[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
You seem to have javascript disabled. Please note that many of the page functionalities won't work as expected without javascript enabled.
 
 
Sign in to use this feature.

Years

Between: -

Subjects

remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline

Journals

Article Types

Countries / Regions

Search Results (55)

Search Parameters:
Keywords = sensor placement identification

Order results
Result details
Results per page
Select all
Export citation of selected articles as:
10 pages, 3161 KiB  
Communication
Design of an Accurate, Planar, Resonant Microwave Sensor for Testing a Wide Range of Liquid Samples
by Smriti Agarwal and Manoj Chandra Garg
Electronics 2024, 13(22), 4510; https://doi.org/10.3390/electronics13224510 - 17 Nov 2024
Viewed by 535
Abstract
In this paper, an inductively coupled capacitively loaded ring resonator (IC-CLRR)-based microwave resonant sensor has been proposed for the accurate identification of any unknown liquid sample and its permittivity estimation. The key element of this work is the sensor’s wide range capability towards [...] Read more.
In this paper, an inductively coupled capacitively loaded ring resonator (IC-CLRR)-based microwave resonant sensor has been proposed for the accurate identification of any unknown liquid sample and its permittivity estimation. The key element of this work is the sensor’s wide range capability towards the non-invasive testing of liquids covering a wide dielectric range of liquid samples, i.e., εr = 2 to 80. The proposed microwave sensor is etched over the FR-4 substrate and is excited by the microstrip line through inductive coupling. The placement of an unknown liquid sample in close proximity to the sensor alters its natural resonant frequency due to a change in effective inductance and capacitance as per the dielectric property of the liquid sample. Further, a mathematical formulation using curve fitting has also been derived. The measurement results show a good accuracy in estimating the permittivity and, thus, the unknown liquid identification capability of the designed sensor with a very low error (nearly 5%). This sensor design is simple to fabricate, cost-friendly, and small in size. Full article
Show Figures

Figure 1

Figure 1
<p>(<b>a</b>) Top view of the proposed inductively coupled capacitively loaded ring resonator (IC-CLRR) microwave resonant sensor, (<b>b</b>) layered perspective view of each plane, and (<b>c</b>) E-field distribution at the resonant frequency (<span class="html-italic">f<sub>r</sub></span> = 3.56 GHz).</p>
Full article ">Figure 2
<p>Demonstration of (<b>a</b>) empty vessel made up of borosilicate glass placed over the IC-CLRR sensor and (<b>b</b>) liquid sample inside the glass vessel placed over the sensor.</p>
Full article ">Figure 3
<p>Scattering parameter (S21) vs. frequency plot: resonant frequency of the sensor under unloaded conditions at 3.56 GHz and under loaded conditions (with the glass vessel) at 3.51 GHz.</p>
Full article ">Figure 4
<p>Scattering parameter (S21) vs. frequency graph for liquid samples with different permittivity.</p>
Full article ">Figure 5
<p>Graph showing permittivity vs. resonant frequency variation (blue curve) and permittivity vs. frequency shift variation (red curve) for liquid samples of varying permittivity (<span class="html-italic">ε<sub>r</sub></span> = 2 to 80).</p>
Full article ">Figure 6
<p>(<b>a</b>) Measurement setup for the proposed IC-CLRR resonant microwave sensor (<b>b</b>) S21 vs. frequency plot (measurement vs. simulated).</p>
Full article ">Figure 7
<p>Measured S21 vs. frequency plot for sample liquids under test, viz., coconut oil, vinegar, methanol, and water.</p>
Full article ">Figure 8
<p>Equivalent circuit modeling of the proposed sensor: (<b>a</b>) lumped circuit schematic and (<b>b</b>) S21 vs. frequency response of the ECM model.</p>
Full article ">
19 pages, 6430 KiB  
Article
An Ensemble Deep Neural Network-Based Method for Person Identification Using Electrocardiogram Signals Acquired on Different Days
by Yeong-Hyeon Byeon and Keun-Chang Kwak
Appl. Sci. 2024, 14(17), 7959; https://doi.org/10.3390/app14177959 - 6 Sep 2024
Viewed by 801
Abstract
Electrocardiogram (ECG) signals are a measure minute electrical signals generated during the cardiac cycle, a biometric signal that occurs during vital human activity. ECG signals are susceptible to various types of noise depending on the data acquisition conditions, with factors such as sensor [...] Read more.
Electrocardiogram (ECG) signals are a measure minute electrical signals generated during the cardiac cycle, a biometric signal that occurs during vital human activity. ECG signals are susceptible to various types of noise depending on the data acquisition conditions, with factors such as sensor placement and the physiological and mental states of the subject contributing to the diverse shapes of these signals. When the data are acquired in a single session, the environmental variables are relatively similar, resulting in similar ECG signals; however, in subsequent sessions, even for the same person, changes in the environmental variables can alter the signal shape. This phenomenon poses challenges for person identification using ECG signals acquired on different days. To improve the performance of individual identification, even when ECG data is acquired on different days, this paper proposes an ensemble deep neural network for person identification by comparing and analyzing the ECG recognition performance under various conditions. The proposed ensemble deep neural network comprises three streams that incorporate two well-known pretrained models. Each network receives the time-frequency representation of ECG signals as input, and a stream reuses the same network structure under different learning conditions with or without data augmentation. The proposed ensemble deep neural network was validated on the Physikalisch-Technische Bundesanstalt dataset, and the results confirmed a 3.39% improvement in accuracy compared to existing methods. Full article
(This article belongs to the Section Applied Biosciences and Bioengineering)
Show Figures

Figure 1

Figure 1
<p>Example of transforming an ECG signal into a time-frequency representation using continuous wavelet transform.</p>
Full article ">Figure 2
<p>ResNet-based classifier for individual identification using ECG signal without data augmentation.</p>
Full article ">Figure 3
<p>ResNet-based classifier for individual identification using ECG signal with data augmentation.</p>
Full article ">Figure 4
<p>Inception-ResNet-v2-based classifier for individual identification using ECG signals without data augmentation.</p>
Full article ">Figure 5
<p>Ensemble neural network for individual identification using ECG signal.</p>
Full article ">Figure 6
<p>Diagnostic categories of the PTB-ECG dataset.</p>
Full article ">Figure 7
<p>The examples of signals with various sample lengths.</p>
Full article ">Figure 8
<p>The results of applying the SGF and BWPF on an ECG signal.</p>
Full article ">Figure 8 Cont.
<p>The results of applying the SGF and BWPF on an ECG signal.</p>
Full article ">Figure 9
<p>Training processes of pretrained ResNet-101 when training and validating with data from different sessions.</p>
Full article ">Figure 10
<p>Training processes of pretrained ResNet-101 with data augmentation when training and validating with data from different sessions.</p>
Full article ">Figure 11
<p>Training processes of pretrained Inception-ResNet-v2 when training and validating with data from different sessions.</p>
Full article ">
22 pages, 2648 KiB  
Article
Damage Detection and Localization Methodology Based on Strain Measurements and Finite Element Analysis: Structural Health Monitoring in the Context of Industry 4.0
by Andrés R. Herrera, Joham Alvarez, Jaime Restrepo, Camilo Herrera, Sven Rodríguez, Carlos A. Escobar, Rafael E. Vásquez and Julián Sierra-Pérez
Aerospace 2024, 11(9), 708; https://doi.org/10.3390/aerospace11090708 - 30 Aug 2024
Viewed by 1264
Abstract
This paper investigates the integration of Structural Health Monitoring (SHM) within the frame of Industry 4.0 (I4.0) technologies, highlighting the potential for intelligent infrastructure management through the utilization of big data analytics, machine learning (ML), and the Internet of Things (IoT). This study [...] Read more.
This paper investigates the integration of Structural Health Monitoring (SHM) within the frame of Industry 4.0 (I4.0) technologies, highlighting the potential for intelligent infrastructure management through the utilization of big data analytics, machine learning (ML), and the Internet of Things (IoT). This study presents a success case focused on a novel SHM methodology for detecting and locating damages in metallic aircraft structures, employing dimensional reduction techniques such as Principal Component Analysis (PCA). By analyzing strain data collected from a network of sensors and comparing it to a baseline pristine condition, the methodology aims to identify subtle changes in local strain distribution indicative of damage. Through extensive Finite Element Analysis (FEA) simulations and a PCA contribution analysis, the research explores the influence of various factors on damage detection, including sensor placement, noise levels, and damage size and type. The findings demonstrate the effectiveness of the proposed methodology in detecting cracks and holes as small as 2 mm in length, showcasing the potential for early damage identification and targeted interventions in diverse sectors such as aerospace, civil engineering, and manufacturing. Ultimately, this paper underscores the synergistic relationship between SHM and I4.0, paving the way for a future of intelligent, resilient, and sustainable infrastructure. Full article
(This article belongs to the Special Issue Aircraft Structural Health Monitoring and Digital Twin)
Show Figures

