Number Of Layers Research Articles

Influence of HTS tape arrangement on the transverse compression performance of copper former CORC cables

Abstract CORC cables are subject to large transverse compression electromagnetic forces in fusion projects. Unfortunately, the electromagnetic force exceeding its critical transverse compression load will cause an irreversible decrease in its critical current. Therefore, it is particularly important to enhance the critical transverse compression load to ensure that the critical current does not decrease during operation. The winding method of HTS (High tempreture superconducting) tape on the central former is variable. So the experimental study on how to increase the critical transverse compression load of CORC cable by changing the winding method of HTS tape is carried out in this paper. Firstly, the influence law of parameters of the number of HTS tapes per layer and the number of HTS tape layers on their transverse compression performance are analysed independently. The results indicate that increasing the number of HTS tapes per layer and the number of HTS tape layers can both improve the transverse compression performance of CORC cables. Whereas, in the case of a cable with a certain critical current demand (the same total number of HTS tapes), increasing the number of HTS tape layers necessarily reduces the number of HTS tapes per layer. Therefore, in order to compare the degree of influence of the above two parameters, we conducted transverse compression experiments on multiple groups of CORC cables with different winding methods (more layers with few tapes per layer or few layers with more tapes per layer) under the same critical current demand. The results show that under the same critical current demand, choosing the winding method that reduces the number of HTS tape layers and increases the number HTS tapes per layer can effectively improve the transverse compression performance of CORC cables. A 3D multilayer CORC cable transverse compression finite element (FE) model is also established to explain the inherent reasons for the differences in transverse compression performance of CORC cables under different HTS tape winding methods.

Carrier transport simulations in twisted bilayer and turbostratic multilayer graphene systems

Abstract The effects of inserting a twisted bilayer graphene (tBLG) between turbostratic graphene layers and a SiO2 substrate on the transport properties are investigated, to explore the possibility that using a tBLG could be an effective way to screen the potential fluctuations due to the impurities on the substrate. The Fermi velocity in a tBLG changes from that of the pristine graphene depending on the twist angle. In the present study, a parameter α is introduced, which is defined as the ratio of the Fermi velocity in the tBLG to that of pristine graphene, and the transport properties are calculated as a function of α. The self-consistent calculation of the Poisson equation with the Thomas-Fermi approximation is performed for multilayer graphene systems consisting of a tBLG and turbostratic graphene layers, and the potential profiles are incorporated into a Monte Carlo simulator to calculate the drift velocities and mobilities. It is shown that the transport properties of the whole system strongly depend on the parameter α and the number of layers in the system.

Effects of Vaccination against Recombinant FSH or LH Receptor Subunits on Gonadal Development and Functioning Male Rats

Luteinizing hormone (LH) and follicle-stimulating hormone (FSH) play key roles in regulating testosterone secretion and spermatogenesis in male mammals, respectively, and they maintain the fertility of male animals by binding to their corresponding receptors. We designed and prepared a recombinant LH receptor (LHR) subunit vaccine and a recombinant FSH receptor (FSHR) subunit vaccine and used male Sprague Dawley (SD) rats as a model to examine their effects on testicular development, spermatogenesis, and testosterone secretion in prepubertal and pubertal mammals. Both vaccines (LHR-DTT and FSHR-DTT) significantly decreased the serum testosterone level in prepubertal rats (p < 0.05) but had no effect on the testosterone secretion in pubertal rats; both vaccines decreased the number of cell layers in the seminiferous tubules and reduced spermatogenesis in prepubertal and pubertal rats. Subunit vaccine FSHR-DTT decreased the sperm density in the epididymis in both prepubertal and pubertal rats (p < 0.01) and lowered testicular index and sperm motility in pubertal rats (p < 0.05), whereas LHR-DTT only reduced the sperm density in the epididymis in pubertal rats (p < 0.05). These results indicate that the FSHR subunit vaccine may be a promising approach for immunocastration, but it still needs improvements in effectiveness.

