
Dr. Jinxi Zhang
Prof. Dr. Driss Boutat
Dr. Dayan Liu
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Applied Sciences is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • fractional-order systems
  • deep learning strategies
  • multi-agent systems
  • prescribed performance control
  • rough set and fuzzy set reasoning
  • genetic algorithms and modelling
  • machine learning
  • recurrent neural networks
  • image processing and computer vision systems.
  • time series forecasting

Benefits of Publishing in a Special Issue

  • Ease of navigation: Grouping papers by topic helps scholars navigate broad scope journals more efficiently.
  • Greater discoverability: Special Issues support the reach and impact of scientific research. Articles in Special Issues are more discoverable and cited more frequently.
  • Expansion of research network: Special Issues facilitate connections among authors, fostering scientific collaborations.
  • External promotion: Articles in Special Issues are often promoted through the journal's social media, increasing their visibility.
  • e-Book format: Special Issues with more than 10 articles can be published as dedicated e-books, ensuring wide and rapid dissemination.

Further information on MDPI's Special Issue policies can be found here.

Published Papers (6 papers)

Download All Papers
Order results
Result details
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:

Research

14 pages, 3164 KiB  
Article
A Local Discrete Feature Histogram for Point Cloud Feature Representation
by Linjing Jia, Cong Li, Guan Xi, Xuelian Liu, Da Xie and Chunyang Wang
Appl. Sci. 2025, 15(5), 2367; https://doi.org/10.3390/app15052367 - 22 Feb 2025
Viewed by 373
Abstract
Local feature descriptors are a critical problem in computer vision; the majority of current approaches find it difficult to achieve a balance between descriptiveness, robustness, compactness, and efficiency. This paper proposes the local discrete feature histogram (LDFH), a novel local feature descriptor, as [...] Read more.
Local feature descriptors are a critical problem in computer vision; the majority of current approaches find it difficult to achieve a balance between descriptiveness, robustness, compactness, and efficiency. This paper proposes the local discrete feature histogram (LDFH), a novel local feature descriptor, as a solution to this problem. The LDFH descriptor is constructed based on a robust local reference frame (LRF). It partitions the local space based on radial distance and calculates three geometric features, including the normal deviation angle, polar angle, and normal lateral angle, in each subspace. These features are then discretized to generate three feature statistical histograms, which are combined using a weighted fusion strategy to generate the final LDFH descriptor. Experiments on public datasets demonstrate that, compared with the existing methods, LDFH strikes an excellent balance between descriptiveness, robustness, compactness, and efficiency, making it suitable for various scenes and sensor datasets. Full article
(This article belongs to the Special Issue Applications of Advanced Deep Learning Technology in Control and Intelligent Systems)
►▼ Show Figures

Figure 1

Figure 1
<p>An illustration of the LDFH descriptor (<b>a</b>) 3D model. (<b>b</b>) Extracting the local surface (blue) around the key point (red). (<b>c</b>) LRF construction at the key point. (<b>d</b>) Dividing the local space along radial distance (for clarity, we set 4 partitions along the radial distance). (<b>e</b>) Calculating three geometric attributes in each subspace. (<b>f</b>) Generating three feature statistical histograms. (<b>g</b>) Generating weighted feature histograms.</p> Full article ">Figure 2
<p>The parameter settings for LDFH descriptor. The solid markers present the selected parameter values.</p> Full article ">Figure 3
<p>Experimental datasets: Two model examples and two scene examples are displayed from left to right.</p> Full article ">Figure 3 Cont.
<p>Experimental datasets: Two model examples and two scene examples are displayed from left to right.</p> Full article ">Figure 4
<p>The RPC results on the B3R dataset.</p> Full article ">Figure 4 Cont.
<p>The RPC results on the B3R dataset.</p> Full article ">Figure 5
<p>The RPC results on the Kinect dataset.</p> Full article ">Figure 6
<p>The compactness of selected descriptors.</p> Full article ">Figure 7
<p>The average computation time of descriptors under different support radii.</p> Full article ">
19 pages, 3146 KiB  
Article
Prediction of Ship-Unloading Time Using Neural Networks
by Zhen Gao, Danning Li, Danni Wang, Zengcai Yu, Witold Pedrycz and Xinhai Wang
Appl. Sci. 2024, 14(18), 8213; https://doi.org/10.3390/app14188213 - 12 Sep 2024
Viewed by 862
Abstract
The prediction of unloading times is crucial for reducing demurrage costs and ensuring the smooth scheduling of downstream processes in a steel plant. The duration of unloading a cargo ship is primarily determined by the unloading schedule established at the raw materials terminal [...] Read more.
The prediction of unloading times is crucial for reducing demurrage costs and ensuring the smooth scheduling of downstream processes in a steel plant. The duration of unloading a cargo ship is primarily determined by the unloading schedule established at the raw materials terminal and the storage operation schedule implemented in the stockyard. This study aims to provide an accurate forecast of unloading times for incoming ships at the raw materials terminal of a steel plant. We propose three neural network-based methods: the Backpropagation Neural Network (BP), the Random Vector Functional Link (RVFL), and the Stochastic Configurations Network (SCN) for this prediction. This issue has not been previously researched using similar methods, particularly in the context of large-scale steel plants. The performance of these three methods is evaluated based on several indices: the Root Mean Square Error (RMSE), the quality of the best solution, convergence, and stability, which are employed for predicting unloading times. The prediction accuracies achieved by the BP, RVFL, and SCN were 76%, 85%, and 87%, respectively. These results demonstrate the effectiveness and potential applications of the proposed methods. Full article
(This article belongs to the Special Issue Applications of Advanced Deep Learning Technology in Control and Intelligent Systems)
►▼ Show Figures