Figure 1

Figure 1
<p>Methodology of SHM and the main techniques nowadays.</p>
Full article ">Figure 2
<p>SHM industry applications within the framework of I4.0.</p>
Full article ">Figure 3
<p>Machine learning classification and main algorithms.</p>
Full article ">Figure 4
<p>Drawing views of the wing structure.</p>
Full article ">Figure 5
<p>Meshed geometry and localization of damages.</p>
Full article ">Figure 6
<p>Geometry and mesh for a 20 mm hole and a crack.</p>
Full article ">Figure 7
<p>Virtual sensor localization.</p>
Full article ">Figure 8
<p>Mesh convergence study.</p>
Full article ">Figure 9
<p>Boundary conditions used for the validation model.</p>
Full article ">Figure 10
<p>Maximum moment applied.</p>
Full article ">Figure 11
<p><span class="html-italic">Q</span> statistic for hole and crack.</p>
Full article ">Figure 12
<p>Damage location comparison between lower and higher F1 Score values for hole.</p>
Full article ">Figure 13
<p>Damage location comparison between lower and higher F1 values for crack.</p>
Full article ">Figure 14
<p>Sensors removed for hole and crack.</p>
Full article ">Figure 15
<p>Signal-to-noise ratio for hole and crack at damage <span class="html-italic">D</span>6.</p>
Full article ">Figure 16
<p>Signal-to-noise ratio for hole and crack at damage <span class="html-italic">D</span>3.</p>
Full article ">
25 pages, 9901 KiB  
Article
A Cost-Effective Fault Diagnosis and Localization Approach for Utility-Scale PV Systems Using Limited Number of Sensors
by Faris E. Alfaris, Essam A. Al-Ammar, Ghazi A. Ghazi and Ahmed A. AL-Katheri
Sustainability 2024, 16(15), 6454; https://doi.org/10.3390/su16156454 - 28 Jul 2024
Cited by 2 | Viewed by 1214
Abstract
As a result of global efforts to combat the rise in global climate change and carbon dioxide emissions, there has been a substantial increase in renewable energy investment for both residential and utility power generation. Solar power facilities are estimated to be among [...] Read more.
As a result of global efforts to combat the rise in global climate change and carbon dioxide emissions, there has been a substantial increase in renewable energy investment for both residential and utility power generation. Solar power facilities are estimated to be among the major contributors to global decarbonization in terms of capacity by 2050. Consequently, the majority of economically significant countries are progressively implementing utility-scale photovoltaic (U-PV) systems. Nevertheless, a major obstacle to the expansion of U-PV generation is the identification and assessment of direct current (DC) faults in the extensive array of PV panels. In order to address this obstacle, it is imperative to provide an evaluation method that can accurately and cost-effectively identify and locate potential DC faults in PV arrays. Therefore, many studies attempted to utilize thermal cameras, voltage and current sensors, power databases, and other detecting elements; however, some of these technologies provide extra hurdles in terms of the quantity and expense of the utilized hardware equipment. This work presents a sophisticated system that aims to diagnose and locate various types of PV faults, such as line-to-ground, line-to-line, inter-string, open-circuit, and partial shading events, within a PV array strings down to a module level. This study primarily depends on three crucial indicators: precise calculation of the PV array output power and current, optimal placement of a limited number of voltage sensors, and execution of specifically specified tests. The estimation of PV array power, along with selectively placed voltage sensors, minimizes the time and equipment required for fault detection and diagnosis. The feasibility of the proposed method is investigated with real field data and the PSCAD simulation platform during all possible weather conditions and array faults. The results demonstrate that the proposed approach can accurately diagnose and localize faults with only NS/2 voltage sensors, where NS is the number of PV array parallel strings. Full article
Show Figures

Figure 1

Figure 1
<p>Sample of a large-scale PV system with <span class="html-italic">M</span> by <span class="html-italic">S</span> array size.</p>
Full article ">Figure 2
<p>Behavior and equivalent circuit of a PV array experiencing an L–G fault in string 1.</p>
Full article ">Figure 3
<p>Behavior and equivalent circuit of a PV array experiencing a module fault in string 1.</p>
Full article ">Figure 4
<p>Behavior and equivalent circuit of a PV array experiencing partial shading at module <span class="html-italic">M31</span>.</p>
Full article ">Figure 5
<p>I–V characteristic of a PV module with 100%, 75%, 50%, and 25% solar radiation intensity.</p>
Full article ">Figure 6
<p>Current versus array voltage and voltage versus array voltage for healthy and faulty stings during partial shading at module <span class="html-italic">M31</span> (with 25% solar radiation intensity).</p>
Full article ">Figure 7
<p>I–V curve of a faulty string during faults with similar power and voltage conditions.</p>
Full article ">Figure 8
<p>Behavior and equivalent circuit of a PV array with an open-string fault in string 1.</p>
Full article ">Figure 9
<p>The effect of L–G, open circuit, inter-string module, and partial shading faults on the PV system: (<b>left</b>) array I–V characteristic; (<b>right</b>) P–V characteristic.</p>
Full article ">Figure 10
<p>Impact of L–G, open circuit, inter-string module, and partial shading faults on the PV system: (<b>left</b>) faulty string to array I–V characteristic; (<b>right</b>) faulty string to array V-V characteristic.</p>
Full article ">Figure 11
<p>Strategic placement of strings differential voltage sensors.</p>
Full article ">Figure 12
<p>Flowchart of the proposed approach for PV fault detection and diagnosis.</p>
Full article ">Figure 13
<p>Altitude, azimuth, tilt, and hour angles for the sun, with respect to the horizontal axis.</p>
Full article ">Figure 14
<p>Flowchart of the conducted process for PV output power estimation.</p>
Full article ">Figure 15
<p>Estimated solar irradiance energy for different days and PV module orientation.</p>
Full article ">Figure 16
<p>Laboratory-scale PV array used for case study analysis.</p>
Full article ">Figure 17
<p>Actual and estimated power for the considered PV system during different days and sky conditions: (<b>a</b>) 27 February; (<b>b</b>) 11 March; (<b>c</b>) 17 May; (<b>d</b>) 22 May.</p>
Full article ">Figure 18
<p>Results from the conducted array tests during L–G fault: (<b>left</b>) during LR<sub>F</sub> L–G fault (with <span class="html-italic">R<sub>F</sub></span> = 0); (<b>right</b>) during HR<sub>F</sub> L–G fault (with <span class="html-italic">R<sub>F</sub></span> = 10 Ω).</p>
Full article ">Figure 19
<p>Results from the conducted array tests during module fault: (<b>left</b>) during LR<sub>F</sub> module fault (with <span class="html-italic">R<sub>F</sub></span> = 0); (<b>right</b>) during HR<sub>F</sub> module fault (with <span class="html-italic">R<sub>F</sub></span> = 10 Ω).</p>
Full article ">Figure 20
<p>Array I–V curve during L–G and module faults in string 1 with HR<sub>F</sub> (<span class="html-italic">R<sub>F</sub></span> = 10 Ω).</p>
Full article ">Figure 21
<p>Results from the conducted array tests: (<b>left</b>) during open-string fault; (<b>right</b>) during partial shading at module <span class="html-italic">M31</span> (with 25% solar radiation intensity).</p>
Full article ">Figure 22
<p>Array I–V curve during partial shading on module <span class="html-italic">M31</span> (with 50% and 25% solar radiation intensity).</p>
Full article ">
25 pages, 4075 KiB  
Article
A Robust Sparse Sensor Placement Strategy Based on Indicators of Noise for Ocean Monitoring
by Qiannan Zhang, Huafeng Wu, Li’nian Liang, Xiaojun Mei, Jiangfeng Xian and Yuanyuan Zhang
J. Mar. Sci. Eng. 2024, 12(7), 1220; https://doi.org/10.3390/jmse12071220 - 19 Jul 2024
Cited by 2 | Viewed by 891
Abstract
A well-performing data-driven sparse sensor deployment strategy is critical for marine monitoring systems, as it enables the optimal reconstruction of marine physical quantities with fewer sensors. However, ocean data typically contain substantial amounts of noise, including outliers (incomplete data) and inherent measurement noise, [...] Read more.
A well-performing data-driven sparse sensor deployment strategy is critical for marine monitoring systems, as it enables the optimal reconstruction of marine physical quantities with fewer sensors. However, ocean data typically contain substantial amounts of noise, including outliers (incomplete data) and inherent measurement noise, which heightens the complexity of sensor deployment. Therefore, this study optimizes the sparse sensor placement model by establishing noise indicators, including small noise weight and large noise weight, which are measured by entropy to minimize the prediction bias. Building on this, a robust sparse sensor placement algorithm is proposed, which utilizes the block coordinate update (BCU) iteration method to solve the function. During the iterative updating process, the proposed algorithm simultaneously updates the selection matrix, reconstruction matrix, and noise matrix. This allows for effective identification and mitigation of noise in the data through evaluation. Consequently, the deployed sensors achieve superior reconstruction performance compared to other deployment methods that do not incorporate noise evaluation. Experiments are also conducted on datasets of sea surface temperature (SST) and global ocean salinity, which demonstrate that our strategy significantly outperforms several other considered methods in terms of reconstruction accuracy while enabling autonomous sensor deployment under noisy conditions. Full article
Show Figures