Numerical and experimental 3-point bending quasi-static flexural behavior of single-strap joints joined using AF163-2K adhesive films

ABSTRACT This paper evaluated the mechanical characteristics of adhesive-single-bonded strap joints with varied numbers of adhesive layers subjected to flexural loads. Six samples were submitted to a 3-point bending test, three with the strap facing up and three with the strap facing down, and the failure behavior was analyzed. A 3-D finite element (FE) model was created in ABAQUS/Explicit. The adhesive layer (AF163-2K) failure was represented by cohesive zone components. It was discovered that the samples with the thickest adhesive layer obtained the highest load values. Furthermore, compared to samples with the strap facing down, those with the strap facing up demonstrated greater strength. Experiments were conducted to validate the constructed numerical model. Given that this type of configuration is one of the most commonly employed, especially in the repair of aeronautical structures, studying single-strap joints is deemed beneficial for designers.

Optically driven plasmons in graphene/hBN van der Waals heterostructures: simulating s-SNOM measurements

Abstract Converting transverse photons into longitudinal two-dimensional plasmon-–polaritons (2D-PP) and vice versa presents a significant challenge within the fields of photonics and plasmonics. Therefore, understanding the mechanism which increases the photon – 2D-PP conversion efficiency could significantly contribute to those efforts. In this study, we theoretically examine how efficiently incident radiation, when scattered by a silver spherical nanoparticle (Ag-NP), can be transformed into 2D-PP within van der Waals (vdW) heterostructures composed of hexagonal boron nitride and graphene (hBN/Gr composites). We show that the Dirac plasmon (DP) excitation efficiency depends on the Ag-NP radius as R 3, and decreases exponentially with Ag-NP height h, so that for a certain Ag-NP geometry up to 25 % of the incident electrical field is channeled into the DP. We demonstrate that the linear plasmons (LPs) excitation efficiency can be manipulated by changing the graphene–graphene distance Δ (or hBN thickness) or by changing the number of graphene layers N. By increasing Δ and/or N the LPs move towards smaller wave vectors Q and become accessible by the Ag-NP dipole field, so that for N ≥ 5 the excitation of more than one LP is possible. These results are supported by recent scattering-type scanning near-field optical microscopy (s-SNOM) measurements. Furthermore, we show that Ag-NPs with specific parameters preferentially hybridizes with DPs of a particular wavelength λ D , facilitating selective excitation of DPs. The obtained tuning possibilities could have a significant impact on applied plasmonics, photonics or optoelectronics.

The application of improved densenet algorithm in accurate image recognition.

Image recognition technology belongs to an important research field of artificial intelligence. In order to enhance the application value of image recognition technology in the field of computer vision and improve the technical dilemma of image recognition, the research improves the feature reuse method of dense convolutional network. Based on gradient quantization, traditional parallel algorithms have been improved. This improvement allows for independent parameter updates layer by layer, reducing communication time and data volume. The introduction of quantization error reduces the impact of gradient loss on model convergence. The test results show that the improvement strategy designed by the research improves the model parameter efficiency while ensuring the recognition effect. Narrowing the learning rate is conducive to refining the updating granularity of model parameters, and deepening the number of network layers can effectively improve the final recognition accuracy and convergence effect of the model. It is better than the existing state-of-the-art image recognition models, visual geometry group and EfficientNet. The parallel acceleration algorithm, which is improved by the gradient quantization, performs better than the traditional synchronous data parallel algorithm, and the improvement of the acceleration ratio is obvious. Compared with the traditional synchronous data parallel algorithm and stale synchronous parallel algorithm, the optimized parallel acceleration algorithm of the study ensures the image data training speed and solves the bottleneck problem of communication data. The model designed by the research improves the accuracy and training speed of image recognition technology and expands the use of image recognition technology in the field of computer vision.Please confirm the affiliation details of [1] is correct.The relevant detailed information in reference [1] has been confirmed to be correct.