Figure 1

Figure 1
<p>The ship-unloading system of a steel plant.</p> Full article ">Figure 2
<p>Fitting curves on DB3.</p> Full article ">Figure 3
<p>Fitting curves on DB4.</p> Full article ">Figure 4
<p>Fitting curves on DB1.</p> Full article ">Figure 5
<p>Fitting curves on DB2.</p> Full article ">Figure 6
<p>Loss curve of training (DB1).</p> Full article ">Figure 7
<p>Loss curve of training (DB2).</p> Full article ">Figure 8
<p>Loss curve of training (DB3).</p> Full article ">Figure 9
<p>Loss curve of training (DB4).</p> Full article ">Figure 10
<p>RMSE within 100 trials (DB1).</p> Full article ">Figure 11
<p>RMSE within 100 trials (DB2).</p> Full article ">Figure 12
<p>RMSE within 100 trials (DB3).</p> Full article ">Figure 13
<p>RMSE within 100 trials (DB4).</p> Full article ">Figure 14
<p>Unloading time predictions for 30, 50, and 100 ships.</p> Full article ">Figure 14 Cont.
<p>Unloading time predictions for 30, 50, and 100 ships.</p> Full article ">
15 pages, 881 KiB  
Article
Lagrange Relaxation for the Capacitated Multi-Item Lot-Sizing Problem
by Zhen Gao, Danning Li, Danni Wang and Zengcai Yu
Appl. Sci. 2024, 14(15), 6517; https://doi.org/10.3390/app14156517 - 25 Jul 2024
Cited by 1 | Viewed by 996
Abstract
The capacitated multi-item lot-sizing problem, referred to as the CLSP, is to determine the lot sizes of products in each period in a given planning horizon of finite periods, meeting the product demands and resource limits in each period, and to minimize the [...] Read more.
The capacitated multi-item lot-sizing problem, referred to as the CLSP, is to determine the lot sizes of products in each period in a given planning horizon of finite periods, meeting the product demands and resource limits in each period, and to minimize the total cost, consisting of the production, inventory holding, and setup costs. CLSPs are often encountered in industry production settings and they are considered NP-hard. In this paper, we propose a Lagrange relaxation (LR) approach for their solution. This approach relaxes the capacity constraints to the objective function and thus decomposes the CLSP into several uncapacitated single-item problems, each of which can be easily solved by dynamic programming. Feasible solutions are achieved by solving the resulting transportation problems and a fixup heuristic. The Lagrange multipliers in the relaxed problem are updated by using subgradient optimization. The experimental results show that the LR approach explores high-quality solutions and has better applicability compared with other commonly used solution approaches in the literature. Full article
(This article belongs to the Special Issue Applications of Advanced Deep Learning Technology in Control and Intelligent Systems)
►▼ Show Figures