Figure 1

Figure 1
<p>Reconstruction for low-rank data matrix of different parameters in RSSPIN. (<b>a</b>) Reconstruction error of different <span class="html-italic">α</span> and <span class="html-italic">β</span>; (<b>b</b>) Execution time of different <span class="html-italic">α</span> and <span class="html-italic">β</span>.</p>
Full article ">Figure 2
<p>Convergence for different numbers of samples in RSSPIN, in which normalized data are used. (<b>a</b>) Convergence rate of total objective results without outliers in iteration; (<b>b</b>) convergence rate of reconstruction errors without outliers in iteration.</p>
Full article ">Figure 3
<p>Convergence rate for different outlier ratios in RSSPIN. (<b>a</b>) Convergence rate of total objective results with different outlier ratios in iteration; (<b>b</b>) convergence rate of reconstruction errors with different outlier ratios in iteration.</p>
Full article ">Figure 4
<p>Reconstruction error of different methods for SST. (<b>a</b>) Reconstruction errors of different outlier rates using Equation (36); (<b>b</b>) reconstruction errors of different samples using Equation (36); (<b>c</b>) reconstruction errors of different outlier rates using Equation (37); (<b>d</b>) reconstruction errors of different samples using Equation (37).</p>
Full article ">Figure 5
<p>Reconstructed SST of different samples by RSSPIN (Ra = 0; Sr = 0). (<b>a</b>) Snapshot of test data; (<b>b</b>) reconstructed SST of 50 samples using RSSPIN; (<b>c</b>) reconstructed SST of 500 samples using RSSPIN.</p>
Full article ">Figure 5 Cont.
<p>Reconstructed SST of different samples by RSSPIN (Ra = 0; Sr = 0). (<b>a</b>) Snapshot of test data; (<b>b</b>) reconstructed SST of 50 samples using RSSPIN; (<b>c</b>) reconstructed SST of 500 samples using RSSPIN.</p>
Full article ">Figure 6
<p>Reconstruction error of different methods for global ocean salinity. (<b>a</b>) Reconstruction errors of different outlier rates; (<b>b</b>) reconstruction errors of different samples.</p>
Full article ">Figure 7
<p>Reconstruction salinity field by RSSPIN. (<b>a</b>) Test salinity; (<b>b</b>) reconstruction salinity with Ra = 0.2; (<b>c</b>) reconstruction salinity with Ra = 0.4.</p>
Full article ">Figure 7 Cont.
<p>Reconstruction salinity field by RSSPIN. (<b>a</b>) Test salinity; (<b>b</b>) reconstruction salinity with Ra = 0.2; (<b>c</b>) reconstruction salinity with Ra = 0.4.</p>
Full article ">
15 pages, 6642 KiB  
Article
Optimization of Sensor Placement for a Measurement System for the Determination of Local Magnetic Material Properties
by Alice Reinbacher-Köstinger, Andreas Gschwentner, Eniz Mušeljić and Manfred Kaltenbacher
Mathematics 2024, 12(14), 2220; https://doi.org/10.3390/math12142220 - 16 Jul 2024
Viewed by 905
Abstract
The aim of this work is to optimize the sensor positions of a sensor–actuator measurement system for identifying local variations in the magnetic permeability of cut steel sheets. Before solving the actual identification problem, i.e., finding the material distribution, the sensor placement of [...] Read more.
The aim of this work is to optimize the sensor positions of a sensor–actuator measurement system for identifying local variations in the magnetic permeability of cut steel sheets. Before solving the actual identification problem, i.e., finding the material distribution, the sensor placement of the measurement setup should be improved in order to reduce the uncertainty of the identification of the material distribution. The Fisher information matrix (FIM), which allows one to quantify the amount of information that the measurements carry about the unknown parameters, is used as the main metric for the objective function of this design optimization. The forward problem is solved by the finite element method. The results show that the proposed method is able to find optimal sensor positions as well as the minimum number of sensors to achieve a desired maximum parameter uncertainty. Full article
(This article belongs to the Section E2: Control Theory and Mechanics)
Show Figures