In Situ Monitoring and Tuning Multilayer Stacking of Polymer Lamellar Crystals in Solution with Aggregation-Induced Emission.

Many types of self-assembled 2D materials with fascinating morphologies and novel properties have been prepared and used in solution. However, it is still a challenge to monitor their in situ growth in solution and to control the number of layers in these materials. Here, we demonstrate that the aggregation-induced emission (AIE) effect can be applied for the in situ decoupled tracing of the lateral growth and multilayer stacking of polymer lamellar crystals in solution. Multilayer stacking considerably enhances the photoluminescence intensity of the AIE molecules sandwiched between two layers of lamellar crystals, which is 2.4 times that on the surface of monolayer crystals. Both variation of the self-seeding temperature of crystal seeds and addition of a trace amount of long polymer chains during growth can control multilayer lamellar stacking, which are applied to produce tunable fluorescent patterns for functional applications.

Direct Synthesis of Layer‐Tunable and Transfer‐Free Graphene on Device‐Compatible Substrates Using Ion Implantation Toward Versatile Applications

Direct synthesis of layer‐tunable and transfer‐free graphene on technologically important substrates is highly valued for various electronics and device applications. State of the art in the field is currently a two‐step process: a high‐quality graphene layer synthesis on metal substrate through chemical vapor deposition (CVD) followed by delicate layer transfer onto device‐relevant substrates. Here, we report a novel synthesis approach combining ion implantation for a precise graphene layer control and dual‐metal smart Janus substrate for a diffusion‐limiting graphene formation to directly synthesize large area, high quality, and layer‐tunable graphene films on arbitrary substrates without the post‐synthesis layer transfer process. Carbon (C) ion implantation was performed on Cu–Ni film deposited on a variety of device‐relevant substrates. A well‐controlled number of layers of graphene, primarily monolayer and bilayer, is precisely controlled by the equivalent fluence of the implanted C‐atoms (1 monolayer ~4 × 1015 C‐atoms/cm2). Upon thermal annealing to promote Cu‐Ni alloying, the pre‐implanted C‐atoms in the Ni layer are pushed toward the Ni/substrate interface by the top Cu layer due to the poor C‐solubility in Cu. As a result, the expelled C‐atoms precipitate into a graphene structure at the interface facilitated by the Cu‐like alloy catalysis. After removing the alloyed Cu‐like surface layer, the layer‐tunable graphene on the desired substrate is directly realized. The layer‐selectivity, high quality, and uniformity of the graphene films are not only confirmed with detailed characterizations using a suite of surface analysis techniques but more importantly are successfully demonstrated by the excellent properties and performance of several devices directly fabricated from these graphene films. Molecular dynamics (MD) simulations using the reactive force field (ReaxFF) were performed to elucidate the graphene formation mechanisms in this novel synthesis approach. With the wide use of ion implantation technology in the microelectronics industry, this novel graphene synthesis approach with precise layer‐tunability and transfer‐free processing has the promise to advance efficient graphene‐device manufacturing and expedite their versatile applications in many fields.

Experimental investigation of dynamic bond behaviors between LRS-FRP and concrete

The dynamic bonding behaviour at the interface between a large-rupture-strain fibre-reinforced polymer (LRS-FRP) and concrete was investigated via drop weight impact tests of notched beam specimens. The main experimental variables included the concrete strength grade, number of LRS-FRP layers, and loading rate. With increasing loading rate, the location of the peeling damage at the interface between the LRS-FRP and concrete shifted from the interior of the concrete layer to the concrete–binder layer interface, binder layer, and even binder layer–FRP interface. The ultimate debonding load of the LRS-FRP strips increased with increasing strain rate, reaching more than 1.7 times the static value at a strain rate of approximately 4/s, while the effective bond length did not show a clear increasing or decreasing tendency with increasing strain rate. For the local bond-slip relationship, both the peak interfacial bond stress and interfacial fracture energy increased with increasing strain rate, and the influences of the concrete strength and number of FRP layers on the strain rate sensitivity were insignificant. Based on the experimental results, improved models to predict the ultimate debonding load and effective bond length at the LRS-FRP–concrete interface under both static and dynamic loading conditions were proposed. Furthermore, dynamic increase factor (DIF) formulas for the peak interfacial bond stress and interfacial fracture energy for bond-slip modelling of the interface between a polyethylene naphthalate (PEN)-FRP and concrete were established.