Figure 1

Figure 1
<p>A 2-opt lot exchange for solution improvement.</p> Full article ">Figure 2
<p>Computational results comparison.</p> Full article ">
17 pages, 4413 KiB  
Article
Super-Resolution Reconstruction of an Array Lidar Range Profile
by Xuelian Liu, Xulang Zhou, Guan Xi, Rui Zhuang, Chunhao Shi and Chunyang Wang
Appl. Sci. 2024, 14(12), 5335; https://doi.org/10.3390/app14125335 - 20 Jun 2024
Viewed by 974
Abstract
Aiming at the problem that the range profile of the current array lidar has a low resolution and contains few target details and little edge information, a super-resolution reconstruction method based on projection onto convex sets (POCS) combining the Lucas–Kanade (LK) optical flow [...] Read more.
Aiming at the problem that the range profile of the current array lidar has a low resolution and contains few target details and little edge information, a super-resolution reconstruction method based on projection onto convex sets (POCS) combining the Lucas–Kanade (LK) optical flow method with a Gaussian pyramid was proposed. Firstly, the reference high-resolution range profile was obtained by the nearest neighbor interpolation of the single low-resolution range profile. Secondly, the LK optical flow method was introduced to achieve the motion estimation of low-resolution image sequences, and the Gaussian pyramid was used to perform multi-scale correction on the estimated vector, effectively improving the accuracy of motion estimation. On the basis of data consistency constraints, gradient constraints were introduced based on the distance value difference between the target edge and the background to enhance the reconstruction ability of the target edge. Finally, the residual between the estimated distance and the actual distance was calculated, and the high-resolution reference range profile was iteratively corrected by using the point spread function according to the residual. Bilinear interpolation, bicubic interpolation, POCS, POCS with adaptive correction threshold, and the proposed method were used to reconstruct the range profile of the datasets and the real datasets. The effectiveness of the proposed method was verified by the range profile reconstruction effect and objective evaluation index. The experimental results show that the index of the proposed method is improved compared to the interpolation method and the POCS method. In the redwood-3dscan dataset experiments, compared to the traditional POCS, the average gradient (AG) of the proposed method is increased by at least 8.04%, and the edge strength (ES) is increased by at least 4.84%. In the real data experiments, compared to the traditional POCS, the AG of the proposed method is increased by at least 5.85%, and the ES is increased by at least 7.01%, which proves that the proposed method can effectively improve the resolution of the reconstructed range map and the quality of the detail edges. Full article
(This article belongs to the Special Issue Applications of Advanced Deep Learning Technology in Control and Intelligent Systems)
►▼ Show Figures