Figure 1

Figure 1
<p>Front view of the 3D model of the sensor–actuator system.</p>
Full article ">Figure 2
<p>Detailed view of the measurement setup showing the sensor array and different measurement positions realized by moving the sensor–actuator system relative to the steel sheets. The distance between two sensors is 1 mm in the x direction and the measuring positions are chosen equidistantly ±1 mm around the middle sensor. For better illustration, only two measuring positions are shown in the upper figure, and the steel sheets are moved instead of the sensor–actuator system.</p>
Full article ">Figure 3
<p>Permeability distribution along two juxtaposed samples for two different materials.</p>
Full article ">Figure 4
<p>Flowchart of the optimization procedure.</p>
Full article ">Figure 5
<p>Absolute values of the sensitivities <math display="inline"><semantics> <mrow> <mo>∂</mo> <mi>B</mi> </mrow> </semantics></math>/<math display="inline"><semantics> <mrow> <mo>∂</mo> <mover accent="true"> <mi>γ</mi> <mo stretchy="false">^</mo> </mover> </mrow> </semantics></math> (<b>left</b>) and <math display="inline"><semantics> <mrow> <mo>∂</mo> <mi>B</mi> </mrow> </semantics></math>/<math display="inline"><semantics> <mrow> <mo>∂</mo> <mi>δ</mi> </mrow> </semantics></math> (<b>right</b>) of all sensors for a material with the parameters <math display="inline"><semantics> <mrow> <msub> <mi mathvariant="bold">θ</mi> <mn>1</mn> </msub> <mo>=</mo> <mrow> <mo>[</mo> <mover accent="true"> <mi>γ</mi> <mo stretchy="false">^</mo> </mover> <mo>,</mo> <mi>δ</mi> <mo>]</mo> </mrow> <mo>=</mo> <mrow> <mo>[</mo> <mn>0.4</mn> <mo>,</mo> <mn>0.3</mn> <mspace width="0.166667em"/> <mi>mm</mi> <mo>]</mo> </mrow> </mrow> </semantics></math>.</p>
Full article ">Figure 6
<p>Absolute values of the sensitivities <math display="inline"><semantics> <mrow> <mo>∂</mo> <mi>B</mi> </mrow> </semantics></math>/<math display="inline"><semantics> <mrow> <mo>∂</mo> <mover accent="true"> <mi>γ</mi> <mo stretchy="false">^</mo> </mover> </mrow> </semantics></math> (<b>left</b>) and <math display="inline"><semantics> <mrow> <mo>∂</mo> <mi>B</mi> </mrow> </semantics></math>/<math display="inline"><semantics> <mrow> <mo>∂</mo> <mi>δ</mi> </mrow> </semantics></math> (<b>right</b>) of all sensors for a material with the parameters <math display="inline"><semantics> <mrow> <msub> <mi mathvariant="bold">θ</mi> <mn>2</mn> </msub> <mo>=</mo> <mrow> <mo>[</mo> <mover accent="true"> <mi>γ</mi> <mo stretchy="false">^</mo> </mover> <mo>,</mo> <mi>δ</mi> <mo>]</mo> </mrow> <mo>=</mo> <mrow> <mo>[</mo> <mn>0.4</mn> <mo>,</mo> <mn>0.7</mn> <mspace width="0.166667em"/> <mi>mm</mi> <mo>]</mo> </mrow> </mrow> </semantics></math>.</p>
Full article ">Figure 7
<p>Absolute values of the sensitivities <math display="inline"><semantics> <mrow> <mo>∂</mo> <mi>B</mi> </mrow> </semantics></math>/<math display="inline"><semantics> <mrow> <mo>∂</mo> <mover accent="true"> <mi>γ</mi> <mo stretchy="false">^</mo> </mover> </mrow> </semantics></math> for a material with the parameters <math display="inline"><semantics> <mrow> <msub> <mi mathvariant="bold">θ</mi> <mn>1</mn> </msub> <mo>=</mo> <mrow> <mo>[</mo> <mover accent="true"> <mi>γ</mi> <mo stretchy="false">^</mo> </mover> <mo>,</mo> <mi>δ</mi> <mo>]</mo> </mrow> <mo>=</mo> <mrow> <mo>[</mo> <mn>0.4</mn> <mo>,</mo> <mn>0.3</mn> <mspace width="0.166667em"/> <mi>mm</mi> <mo>]</mo> </mrow> </mrow> </semantics></math> (<b>left</b>) and <math display="inline"><semantics> <mrow> <msub> <mi mathvariant="bold">θ</mi> <mn>2</mn> </msub> <mo>=</mo> <mrow> <mo>[</mo> <mover accent="true"> <mi>γ</mi> <mo stretchy="false">^</mo> </mover> <mo>,</mo> <mi>δ</mi> <mo>]</mo> </mrow> <mo>=</mo> <mrow> <mo>[</mo> <mn>0.4</mn> <mo>,</mo> <mn>0.7</mn> <mspace width="0.166667em"/> <mi>mm</mi> <mo>]</mo> </mrow> </mrow> </semantics></math> (<b>right</b>) of all sensors between the poles along <math display="inline"><semantics> <mrow> <mi>z</mi> <mo>=</mo> <mn>0</mn> </mrow> </semantics></math>.</p>
Full article ">Figure 8
<p>Convergence and constraint violation of the optimization problem as formulated in (<a href="#FD22-mathematics-12-02220" class="html-disp-formula">22</a>). Also, the standard deviations of the material parameter combination with the highest uncertainty values are shown.</p>
Full article ">Figure 9
<p>Optimized sensor positions with <math display="inline"><semantics> <mrow> <msub> <mi>n</mi> <mrow> <mi mathvariant="normal">s</mi> <mo>,</mo> <mi>max</mi> </mrow> </msub> <mo>=</mo> <mn>9</mn> </mrow> </semantics></math>, <math display="inline"><semantics> <mrow> <msub> <mi>n</mi> <mi mathvariant="normal">m</mi> </msub> <mo>=</mo> <mn>5</mn> </mrow> </semantics></math> different parameter configurations <math display="inline"><semantics> <mi mathvariant="bold">θ</mi> </semantics></math>, and using the E-criterion. The vertical gray lines represent the different measurement positions as described in <a href="#sec5dot1-mathematics-12-02220" class="html-sec">Section 5.1</a>.</p>
Full article ">Figure 10
<p>Confidence ellipses of different material parameter combinations using the optimized sensor configuration, as shown in <a href="#mathematics-12-02220-f009" class="html-fig">Figure 9</a>.</p>
Full article ">Figure 11
<p>Degradation profile for the material parameter set <math display="inline"><semantics> <mrow> <mi mathvariant="bold">θ</mi> <mo>=</mo> <mo>[</mo> <mn>0.5</mn> <mo>,</mo> <mn>0.5</mn> <mspace width="0.166667em"/> <mi>mm</mi> <mo>]</mo> </mrow> </semantics></math> and with an uncertainty of <math display="inline"><semantics> <mrow> <msub> <mi>σ</mi> <mi mathvariant="normal">t</mi> </msub> <mo>=</mo> <mrow> <mo>[</mo> <mn>0.05</mn> <mo>,</mo> <mn>0.05</mn> <mspace width="0.166667em"/> <mi>mm</mi> <mo>]</mo> </mrow> </mrow> </semantics></math>. Thus, the confidence interval is <math display="inline"><semantics> <mrow> <mo>(</mo> <mn>0.5</mn> <mo>−</mo> <mn>0.1</mn> <mo>,</mo> <mn>0.5</mn> <mo>+</mo> <mn>0.1</mn> <mo>]</mo> </mrow> </semantics></math> for both parameters.</p>
Full article ">Figure 12
<p>Optimized sensor positions with <math display="inline"><semantics> <mrow> <msub> <mi>n</mi> <mrow> <mi mathvariant="normal">s</mi> <mo>,</mo> <mi>max</mi> </mrow> </msub> <mo>=</mo> <mn>10</mn> </mrow> </semantics></math> and <math display="inline"><semantics> <mrow> <msub> <mi>n</mi> <mi mathvariant="normal">m</mi> </msub> <mo>=</mo> <mn>5</mn> </mrow> </semantics></math> different parameter configurations <math display="inline"><semantics> <mi mathvariant="bold">θ</mi> </semantics></math> and using the E-criterion.</p>
Full article ">Figure 13
<p>Confidence ellipses of different material parameter combinations using the optimized sensor configuration, as shown in <a href="#mathematics-12-02220-f012" class="html-fig">Figure 12</a>.</p>
Full article ">
26 pages, 5964 KiB  
Article
State Observer-Based Conditioned Reverse-Path Method for Nonlinear System Identification
by Atta Oveisi, Umaaran Gogilan, Jafar Keighobadi and Tamara Nestorović
Actuators 2024, 13(4), 142; https://doi.org/10.3390/act13040142 - 11 Apr 2024
Viewed by 1140
Abstract
In light of the complex behavior of vibrating structures, their reliable modeling plays a crucial role in the analysis and system design for vibration control. In this paper, the reverse-path (RP) method is revisited, further developed, and applied to modeling a nonlinear system, [...] Read more.
In light of the complex behavior of vibrating structures, their reliable modeling plays a crucial role in the analysis and system design for vibration control. In this paper, the reverse-path (RP) method is revisited, further developed, and applied to modeling a nonlinear system, particularly with respect to the identification of the frequency response function for a nominal underlying linear system and the determination of the structural nonlinearities. The present approach aims to overcome the requirement for measuring all nonlinear system states all the time during operation. Especially in large-scale systems, this might be a tedious task and often practically infeasible since it would require having individual sensors assigned for each state involved in the design process. In addition, the proper placement and simultaneous operation of a large number of transducers would represent further difficulty. To overcome those issues, we have proposed state estimation in light of the observability criteria, which significantly reduces the number of required sensor elements. To this end, relying on the optimal sensor placement problem, the state estimation process reduces to the solution of Kalman filtering. On this ground, the problem of nonlinear system identification for large-scale systems can be addressed using the observer-based conditioned RP method (OBCRP) proposed in this paper. In contrast to the classical RP method, the current one can potentially handle local and distributed nonlinearities. Moreover, in addition to the state estimation and in comparison to the orthogonal RP method, a new frequency-dependent weighting is introduced in this paper, which results in superior nonlinear system identification performances. Implementation of the method is demonstrated on a multi-degree-of-freedom discretized lumped mass system, representing a substitute model of a physical counterpart used for the identification of the model parameters. Full article
Show Figures

Figure 1

Figure 1
<p>Spectrum of the odd-random excitation input signal and the spectrum of the output signal with detection lines (red) and neighboring lines (green).</p>
Full article ">Figure 2
<p>A substitute lumped mass discrete model with multi-degree-of-freedom.</p>
Full article ">Figure 3
<p>Wavelet-transformed output signal on a non-logarithmic scale, obtained for different values of excitation amplitudes measured at different mass positions.</p>
Full article ">Figure 4
<p>Acceleration surface across the junction between the fixed base and the mass <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>m</mi> </mrow> <mrow> <mn>1</mn> </mrow> </msub> </mrow> </semantics></math>.</p>
Full article ">Figure 5
<p>Corresponding side-view (acceleration vs. relative displacement) of the acceleration surface used for nonparametric modeling of nonlinearity at (<b>a</b>) mass <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>m</mi> </mrow> <mrow> <mn>1</mn> </mrow> </msub> </mrow> </semantics></math> and the (<b>b</b>) junction between masses <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>m</mi> </mrow> <mrow> <mn>1</mn> </mrow> </msub> <mtext> </mtext> <mi mathvariant="normal">a</mi> <mi mathvariant="normal">n</mi> <mi mathvariant="normal">d</mi> <mtext> </mtext> <msub> <mrow> <mi>m</mi> </mrow> <mrow> <mn>2</mn> </mrow> </msub> </mrow> </semantics></math>.</p>
Full article ">Figure 6
<p>Corresponding side-view (acceleration vs. relative displacement) of the acceleration surface used for nonparametric modeling of nonlinearity at the (<b>a</b>) junction between masses <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>m</mi> </mrow> <mrow> <mn>2</mn> </mrow> </msub> <mtext> </mtext> <mi mathvariant="normal">a</mi> <mi mathvariant="normal">n</mi> <mi mathvariant="normal">d</mi> <mtext> </mtext> <msub> <mrow> <mi>m</mi> </mrow> <mrow> <mn>3</mn> </mrow> </msub> </mrow> </semantics></math> and the (<b>b</b>) junction between masses <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>m</mi> </mrow> <mrow> <mn>3</mn> </mrow> </msub> </mrow> </semantics></math> and <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>m</mi> </mrow> <mrow> <mn>4</mn> </mrow> </msub> </mrow> </semantics></math>.</p>
Full article ">Figure 7
<p>(<b>a</b>) The transient contribution. (<b>b</b>) The robust LPM used to obtain the noise/nonlinearity covariance and BLA.</p>
Full article ">Figure 8
<p>The poles of the simulation system (<span style="color:#816E5F">×</span>) in comparison to the poles of the steady-state Kalman state observer (<span style="color:#4472C4">×</span>).</p>
Full article ">Figure 9
<p>The observation error in states.</p>
Full article ">Figure 10
<p>The real part of the estimated coefficients compared to the actual values.</p>
Full article ">Figure 11
<p><b>Top</b>: The imaginary part of the estimated spectrum of the cubic term <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>β</mi> </mrow> <mrow> <mn>2</mn> </mrow> </msub> </mrow> </semantics></math> and calculated envelopes. <b>Bottom</b>: The calculated weight based on Algorithm 2.</p>
Full article ">Figure 12
<p>Conditioned and unconditioned FRFs in comparison to the correct linear estimate represented for selected states <span class="html-italic">x</span><sub>1</sub> and <span class="html-italic">x</span><sub>5</sub>.</p>
Full article ">Figure 13
<p>(<b>a</b>) The frequency content of the sampled colored noise acting on the measurement outputs. (<b>b</b>) The extracted FRFs under the prescribed measurement noise levels in <a href="#actuators-13-00142-t006" class="html-table">Table 6</a>.</p>
Full article ">
51 pages, 1384 KiB  
Review
Fundamentals, Algorithms, and Technologies of Occupancy Detection for Smart Buildings Using IoT Sensors
by Pratiksha Chaudhari, Yang Xiao, Mark Ming-Cheng Cheng and Tieshan Li
Sensors 2024, 24(7), 2123; https://doi.org/10.3390/s24072123 - 26 Mar 2024
Cited by 5 | Viewed by 4104
Abstract
Smart buildings use advanced technologies to automate building functions. One important function is occupancy detection using Internet of Things (IoT) sensors for smart buildings. Occupancy information is useful information to reduce energy consumption by automating building functions such as lighting, heating, ventilation, and [...] Read more.
Smart buildings use advanced technologies to automate building functions. One important function is occupancy detection using Internet of Things (IoT) sensors for smart buildings. Occupancy information is useful information to reduce energy consumption by automating building functions such as lighting, heating, ventilation, and air conditioning systems. The information is useful to improve indoor air quality by ensuring that ventilation systems are used only when and where they are needed. Additionally, it is useful to enhance building security by detecting unusual or unexpected occupancy levels and triggering appropriate responses, such as alarms or alerts. Occupancy information is useful for many other applications, such as emergency response, plug load energy management, point-of-interest identification, etc. However, the accuracy of occupancy detection is limited by factors such as real-time occupancy data, sensor placement, privacy concerns, and the presence of pets or objects that can interfere with sensor reading. With the rapid development of IoT sensor technologies and the increasing need for smart building solutions, there is a growing interest in occupancy detection techniques. There is a need to provide a comprehensive survey of these technologies. Although there are some exciting survey papers, they all have limited scopes with different focuses. Therefore, this paper provides a comprehensive overview of the current state-of-the-art occupancy detection methods (including both traditional algorithms and machine learning algorithms) and devices with their advantages and limitations. It surveys and compares fundamental technologies (such as sensors, algorithms, etc.) for smart buildings. Furthermore, the survey provides insights and discussions, which can help researchers, practitioners, and stakeholders develop more effective occupancy detection solutions for smart buildings. Full article
Show Figures