Numerical Investigation on Strengthening of Steel Beams for Corrosion Damage or Web Openings Using Carbon Fiber Reinforced Polymer Sheets

Fiber-reinforced polymers (FRPs) have been widely used to strengthen steel structures, which could suffer from corrosion or the introduction of web openings, for utilities such as ductwork, plumbing, electrical conduits, and HVAC systems. The present numerical study involves the application of unidirectional carbon FRP (CFRP) sheets to steel I-beams, damaged due to corrosion or web openings, to regain their lost load-carrying capacity. Finite element analysis (FEA) was utilized to develop and validate three beam models against existing experimentally tested specimens. Subsequently, a parametric study was conducted investigating the effect of various corrosion levels and the number of circular web openings on the yield and ultimate load capacities of the beams. The optimum number of CFRP layers needed to strengthen corroded beams was determined and six CFRP strengthening scenarios were adopted to determine the best configurations to retrofit steel beams with openings (SBWOs). The results revealed that corrosion, introduced by thinning the bottom flange, reduced both yield and ultimate load capacities, with a nearly perfect linear reduction in ultimate load for each 2.5% thickness loss. The optimum number of CFRP layers depended on the level of corrosion damage. Furthermore, while maintaining a constant total opening area, beams with a greater number of smaller circular web openings demonstrated higher yield and ultimate load capacities than those with fewer larger openings. Out of the six adopted CFRP strengthening scenarios, three configurations that involved applying CFRP sheets to both flanges and the web effectively restored the strength of SBWOs, when adequate CFRP layers were used.

Deep-learning-assisted design of multi-degree-of-freedom metamaterial absorber

In recent years, the development of metamaterial absorbers (MAs) based on deep learning methods has become a popular research topic. Nevertheless, conventional analytical approaches have limitations, leading to MAs characterized by low-degree-of-freedom cell structures, thereby restricting their absorption characteristics. In this study, we propose a multi-degree-of-freedom MA with multiple variable parameters in terms of structures, materials, and number of layers, which are distinguished by coding. Additionally, deep learning models, including a fully connected neural network, recurrent neural network, AlexNet, and residual neural network, are designed to predict structures and absorption spectra. We compare the prediction accuracy of these four neural network models and identify the optimal network model for spectral prediction and on-demand design. Using trained neural network models, we successfully designed broadband, dual-band, and single-band MAs in on-demand design. Compared with other design methods, this method provides a greater degree of freedom in device design, which allows the design of MAs to have more absorption characteristics

A Performance Comparison of Japanese Sign Language Recognition with ViT and CNN Using Angular Features

In recent years, developments in deep learning technology have driven significant advancements in research aimed at facilitating communication with individuals who have hearing impairments. The focus has been on enhancing automatic recognition and translation systems for sign language. This study proposes a novel approach using a vision transformer (ViT) for recognizing Japanese Sign Language. Our method employs a pose estimation library, MediaPipe, to extract the positional coordinates of each finger joint within video frames and generate one-dimensional angular feature data from these coordinates. Then, the code arranges these feature data in a temporal sequence to form a two-dimensional input vector for the ViT model. To determine the optimal configuration, this study evaluated recognition accuracy by manipulating the number of encoder layers within the ViT model and compared against traditional convolutional neural network (CNN) models to evaluate its effectiveness. The experimental results showed 99.7% accuracy for the method using the ViT model and 99.3% for the results using the CNN. We demonstrated the efficacy of our approach through real-time recognition experiments using Japanese sign language videos.

Effect of Particle-Substrate Interactions on Colloidal Layer Structure Prepared by Convective Self-Assembly Using Polyelectrolyte-Grafted Silica Particles.