Figure 1

Figure 1
<p>Gaussian pyramid optical flow iteration map.</p> Full article ">Figure 2
<p>Flowchart of super-resolution reconstruction based on POCS with an LK Gaussian pyramid.</p> Full article ">Figure 3
<p>Super-resolution rendering of the sofa range profile with (<b>a</b>) the original sofa range profile; (<b>b</b>) bilinear sofa range profile; (<b>c</b>) bicubic sofa range profile; (<b>d</b>) POCS sofa range profile; (<b>e</b>) POCS [<a href="#B17-applsci-14-05335" class="html-bibr">17</a>] sofa range profile; (<b>f</b>) POCS with the LK sofa range profile; (<b>g</b>) POCS with gradient sofa range profile; and (<b>h</b>) the proposed sofa range profile.</p> Full article ">Figure 4
<p>Super-resolution rendering of the car range profile with (<b>a</b>) the original car range profile; (<b>b</b>) bilinear car range profile; (<b>c</b>) bicubic car range profile; (<b>d</b>) POCS car range profile; (<b>e</b>) POCS [<a href="#B17-applsci-14-05335" class="html-bibr">17</a>] car range profile; (<b>f</b>) POCS with the LK car range profile; (<b>g</b>) POCS with gradient car range profile; and (<b>h</b>) the proposed car range profile.</p> Full article ">Figure 5
<p>(<b>a</b>) Schematic diagram of the GM-APD lidar system. (<b>b</b>) Schematic diagram of the GM-APD lidar range profile acquisition.</p> Full article ">Figure 6
<p>Super-resolution rendering of the tank model range profile with (<b>a</b>) the original tank model range profile; (<b>b</b>) bilinear tank model range profile; (<b>c</b>) bicubic tank model range profile; (<b>d</b>) POCS tank model range profile; (<b>e</b>) POCS [<a href="#B17-applsci-14-05335" class="html-bibr">17</a>] tank model range profile; and (<b>f</b>) the proposed tank model range profile.</p> Full article ">Figure 7
<p>Super-resolution rendering of the armored car model range profile with (<b>a</b>) the original armored car model range profile; (<b>b</b>) bilinear armored car model profile; (<b>c</b>) bicubic armored car model range profile; (<b>d</b>) POCS armored car model range profile; (<b>e</b>) POCS [<a href="#B17-applsci-14-05335" class="html-bibr">17</a>] armored car model range profile; and (<b>f</b>) the proposed armored car model range profile.</p> Full article ">Figure 8
<p>Super-resolution rendering of the outdoor scene range profile with (<b>a</b>) the original outdoor scene range profile; (<b>b</b>) bilinear outdoor scene range profile; (<b>c</b>) bicubic outdoor scene range profile; (<b>d</b>) POCS outdoor scene range profile; (<b>e</b>) POCS [<a href="#B17-applsci-14-05335" class="html-bibr">17</a>] outdoor scene range profile; and (<b>f</b>) the proposed outdoor scene range profile.</p> Full article ">
10 pages, 480 KiB  
Article
Raw Material Purchasing Optimization Using Column Generation
by Zhen Gao, Danning Li, Danni Wang and Zengcai Yu
Appl. Sci. 2024, 14(11), 4375; https://doi.org/10.3390/app14114375 - 22 May 2024
Viewed by 1509
Abstract
The raw material purchasing (RMP) problem is to determine the purchasing quantities of raw materials in given periods or cycles. Raw material purchasing optimization is crucial for large-scale steel plants because it is closely related to the supply of raw materials and cost [...] Read more.
The raw material purchasing (RMP) problem is to determine the purchasing quantities of raw materials in given periods or cycles. Raw material purchasing optimization is crucial for large-scale steel plants because it is closely related to the supply of raw materials and cost savings. The raw material purchasing of large-scale steel enterprises is characterized by many varieties, large quantities, and high costs. The RMP objective is to minimize the total purchasing cost, consisting of the price of raw materials, purchasing set-up costs, and inventory costs, and meet product demand. We present a mixed integer linear programming (MILP) model and a column generation (CG) solution for the resulting optimization problem. The column generation algorithm is the generalization of the branch and bound algorithm while solving the linear programming (LP) relaxation of MILP using column generation (CG), and its advantage is to handle large-sized MILPs. Experimental results show the effectiveness and efficiency of the solution. Full article
(This article belongs to the Special Issue Applications of Advanced Deep Learning Technology in Control and Intelligent Systems)
18 pages, 4592 KiB  
Article
Text Triplet Extraction Algorithm with Fused Graph Neural Networks and Improved Biaffine Attention Mechanism
by Yinghao Piao and Jin-Xi Zhang
Appl. Sci. 2024, 14(8), 3524; https://doi.org/10.3390/app14083524 - 22 Apr 2024
Cited by 2 | Viewed by 1232
Abstract
In the realm of aspect-based sentiment analysis (ABSA), a paramount task is the extraction of triplets, which define aspect terms, opinion terms, and their respective sentiment orientations within text. This study introduces a novel extraction model, BiLSTM-BGAT-GCN, which seamlessly integrates graph neural networks [...] Read more.
In the realm of aspect-based sentiment analysis (ABSA), a paramount task is the extraction of triplets, which define aspect terms, opinion terms, and their respective sentiment orientations within text. This study introduces a novel extraction model, BiLSTM-BGAT-GCN, which seamlessly integrates graph neural networks with an enhanced biaffine attention mechanism. This model amalgamates the sophisticated capabilities of both graph attention and convolutional networks to process graph-structured data, substantially enhancing the interpretation and extraction of textual features. By optimizing the biaffine attention mechanism, the model adeptly uncovers the subtle interplay between aspect terms and emotional expressions, offering enhanced flexibility and superior contextual analysis through dynamic weight distribution. A series of comparative experiments confirm the model’s significant performance improvements across various metrics, underscoring its efficacy and refined effectiveness in ABSA tasks. Full article
(This article belongs to the Special Issue Applications of Advanced Deep Learning Technology in Control and Intelligent Systems)
►▼ Show Figures

Figure 1

Figure 1
<p>The GCN architecture.</p> Full article ">Figure 2
<p>GAT with self attention.</p> Full article ">Figure 3
<p>GAT with multi-head attention.</p> Full article ">Figure 4
<p>The Biaffine attention architecture.</p> Full article ">Figure 5
<p>The BiLSTM architecture.</p> Full article ">Figure 6
<p>The architecture of the proposed BiLSTM-BGAT-GCN model.</p> Full article ">Figure 7
<p>A grid marking of a sentence.</p> Full article ">Figure 8
<p>Four types of features for a sentence.</p> Full article ">Figure 9
<p>The results on <math display="inline"><semantics> <mrow> <mi>D</mi> <mn>1</mn> </mrow> </semantics></math> and <math display="inline"><semantics> <mrow> <mi>D</mi> <mn>2</mn> </mrow> </semantics></math> datasets.</p> Full article ">Figure 10
<p>Comparison results on <math display="inline"><semantics> <mrow> <mi>D</mi> <mn>1</mn> </mrow> </semantics></math> and <math display="inline"><semantics> <mrow> <mi>D</mi> <mn>2</mn> </mrow> </semantics></math> datasets.</p> Full article ">Figure 11
<p>Attention distribution on word.</p> Full article ">
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-6
Back to TopTop