Figure 1

Figure 1
<p>The structure of the paper outlines the detection of occupancy within smart buildings.</p>
Full article ">Figure 2
<p>How the occupancy detection for smart buildings using IoT sensors system works [<a href="#B14-sensors-24-02123" class="html-bibr">14</a>].</p>
Full article ">Figure 3
<p>IoT system architecture for occupancy sensing [<a href="#B11-sensors-24-02123" class="html-bibr">11</a>].</p>
Full article ">Figure 4
<p>Occupancy detection using the BOD Algorithm.</p>
Full article ">Figure 5
<p>The common steps of the process of occupancy detection via ML/DL.</p>
Full article ">Figure 6
<p>Optimal hyperplane using the SVM algorithm [<a href="#B71-sensors-24-02123" class="html-bibr">71</a>].</p>
Full article ">Figure 7
<p>Occupancy classification using the KNN Algorithm [<a href="#B71-sensors-24-02123" class="html-bibr">71</a>].</p>
Full article ">Figure 8
<p>Occupancy classification using the RF algorithm [<a href="#B71-sensors-24-02123" class="html-bibr">71</a>].</p>
Full article ">Figure 9
<p>FNN architecture with 3 layers [<a href="#B80-sensors-24-02123" class="html-bibr">80</a>].</p>
Full article ">Figure 10
<p>CNN architecture with five layers [<a href="#B82-sensors-24-02123" class="html-bibr">82</a>].</p>
Full article ">Figure 11
<p>RNN Architecture [<a href="#B85-sensors-24-02123" class="html-bibr">85</a>].</p>
Full article ">Figure 12
<p>LSTM architecture [<a href="#B88-sensors-24-02123" class="html-bibr">88</a>].</p>
Full article ">
16 pages, 2895 KiB  
Article
Dynamic Measurement of a Cancer Biomarker: Towards In Situ Application of a Fiber-Optic Ball Resonator Biosensor in CD44 Protein Detection
by Zhuldyz Myrkhiyeva, Kanagat Kantoreyeva, Aliya Bekmurzayeva, Anthony W. Gomez, Zhannat Ashikbayeva, Meruyert Tilegen, Tri T. Pham and Daniele Tosi
Sensors 2024, 24(6), 1991; https://doi.org/10.3390/s24061991 - 21 Mar 2024
Cited by 2 | Viewed by 1727
Abstract
The accuracy and efficacy of medical treatment would be greatly improved by the continuous and real-time monitoring of protein biomarkers. Identification of cancer biomarkers in patients with solid malignant tumors is receiving increasing attention. Existing techniques for detecting cancer proteins, such as the [...] Read more.
The accuracy and efficacy of medical treatment would be greatly improved by the continuous and real-time monitoring of protein biomarkers. Identification of cancer biomarkers in patients with solid malignant tumors is receiving increasing attention. Existing techniques for detecting cancer proteins, such as the enzyme-linked immunosorbent assay, require a lot of work, are not multiplexed, and only allow for single-time point observations. In order to get one step closer to clinical usage, a dynamic platform for biosensing the cancer biomarker CD44 using a single-mode optical fiber-based ball resonator biosensor was designed, constructed and evaluated in this work. The main novelty of the work is an in-depth study of the capability of an in-house fabricated optical fiber biosensor for in situ detection of a cancer biomarker (CD44 protein) by conducting several types of experiments. The main results of the work are as follows: (1) Calibration of the fabricated fiber-optic ball resonator sensors in both static and dynamic conditions showed similar sensitivity to the refractive index change demonstrating its usefulness as a biosensing platform for dynamic measurements; (2) The fabricated sensors were shown to be insensitive to pressure changes further confirming their utility as an in situ sensor; (3) The sensor’s packaging and placement were optimized to create a better environment for the fabricated ball resonator’s performance in blood-mimicking environment; (4) Incubating increasing protein concentrations with antibody-functionalized sensor resulted in nearly instantaneous signal change indicating a femtomolar detection limit in a dynamic range from 7.1 aM to 16.7 nM; (5) The consistency of the obtained signal change was confirmed by repeatability studies; (6) Specificity experiments conducted under dynamic conditions demonstrated that the biosensors are highly selective to the targeted protein; (7) Surface morphology studies by AFM measurements further confirm the biosensor’s exceptional sensitivity by revealing a considerable shift in height but no change in surface roughness after detection. The biosensor’s ability to analyze clinically relevant proteins in real time with high sensitivity offers an advancement in the detection and monitoring of malignant tumors, hence improving patient diagnosis and health status surveillance. Full article
Show Figures