Cationic and anionic polyelectrolytes, poly(vinylbenzyl trimethylammonium chloride) (PVBTA) and poly(sodium styrene sulfate) (PSSS), were grafted on the surface of the silica particles, respectively, and then these two types of polyelectrolyte-grafted silica particles were applied to the colloidal layer preparation by convective self-assembly (CSA) using hydrophilic or hydrophobic glass substrates to investigate the effect of the interactions between the particles and the substrate surface on the resultant layer structures. When the PVBTA-grafted silica particle (PVBTA-Si) was used, the colloidal monolayers with a non-close-packed (NCP) structure were formed on both hydrophilic and hydrophobic glass substrates, where the NCP colloidal layers on the hydrophobic glass substrate have a somewhat more ordered structure than those on the hydrophilic glass substrate. In the case of the PSSS-grafted silica particle (PSSS-Si), on the other hand, stripe patterns with close-packed (CP) colloidal layers were obtained on both types of the glass substrates. The number of layers of the stripes on the hydrophilic glass substrate was less than that on the hydrophobic glass substrate, while the spacing and width of the stripes on both substrates were similar to each other. The difference in the structures of the colloidal layers obtained here indicates that an attractive interaction, such as an electrostatic attraction and a hydrophobic interaction, between the particle and the substrate surface is necessary to achieve the NCP structure by the CSA process using polyelectrolyte-grafted silica particles.

Restoration of two-span RC beams with severely corroded reinforcement using C-FRCM composites

The feasibility and viability of using carbon fabric-reinforced cementitious matrix (C-FRCM) composites for restoration of continuous reinforced concrete (RC) beams with severely corroded reinforcement were examined. Five large-scale dual-span RC beams were fabricated and tested. One beam was not corroded to serve as a control. Four beams were subjected to accelerated corrosion that corresponded to a tensile steel cross-sectional loss of 40% in either the sagging or the hogging region. Such a steel cross-sectional loss would cause a comparable reduction in the yield load of bare steel reinforcing bars. Two of the corroded beams were repaired with C-FRCM composites. An analytical model was formulated and utilized to design the number of C-FRCM layers required to fully restore the original load capacity. Experimental results verified the redistribution of moments between the sagging and hogging regions due to yielding of the tensile steel at critical sections. Both sagging and hogging sections contributed to the load-bearing capacity, and hence, the reduction in the load capacity caused by corrosion in one of the critical sections was smaller than the tension steel cross-sectional loss caused by corrosion. It was possible to predict the load capacity of the tested beams with reasonable accuracy considering the residual cross-sectional area of the corroded tension steel bars while keeping their yield strength and Young’s modulus unaltered in addition to implementing other conventional assumptions adopted by international codes and guidelines. As predicted analytically and verified experimentally, rehabilitation of corroded continuous beams with appropriate number of C-FRCM composite layers fully restored their original load capacity. Research findings would assist practitioners in conducting a proper analysis of continuous RC beams with corroded reinforcement before and after repair with C-FRCM composites.

Thickness-Dependent Nanoscale Elastic Stiffening of Chemical Vapor Deposited Atomically Thin 2H-MoS2 Films.

Understanding the nanoscale elastic-size-effects of atomically thin transition-metal dichalcogenides (TMDs) as a function of thickness underpins the avenue of flexible 2D electronics. In this work, we employed the atomic force acoustic microscopy (AFAM) technique to investigate the thickness-dependent elastic properties of CVD grown 2H-MoS2 films. The monolayer MoS2 exhibited a Young's modulus of 273 ± 27 GPa. Our systematic analysis from bulk to monolayer suggests that the 2H-MoS2 phase exhibits nanoscale elastic-stiffening behavior with decreasing number of layers (thickness). The Young's modulus increased by a factor of ∼2.7 for monolayer MoS2 when compared with the bulk. First-principle DFT calculations affirm the nanoscale elastic-stiffening behavior of MoS2 with decreasing number of layers. Our findings suggest that the observed elastic stiffening is due to the interlayer sliding, which may be facilitated by defects in MoS2 layers. The observed elastic stiffening may be of potential importance for understanding TMD based nanomechanical devices.