Figure 1

Figure 1
<p>The sequential procedure of fabricating an optical fiber ball resonator using the Fujikura LZM-100. The location in the splicer where the fiber is inserted to fabricate the ball resonator is indicated in the orange box in the bottom image. The upper image shows the schematic images of the fabrication procedures, which include aligning and splicing the optical fiber before heating and rotating it with a CO<sub>2</sub> laser. This process results in the formation of the ball resonator structure.</p>
Full article ">Figure 2
<p>Pressure characterization setup for an optical fiber ball resonator. A 499 μm resonator is placed into the tip of a burette and then systematically filled with DI water at different levels. The pressure data recorded at water column heights ranging from 16 cm to 66 cm demonstrates that changes in pressure do not alter the detected signals.</p>
Full article ">Figure 3
<p>An illustration of the setup for the dynamic measurement of the CD44 protein. The setup includes a Legato 100 KD Scientific syringe pump, which operates at a flow rate of 20 mL/min to mimic venous circulation, and a 20-gauge polyurethane cannula that protects the optical fiber ball resonator, coupled with a LUNA OBR 4600 device for accurate detection.</p>
Full article ">Figure 4
<p>Example of a calibration of a ball resonator for RI detection. (<b>a</b>) Spectrum of the ball resonator probe, for various RI values. (<b>b</b>) Inset displaying the spectrum in proximity of the detected spectral feature. (<b>c</b>) Change of intensity as a function of the refractive index.</p>
Full article ">Figure 5
<p>Analysis of pressure effects on the 499 μm optical fiber ball resonator.</p>
Full article ">Figure 6
<p>Representative images of the biosensors’ surface morphologies obtained from AFM measurements at various stages of functionalization. Images of different stages of functionalization in a 1 μm × 1 μm section of an optical fiber ball resonator’s surface are displayed: (<b>a</b>) Piranha pre-treatment, (<b>b</b>) silanization with APTMS, (<b>c</b>) heat treatment, (<b>d</b>) cross-linking with GA, (<b>e</b>) immobilization of antibodies, (<b>f</b>) blocking with mPEG-amine, and (<b>g</b>) CD44 protein detected by a ball resonator.</p>
Full article ">Figure 7
<p>Quantitative analysis of the height and surface roughness of various functionalization phases for optical fiber ball resonator biosensors. Assessing the variations in height (<b>a</b>) and RMS-roughness (<b>b</b>) at every stage of functionalization. * <span class="html-italic">p</span> ≤ 0.05, ** <span class="html-italic">p</span> ≤ 0.01, *** <span class="html-italic">p</span> ≤ 0.001, ns, <span class="html-italic">p</span> &gt; 0.05.</p>
Full article ">Figure 8
<p>An optical fiber ball resonator biosensor’s efficacy in detecting CD44 was analyzed. The sensorgram shows the change in signal intensity over time, showing how a functionalized optical fiber ball resonator reacts to increasing amounts of CD44 in serum, ranging from 7.1 aM to 16.7 nM in diluted calf serum; results from the 497 μm diameter sensor are highlighted.</p>
Full article ">Figure 9
<p>This 3D sensorgram shows the real-time detection of CD44 using a biosensor with a 497 μm optical fiber ball resonator. The sensorgram plots the spectral intensity against wavelength and time. The color gradient change from blue to yellow visually demonstrates the biosensor’s response to increasing concentrations of CD44 in serum, which range from 7.1 aM to 16.7 nM. Each peak correlates to a specific CD44 concentration, demonstrating the sensor’s ability to measure.</p>
Full article ">Figure 10
<p>Evaluation of optical fiber ball resonator biosensors’ specificity for CD44 in comparison to thrombin and gamma-globulin. The bar graph illustrates how biosensors with diameters varying from 492 to 497 μm responded differently to CD44, while showing minimal responses to gamma-globulin (489 μm) and thrombin (518 μm) at concentrations ranging from 9.3 fM to 16.7 nM. This indicates the biosensor’s specific affinity for CD44. Error bars represent the measurements standard deviation, emphasizing the consistency and reliability of the sensor’s specificity.</p>
Full article ">Figure 11
<p>Results for CD44 detection from three biological replicates. (<b>a</b>) The graph shows the sensor’s response percentages in a logarithmic range of CD44 concentrations, demonstrating the sensors’ consistent performance in several trials. This is visible from the overlapping data points and the shaded confidence interval. (<b>b</b>) The graph compares intensity change (dB) for three sensors with varying diameters (492 μm, 496 μm, and 497 μm) to the logarithmic concentration of CD44. It indicates that the sensors operate consistently when CD44 levels increase. Error bars are used to indicate the standard deviation, which highlights the accuracy of the sensors while taking several measurements.</p>
Full article ">
23 pages, 7621 KiB  
Article
Accurate Liquid Level Measurement with Minimal Error: A Chaotic Observer Approach
by Vighnesh Shenoy, Prathvi Shenoy and Santhosh Krishnan Venkata
Computation 2024, 12(2), 29; https://doi.org/10.3390/computation12020029 - 6 Feb 2024
Viewed by 1857
Abstract
This paper delves into precisely measuring liquid levels using a specific methodology with diverse real-world applications such as process optimization, quality control, fault detection and diagnosis, etc. It demonstrates the process of liquid level measurement by employing a chaotic observer, which senses multiple [...] Read more.
This paper delves into precisely measuring liquid levels using a specific methodology with diverse real-world applications such as process optimization, quality control, fault detection and diagnosis, etc. It demonstrates the process of liquid level measurement by employing a chaotic observer, which senses multiple variables within a system. A three-dimensional computational fluid dynamics (CFD) model is meticulously created using ANSYS to explore the laminar flow characteristics of liquids comprehensively. The methodology integrates the system identification technique to formulate a third-order state–space model that characterizes the system. Based on this mathematical model, we develop estimators inspired by Lorenz and Rossler’s principles to gauge the liquid level under specified liquid temperature, density, inlet velocity, and sensor placement conditions. The estimated results are compared with those of an artificial neural network (ANN) model. These ANN models learn and adapt to the patterns and features in data and catch non-linear relationships between input and output variables. The accuracy and error minimization of the developed model are confirmed through a thorough validation process. Experimental setups are employed to ensure the reliability and precision of the estimation results, thereby underscoring the robustness of our liquid-level measurement methodology. In summary, this study helps to estimate unmeasured states using the available measurements, which is essential for understanding and controlling the behavior of a system. It helps improve the performance and robustness of control systems, enhance fault detection capabilities, and contribute to dynamic systems’ overall efficiency and reliability. Full article
Show Figures

Figure 1

Figure 1
<p>Orifice meter flow lines and their pressure characteristics.</p>
Full article ">Figure 2
<p>Experimental setup.</p>
Full article ">Figure 3
<p>Simulink model of the system with the estimator.</p>
Full article ">Figure 4
<p>Flow chart for designing the estimator.</p>
Full article ">Figure 5
<p>Architecture of an artificial neural network.</p>
Full article ">Figure 6
<p>Practical setup.</p>
Full article ">Figure 7
<p>The workflow of CFD simulation, design of the estimator, ANN prediction model, and real-time implementation.</p>
Full article ">Figure 8
<p>Liquid level estimation with the initial condition (0.02, 0.1, 0.1) using the Lorenz estimator.</p>
Full article ">Figure 9
<p>Error estimation between the plant and the estimator with the initial condition (0.02, 0.1, 0.1) using the Lorenz estimator.</p>
Full article ">Figure 10
<p>Liquid level estimation with the initial condition (0.025, 0.15, 0.15) using the Lorenz estimator.</p>
Full article ">Figure 11
<p>Error estimation between the plant and the Lorenz estimator with the initial condition (0.025, 0.15, 0.15) using the Lorenz estimator.</p>
Full article ">Figure 12
<p>Liquid level estimation with the initial condition (0.02, 0.1, 0.1) using the Lorenz estimator with a noise magnitude of 10<sup>−6</sup>.</p>
Full article ">Figure 13
<p>Error estimation between the plant and the Lorenz estimator with the initial condition (0.02, 0.1, 0.1) using the Lorenz estimator.</p>
Full article ">Figure 14
<p>Liquid level estimation with the initial condition (0.02, 0.1, 0.1) using the Rossler estimator.</p>
Full article ">Figure 15
<p>Error estimation between the plant and the estimator with the initial condition (0.02, 0.1, 0.1) using the Rossler estimator.</p>
Full article ">Figure 16
<p>The liquid level estimation with the initial condition (0.025, 0.15, 0.15) using the Rossler estimator.</p>
Full article ">Figure 17
<p>Error estimation between the plant and the estimator with the initial condition (0.025, 0.15, 0.15) using the Rossler estimator.</p>
Full article ">Figure 18
<p>Liquid level estimation with the initial condition (0.02, 0.1, 0.1) using the Rossler estimator with a noise magnitude of 10<sup>−6</sup>.</p>
Full article ">Figure 19
<p>The noise between the plant and the estimator with the initial condition (0.02, 0.1, 0.1) using the Rossler estimator.</p>
Full article ">Figure 20
<p>Prediction for inlet velocity, y = 0.6 m/s, temperature = 50 °C, density = 900 kg/m<sup>3</sup>, and placement of the sensor at 20 inches from the pipe inlet using an ANN.</p>
Full article ">Figure 21
<p>Loss functions for 50 epochs.</p>
Full article ">
17 pages, 3495 KiB  
Article
Semi-Supervised Domain Adaptation for Individual Identification from Electrocardiogram Signals
by Yeong-Hyeon Byeon and Keun-Chang Kwak
Appl. Sci. 2023, 13(24), 13259; https://doi.org/10.3390/app132413259 - 14 Dec 2023
Viewed by 1148
Abstract
When acquiring electrocardiogram (ECG) signals, the placement of electrode patches is crucial for acquiring electrocardiographic signals. Constant displacement positions are essential for ensuring the consistency of the ECG signal when used for individual identification. However, achieving constant placement of ECG electrode patches in [...] Read more.
When acquiring electrocardiogram (ECG) signals, the placement of electrode patches is crucial for acquiring electrocardiographic signals. Constant displacement positions are essential for ensuring the consistency of the ECG signal when used for individual identification. However, achieving constant placement of ECG electrode patches in every trial for data acquisition is challenging. This is because different individuals may attach patches, and even when the same person attaches them, it may be difficult to specify the exact position. Therefore, gathering ECG data from various locations is necessary. However, this process requires a substantial amount of labor and time, owing to the requirement for multiple attempts. Nonetheless, persisting with these efforts enables the endurance of some ECG differences. To reduce labor and time, we propose a semi-supervised domain adaptation for individual identification using ECG signals. The method operates with a full set of original ECG signals and a small set of ECG signals from different placements to account for the differences between the signals in the generative adversarial network (CycleGAN). Specifically, to train the CycleGAN, the ECG signals were transformed into time–frequency representations, and the trained generator was used to generate ECG signals to expand the small set of ECG signals from different placements. Subsequently, both the original and generated signals were used to train the classifier for individual identification. This scenario can also be applied to the classification of ECG signals from different sensors. The PTB-ECG dataset was used for this experiment. We found that the proposed method showed higher accuracy than when only the original ECG signals were used for the training classifier. Full article
Show Figures

Figure 1

Figure 1
<p>Examples of the time-frequency representations of ECG signals using continuous wavelet transform (CWT).</p>
Full article ">Figure 2
<p>Generator of CycleGAN for domain transfer between channels.</p>
Full article ">Figure 3
<p>ResNet-based individual identification model.</p>
Full article ">Figure 4
<p>Training course of CycleGAN on class 1.</p>
Full article ">Figure 5
<p>Training course of CycleGAN on class 2.</p>
Full article ">Figure 6
<p>Comparison of original and generated data using CycleGAN to transform channel 1 to channel 2 ECG data.</p>
Full article ">
13 pages, 3417 KiB  
Article
Development and Field Application of an Automated Pipe Jacking Friction Resistance Identification and Warning System in Construction
by Liang-Hai Jin, Bang-Jie Wu, Xia-Zhong Zheng and Shu Chen
Appl. Sci. 2023, 13(23), 12814; https://doi.org/10.3390/app132312814 - 29 Nov 2023
Cited by 1 | Viewed by 1316
Abstract
Minimizing frictional resistance is crucial for ensuring the safety and smooth progress of pipe jacking construction. However, due to the unpredictability of geological conditions, it is difficult to grasp the frictional resistance during construction, which poses challenges to safe and smooth construction. In [...] Read more.
Minimizing frictional resistance is crucial for ensuring the safety and smooth progress of pipe jacking construction. However, due to the unpredictability of geological conditions, it is difficult to grasp the frictional resistance during construction, which poses challenges to safe and smooth construction. In order to reduce the frictional resistance during the process of pipe jacking, an automated pipe jacking friction resistance identification and warning system is thus innovatively proposed. This system uses jacking resistance sensors to identify resistance during the jacking process. When the jacking resistance exceeds a certain threshold, the system will send alerts, which could prompt construction workers to adjust the rheological slurry ratio according to the on-site soil conditions. This system includes the following major components: (1) an analysis of primary factors influencing pipe frictional resistance and a model for resistance calculation, (2) the examination of forces exerted on disturbed soil during pipe jacking construction to determine the optimal placement of resistance sensors, (3) the design and operational principles for an automated resistance identification and warning system, and (4) the application of a slurry shield construction method for resistance reduction. The research has practical significance in providing a reference for developing intelligent pipe jacking and contributing to the improvement in construction safety levels. Full article
(This article belongs to the Section Civil Engineering)
Show Figures

Figure 1

Figure 1
<p>(<b>a</b>) Full contact schematic diagram and (<b>b</b>) pipe–soil partial contact model in grouting.</p>
Full article ">Figure 2
<p>Persson contact plane strain model.</p>
Full article ">Figure 3
<p>Longitudinal disturbance soil partition and resistance sensor installation location.</p>
Full article ">Figure 4
<p>Lateral disturbance soil partition and resistance sensor installation location.</p>
Full article ">Figure 5
<p>Fundamental structure of the system.</p>
Full article ">Figure 6
<p>Diagram of the automatic resistance identification and warning system.</p>
Full article ">Figure 7
<p>Diagram of the resistance threshold pre-alarm control process.</p>
Full article ">Figure 8
<p>WX2–WX13 stratigraphic section.</p>
Full article ">Figure 9
<p>Resistance values.</p>
Full article ">
23 pages, 7828 KiB  
Article
Methodologies and Challenges for Optimal Sensor Placement in Historical Masonry Buildings
by Estefanía Chaves, Alberto Barontini, Nuno Mendes, Víctor Compán and Paulo B. Lourenço
Sensors 2023, 23(23), 9304; https://doi.org/10.3390/s23239304 - 21 Nov 2023
Cited by 1 | Viewed by 1317
Abstract
As ageing structures and infrastructures become a global concern, structural health monitoring (SHM) is seen as a crucial tool for their cost-effective maintenance. Promising results obtained for modern and conventional constructions suggested the application of SHM to historical masonry buildings as well. However, [...] Read more.
As ageing structures and infrastructures become a global concern, structural health monitoring (SHM) is seen as a crucial tool for their cost-effective maintenance. Promising results obtained for modern and conventional constructions suggested the application of SHM to historical masonry buildings as well. However, this presents peculiar shortcomings and open challenges. One of the most relevant aspects that deserve more research is the optimisation of the sensor placement to tackle well-known issues in ambient vibration testing for such buildings. The present paper focuses on the application of optimal sensor placement (OSP) strategies for dynamic identification in historical masonry buildings. While OSP techniques have been extensively studied in various structural contexts, their application in historical masonry buildings remains relatively limited. This paper discusses the challenges and opportunities of OSP in this specific context, analysing and discussing real-world examples, as well as a numerical benchmark application to illustrate its complexities. This article aims to shed light on the progress and issues associated with OSP in masonry historical buildings, providing a detailed problem formulation, identifying ongoing challenges and presenting promising solutions for future improvements. Full article
(This article belongs to the Special Issue Feature Papers in Fault Diagnosis & Sensors 2023)
Show Figures

Figure 1

Figure 1
<p>Location of the European case studies.</p>
Full article ">Figure 2
<p>Number of candidate locations per direction for each case. Logarithmic scale. In Saint John Cathedral, the resultant of the three directions was used. No available information for the Slottsfjell tower.</p>
Full article ">Figure 3
<p>Number of global and local modes used in the optimisation in each case study.</p>
Full article ">Figure 4
<p>Number of sensors selected. Dark orange represents the cases where the number of sensors was optimised. Light orange represents the cases where the minimum number was pre-defined.</p>
Full article ">Figure 5
<p>Candidate nodes (black: 113-node selection; black and blue: 179-node selection).</p>
Full article ">Figure 6
<p>Modal shapes of the reference scenario used for the optimisation: (<b>a</b>) mode 1; (<b>b</b>) mode 2; (<b>c</b>) mode 3; (<b>d</b>) mode 4; (<b>e</b>) mode 5; (<b>f</b>) mode 6; (<b>g</b>) mode 9. Maximum displacemnt represented by red, zero displacement represented by blue.</p>
Full article ">Figure 7
<p>OSP results. Outcomes of the 10-sensor optimisation are shown with red arrows, with the five additional sensors (15-sensor optimisation) in green: (<b>a</b>) 179-node EfI method; (<b>b</b>) 179-node offMAC method; (<b>c</b>) 113-node EfI method; (<b>d</b>) 113-node offMAC method.</p>
Full article ">Figure 8
<p>Auto-MAC matrix of the reference scenario, considering all the modes up to 5 Hz and the EfI 10-sensor 179-node optimisation. MAC values represented by colour scale and number.</p>
Full article ">Figure 9
<p>Modal shapes of the reference scenario: (<b>a</b>) mode 1; (<b>b</b>) mode 8; (<b>c</b>) mode 12. Maximum displacemnt represented by red, zero displacement represented by blue.</p>
Full article ">Figure 10
<p>Auto-MAC tables considering all the modes under 5 Hz and the EfI 10-sensors 179-node optimisation for the reference scenario: (<b>a</b>) SCN01; (<b>b</b>) SCN02; (<b>c</b>) SCN03; (<b>d</b>) SCN04. MAC values represented by colour scale: red represents 1, white represents 0.</p>
Full article ">Figure 11
<p>Cross-MAC tables. Comparison between the reference scenario and (<b>a</b>) SCN01, (<b>b</b>) SCN02, (<b>c</b>) SCN03 and (<b>d</b>) SCN04. MAC values represented by colour scale: red represents 1, white represents 0.</p>
Full article ">Figure 12
<p>OSP results of the alternative scenarios: (<b>a</b>) SCN01; (<b>b</b>) SCN02; (<b>c</b>) SCN03; (<b>d</b>) SCN04.</p>
Full article ">Figure 13
<p>Auto-MAC tables considering the result from the offMAC 15-sensor 179-node optimisation of (<b>a</b>) SCN01, (<b>b</b>) SCN02, (<b>c</b>) SCN03 and (<b>d</b>) SCN04. MAC values represented by colour scale: red represents 1, white represents 0.</p>
Full article ">
21 pages, 15407 KiB  
Article
Iterative-Based Impact Force Identification on a Bridge Concrete Deck
by Maria Rashidi, Shabnam Tashakori, Hamed Kalhori, Mohammad Bahmanpour and Bing Li
Sensors 2023, 23(22), 9257; https://doi.org/10.3390/s23229257 - 18 Nov 2023
Viewed by 1357
Abstract
Steel-reinforced concrete decks are prominently utilized in various civil structures such as bridges and railways, where they are susceptible to unforeseen impact forces during their operational lifespan. The precise identification of the impact events holds a pivotal role in the robust health monitoring [...] Read more.
Steel-reinforced concrete decks are prominently utilized in various civil structures such as bridges and railways, where they are susceptible to unforeseen impact forces during their operational lifespan. The precise identification of the impact events holds a pivotal role in the robust health monitoring of these structures. However, direct measurement is not usually possible due to structural limitations that restrict arbitrary sensor placement. To address this challenge, inverse identification emerges as a plausible solution, albeit afflicted by the issue of ill-posedness. In tackling such ill-conditioned challenges, the iterative regularization technique known as the Landweber method proves valuable. This technique leads to a more reliable and accurate solution compared with traditional direct regularization methods and it is, additionally, more suitable for large-scale problems due to the alleviated computation burden. This paper employs the Landweber method to perform a comprehensive impact force identification encompassing impact localization and impact time–history reconstruction. The incorporation of a low-pass filter within the Landweber-based identification procedure is proposed to augment the reconstruction process. Moreover, a standardized reconstruction error metric is presented, offering a more effective means of accuracy assessment. A detailed discussion on sensor placement and the optimal number of regularization iterations is presented. To automatedly localize the impact force, a Gaussian profile is proposed, against which reconstructed impact forces are compared. The efficacy of the proposed techniques is illustrated by utilizing the experimental data acquired from a bridge concrete deck reinforced with a steel beam. Full article
(This article belongs to the Special Issue Advanced Sensing for Mechanical Vibration and Fault Diagnosis)
Show Figures