Stacked U-Net with Time–Frequency Attention and Deep Connection Net for Single Channel Speech Enhancement

Deep neural networks have significantly promoted the progress of speech enhancement technology. However, a great number of speech enhancement approaches are unable to fully utilize context information from various scales, hindering performance enhancement. To tackle this issue, we introduce a method called TFADCSU-Net (Stacked U-Net with Time-Frequency Attention (TFA) and Deep Connection Layer (DCL)) for enhancing noisy speech in the time–frequency domain. TFADCSU-Net adopts an encoder-decoder structure with skip links. Within TFADCSU-Net, a multiscale feature extraction layer (MSFEL) is proposed to effectively capture contextual data from various scales. This allows us to leverage both global and local speech features to enhance the reconstruction of speech signals. Moreover, we incorporate deep connection layer and TFA mechanisms into the network to further improve feature extraction and aggregate utterance level context. The deep connection layer effectively captures rich and precise features by establishing direct connections starting from the initial layer to all subsequent layers, rather than relying on connections from earlier layers to subsequent layers. This approach not only enhances the information flow within the network but also avoids a significant rise in computational complexity as the number of network layers increases. The TFA module consists of two attention branches operating concurrently: one directed towards the temporal dimension and the other towards the frequency dimension. These branches generate distinct forms of attention — one for identifying relevant time frames and another for selecting frequency wise channels. These attention mechanisms assist the models in discerning “where” and “what” to prioritize. Subsequently, the TA and FA branches are combined to produce a comprehensive attention map in two dimensions. This map assigns specific attention weights to individual spectral components in the time–frequency representation, enabling the networks to proficiently capture the speech characteristics in the T-F representation. The results confirm that the proposed method outperforms other models in terms of objective speech quality as well as intelligibility.

Hemodynamics in nutcracker syndrome: implications for diagnosis.

Nutcracker syndrome is a disease characterized by complex symptoms, making its diagnosis challenging and often delayed, often resulting in a painful experience for the patients. This study aimed to investigate the pathogenesis of nutcracker syndrome through the perspective of hemodynamics by simulating blood flow with varying compression degrees of the left renal vein. 3D patient-specific vascular models of the abdominal aorta, superior mesenteric artery and left renal vein were constructed based on CT images of patients suspected of having nutcracker syndrome. A hemodynamic simulation was then conducted using computational fluid dynamics to identify the correlation between alterations in hemodynamic parameters and varying degrees of compression. The study indicated the presence of anevident gradient in velocity distribution over the left renal vein with relatively high degrees of stenosis (α ≤ 50°), with maximum velocity in the central region of the stenosis. Additionally, when the compression degree of the left renal vein increases, the pressure distribution of the left renal vein presents an increasing number of gradient layers. Furthermore, the wall shear stress shows a correlation with the variation of blood flow velocity, i.e., the increase of wall shear stress correlates with the acceleration of the blood flow velocity. Using computational fluid dynamics as a non-invasive instrument to obtain the hemodynamic characteristics of nutcracker syndrome is feasible and could provide insights into the pathological mechanisms of thenutcracker syndrome supporting clinicians in diagnosis.