Figure 1

Figure 1
<p>Error schematic showing asymptotic and semi-convergence phenomena [<a href="#B35-sensors-23-09257" class="html-bibr">35</a>].</p>
Full article ">Figure 2
<p>Experimental setup: a concrete deck reinforced with a steel beam with accelerometers placed at positions S<sub>i</sub> (i = 1, …, 10), and the potential impact locations labeled as L<sub>j</sub> (j = 1, …, 7).</p>
Full article ">Figure 3
<p>Cross-sectional view of the concrete deck setup.</p>
Full article ">Figure 4
<p>Reconstruction of the impact force applied at L<sub>1</sub>, using accelerometers placed at S<sub>1</sub> to S<sub>10</sub>.</p>
Full article ">Figure 5
<p>Reconstruction of the impact force applied at L<sub>2</sub>, using accelerometers placed at S<sub>1</sub> to S<sub>10</sub>.</p>
Full article ">Figure 6
<p>Reconstruction of the impact force applied at L<sub>3</sub>, using accelerometers placed at S<sub>1</sub> to S<sub>10</sub>.</p>
Full article ">Figure 7
<p>Reconstruction of the impact force applied at L<sub>4</sub>, using accelerometers placed at S<sub>1</sub> to S<sub>10</sub>.</p>
Full article ">Figure 8
<p>Reconstruction of the impact force applied at L<sub>5</sub>, using accelerometers placed at S<sub>1</sub> to S<sub>10</sub>.</p>
Full article ">Figure 9
<p>Reconstruction of the impact force applied at L<sub>6</sub>, using accelerometers placed at S<sub>1</sub> to S<sub>10</sub>.</p>
Full article ">Figure 10
<p>Reconstruction of the impact force applied at L<sub>7</sub>, using accelerometers placed at S<sub>1</sub> to S<sub>10</sub>.</p>
Full article ">Figure 11
<p>Correlation error for initial and filtered reconstruction at different locations using accelerometer S<sub>1</sub>.</p>
Full article ">Figure 12
<p>Peak error for initial and filtered reconstruction at different locations using accelerometer S<sub>1</sub>.</p>
Full article ">Figure 13
<p>Reconstruction error at impact location L<sub>1</sub> for different number of iterations in Landweber regularization, using accelerometers placed at S<sub>1</sub> to S<sub>10</sub>.</p>
Full article ">Figure 14
<p>Reconstruction error at impact location L<sub>2</sub> for different number of iterations in Landweber regularization, using accelerometers placed at S<sub>1</sub> to S<sub>10</sub>.</p>
Full article ">Figure 15
<p>Reconstruction error at impact location L<sub>3</sub> for different number of iterations in Landweber regularization, using accelerometers placed at S<sub>1</sub> to S<sub>10</sub>.</p>
Full article ">Figure 16
<p>Reconstruction error at impact location L<sub>4</sub> for different number of iterations in Landweber regularization, using accelerometers placed at S<sub>1</sub> to S<sub>10</sub>.</p>
Full article ">Figure 17
<p>Reconstruction error at impact location L<sub>5</sub> for different number of iterations in Landweber regularization, using accelerometers placed at S<sub>1</sub> to S<sub>10</sub>.</p>
Full article ">Figure 18
<p>Reconstruction error at impact location L<sub>6</sub> for different number of iterations in Landweber regularization, using accelerometers placed at S<sub>1</sub> to S<sub>10</sub>.</p>
Full article ">Figure 19
<p>Reconstruction error at impact location L<sub>7</sub> for different number of iterations in Landweber regularization, using accelerometers placed at S<sub>1</sub> to S<sub>10</sub>.</p>
Full article ">Figure 20
<p>Identification of impact location using accelerometer S<sub>7</sub> at all impact locations L<sub>1</sub> to L<sub>7</sub> with the correlation errors (1-cv)% reported.</p>
Full article ">Figure 21
<p>Identification of impact location using accelerometer S<sub>4</sub> at true impact location L<sub>4</sub> with the correlation errors (1-cv)% reported.</p>
Full article ">
23 pages, 7523 KiB  
Article
Efficient Modal Identification and Optimal Sensor Placement via Dynamic DIC Measurement and Feature-Based Data Compression
by Weizhuo Wang
Vibration 2023, 6(4), 820-842; https://doi.org/10.3390/vibration6040050 - 6 Oct 2023
Viewed by 2424
Abstract
Full-field non-contact vibration measurements provide a rich dataset for analysing structural dynamics. However, implementing the identification algorithm directly using high-spatial resolution data can be computationally expensive in modal identification. To address this challenge, performing identification in a shape-preserving but lower-dimensional feature space is [...] Read more.
Full-field non-contact vibration measurements provide a rich dataset for analysing structural dynamics. However, implementing the identification algorithm directly using high-spatial resolution data can be computationally expensive in modal identification. To address this challenge, performing identification in a shape-preserving but lower-dimensional feature space is more feasible. The full-field mode shapes can then be reconstructed from the identified feature mode shapes. This paper discusses two approaches, namely data-dependent and data-independent, for constructing the feature spaces. The applications of these approaches to modal identification on a curved plate are studied, and their performance is compared. In a case study involving a curved plate, it was found that a spatial data compression ratio as low as 1% could be achieved without compromising the integrity of the shape features essential for a full-field modal. Furthermore, the paper explores the optimal point-wise sensor placement using the feature space. It presents an alternative, data-driven method for optimal sensor placement that eliminates the need for a normal model, which is typically required in conventional approaches. Combining a small number of point-wise sensors with the constructed feature space can accurately reconstruct the full-field response. This approach demonstrates a two-step structural health monitoring (SHM) preparation process: offline full-field identification of the structure and the recommended point-wise sensor placement for online long-term monitoring. Full article
Show Figures

Figure 1

Figure 1
<p>Experimental setup.</p>
Full article ">Figure 2
<p>Domain/grids from a DIC measurement of the PP plate.</p>
Full article ">Figure 3
<p>The constructed basis function from principal directions by SVD. The subplots are ordered by the descending singular values of <math display="inline"><semantics> <mrow> <mi>σ</mi> </mrow> </semantics></math>.</p>
Full article ">Figure 4
<p>The constructed orthonormal basis function from 2-monomial on the DIC measured domains. (m,n) are orders of the power of the monomials in the two perpendicular directions.</p>
Full article ">Figure 5
<p>Comparison of full-field displacement at each temporal step between the DIC measurement (2320 spatial points at each time step) and the reconstructed displacement field from only 25 terms of data-independent shape descriptors (AGMD). The shape compression ratio is about 1% (25/2320). The 25-dimensional sub-space accurately approximates the measured data with 2320 dimensions as the correlation between the two is always greater than 99.65%, with most steps greater than 99.99%.</p>
Full article ">Figure 6
<p>Reconstructed mode shapes from eigenvectors of 20 principal components. (<b>a</b>) mode shapes; (<b>b</b>) nodal lines.</p>
Full article ">Figure 7
<p>autoMAC of the reconstructed mode shapes from eigenvectors of principal components.</p>
Full article ">Figure 8
<p>Reconstructed mode shapes from eigenvectors of 25 AGMDs. (<b>a</b>) mode shapes; (<b>b</b>) nodal lines.</p>
Full article ">Figure 9
<p>autoMAC of the reconstructed mode shapes from eigenvectors of AGMDs.</p>
Full article ">Figure 10
<p>Reconstructed mode shapes MAC values between AGMDs and principal components.</p>
Full article ">Figure 11
<p>Nodal lines (in red) from the AGMD reconstructed modes, and the proposed point wise sensors placements as annotated as pentagrams (star symbol).</p>
Full article ">Figure 12
<p>Full-field mode shape expanded from point-wise sparse sensors with basis functions.</p>
Full article ">Figure 13
<p>Reconstructed mode shapes’ MAC between AGMDs and expanded from point-wise sparse sensor.</p>
Full article ">
Back to TopTop