Thermal buckling analysis for hybrid and composite laminated plate by using new displacement function

Abstract The analysis of a critical buckling temperature is dependent on the new displacement function field provided in the literature and uses third-order shear deformation theory. The value of the parameter “m,” which determines the new displacement function, closed with the three-dimensional (3D) elasticity results. Thermal critical buckling is analyzed for thick and thin laminated plates by using the Navier solution for symmetric and anti-symmetric cross-ply simply supported boundary conditions. Several design parameters are considered such as extension thermal coefficient ratio (α 1⁄α 2), number of schemes, thickness ratio (a/h), aspect ratio (a/b), and modular ratio (E 1/E 2) when analyzing the dynamic behavior of the critical buckling temperature for different materials, including composite (glass/epoxy) and hybrid (glass/carbon/epoxy), under uniformly distributed temperature load. The accuracy for theoretical results by using Matlab R2019b was checked with other results for different researchers and gave good agreement. Increasing orthotropic ratio and aspect ratio resulted in increasing critical buckling temperature in M1 than M2, whereas increasing thickness ratio, thermal coefficient expansion, and number of layers resulted in decrease in critical buckling temperature in M2 than M1.

Modeling and characterization of deeply etched multilayer resonators under partial coherent excitation via multimode optical fibers

Abstract Recently, there has been a resurgence of interest in multimode optical fibers illuminated by a white light source. Largely, in anticipation of many integrated applications in the biomedical domain and spectral sensing benefiting from the broad spectral range and high numerical aperture. Along these lines, the output light from these fibers can be captured by the physics of partially coherent sources. While the Gaussian Schell model has provided a framework for studying partial coherence, to our knowledge, its impact on microstructures remains unexplored. As the sheer complexity arising from the interplay between partial coherence and microstructures transfer function has posed fundamental challenges in deciphering their response. In this work, we introduce a comprehensive numerical model paired with experimental validation to assess the performance of multilayer optical resonators, which are meticulously crafted through high aspect ratio silicon etching under the influence of a partially coherent optical source. The model studies the effects of optical fiber numerical aperture, Bragg mirror order, cavity length, and surface roughness of the microstructures on the output of the resonator. The results show that the response under standard multimode fiber (partial coherent source) has lower insertion loss, more asymmetry versus wavelength, and larger full width at half maximum (FWHM) than the standard single mode fiber (full coherent source). A silicon-on-insulator chip is fabricated using 130-µm deep etching of silicon for Bragg mirrors with 2.25, 3, and 3.25 µm silicon layer widths and a different number of layers. The structures are characterized using a multimode fiber of 62.5-µm core diameter illuminated by an infrared white light source. The theoretical results have been compared with the experimental results and a good agreement has been obtained.

A Hybrid Soft Sensor Model for Measuring the Oxygen Content in Boiler Flue Gas.

As an indispensable component of coal-fired power plants, boilers play a crucial role in converting water into high-pressure steam. The oxygen content in the flue gas is a crucial indicator, which indicates the state of combustion within the boiler. The oxygen content not only affects the thermal efficiency of the boiler and the energy utilization of the generator unit, but also has adverse impacts on the environment. Therefore, accurate measurement of the flue gas's oxygen content is of paramount importance in enhancing the energy utilization efficiency of coal-fired power plants and reducing the emissions of waste gas and pollutants. This study proposes a prediction model for the oxygen content in the flue gas that combines the whale optimization algorithm (WOA) and long short-term memory (LSTM) networks. Among them, the whale optimization algorithm (WOA) was used to optimize the learning rate, the number of hidden layers, and the regularization coefficients of the long short-term memory (LSTM). The data used in this study were obtained from a 350 MW power generation unit in a coal-fired power plant to validate the practicality and effectiveness of the proposed hybrid model. The simulation results demonstrated that the whale optimization algorithm-long short-term memory (WOA-LSTM) model achieved an MAE of 0.16493, an RMSE of 0.12712, an MAPE of 2.2254%, and an R2 value of 0.98664. The whale optimization algorithm-long short-term memory (WOA-LSTM) model demonstrated enhancements in accuracy compared with the least squares support vector machine (LSSVM), long short-term memory (LSTM), particle swarm optimization-least squares support vector machine (PSO-LSSVM), and particle swarm optimization-long short-term memory (PSO-LSTM), with improvements of 4.93%, 4.03%, 1.35%, and 0.49%, respectively. These results indicated that the proposed soft sensor model exhibited more accurate performance, which can meet practical requirements of coal-fired power plants.