Large Complex Research Articles

Deep knowledge transfer powered ultrasonic guided wave damage monitoring under incomplete database scenarios: Theories, applications and challenges

Abstract Ultrasonic guided waves can travel long distances within the detected structures, which is of great significance for monitoring large complex engineering systems. However, the multimodal and dispersive properties of the specific research object making this promising whole structure monitoring difficult to interpret the signal mathematically and physically. With the development and maturity of deep learning and big data mining technologies, many scholars have noticed artificial intelligence algorithms such as deep learning can provide a new tool in ultrasonic guided wave signal processing, avoiding the mechanism analysis difficulties in the application of ultrasonic guided wave. But the integrity of structural state data sets has become a new pain point in engineering applications under this new approach, and how to apply the knowledge obtained from the existing data set to different but related fields through knowledge transfer in such cases begin to attract the attention of scholars and engineers. Although several systematic and valuable review articles on data-driven ultrasonic guided wave monitoring methods have been published, they only summarized relevant studies from the perspective of data-driven algorithms, ignoring the knowledge transfer process in practical application scenarios, and the intelligent ultrasonic guided wave monitoring methods based on knowledge transfer of incomplete sets are still lacking a comprehensive review. This paper focuses on the ultrasonic guided wave transfer monitoring technology when the training sample is missing, explores the feature correlation between samples in different domains, improves the transfer ability of the structural monitoring model under different conditions, and analyzes the ultrasonic guided wave intelligent monitoring methods for structural state under different sample missing conditions from three aspects: semi-supervised monitoring, multi-task transfer and cross-structure transfer. It is also expected to provide a new method and approach to solve the condition monitoring problems in other complex scenarios.

Innovative Strategies for Exploring the Potential of China’s Market

As the world's second-largest economy, China plays an important role in the development of world trade. With nearly one-fifth of the world's population, China's market has huge trade potential for each country. In the process of expanding exports and investment to China, many countries face a lot of trade barriers due to China's large market span, complex structure, and different consumer preferences, so exploring China’s market has become a difficult research topic for many economists and entrepreneurs. It is also a complicated strategy for different companies choose different cities and regions when entering China’s market or expanding their market share in China. The research subject of this paper is exploring the potential of China’s market, and the research object is three dimensions (geographical, economic, and import potential) analysis of market entry approaches in China. According to the depth, breadth and market potential of China’s market, this paper makes a comprehensive comparison of different methods to enter China’s market. The research results will help a company improve the efficiency of penetrating China’s market. In the conclusion part, this paper summarizes new scientific methods and steps for international companies to explore the potential of China’s market.

Kinetic principles of chemical cross-link formation for protein–protein interactions

Proteins play a central role in most biological processes within the cell, and deciphering how they interact is key to understand their function. Cross-linking coupled with mass spectrometry is an essential tool for elucidating protein-protein interactions (PPIs). Despite its importance, we still know surprisingly little about the principles that underlie the process of chemical cross-link formation itself and how it is influenced by different physicochemical factors. To understand the molecular details of cross-link formation, we have set up a comprehensive kinetic model and carried out simulations of protein cross-linking on large protein complexes. We dissect the contribution on the cross-link yield of parameters such as amino acid reactivity, cross-linker concentration, and hydrolysis rate. Our model can compute cross-link formation based solely on the structure of a protein complex, thereby enabling realistic predictions for a diverse set of systems. We quantitatively show how cross-links and mono-links are in direct competition and how the hydrolysis rate and abundance of cross-linker and proteins directly influence their relative formation. We show how cross-links and mono-links exist in an "all-against-all" competition due to their simultaneous formation, resulting in a nonintuitive network of interdependence. We show that this interdependence is locally confined and mainly limited to direct neighbors or residues in direct vicinity. These results enable us to identify the optimal cross-linker concentration at which the maximal number of cross-links is formed. Taken together, our study establishes a comprehensive kinetic model to quantitatively describe cross-link formation for PPIs.

Characterising the molecular line emission in the asymmetric Oph-IRS 48 dust trap: Temperatures, timescales, and sub-thermal excitation

The ongoing physical and chemical processes in planet-forming disks set the stage for planet (and comet) formation. The asymmetric disk around the young star Oph-IRS 48 has one of the most well-characterised chemical inventories, showing molecular emission from a wide variety of species at the dust trap: from simple molecules, such as CO SO SO_2 and H_2CO to large complex organics, such as CH_3OH CH_3OCHO and CH_3OCH_3 . One of the explanations for the asymmetric structure in the disk is dust trapping by a perturbation-induced vortex. We aimed to constrain the excitation properties of the molecular species SO_2 CH_3OH and H_2CO for which we have used 13, 22, and 7 transitions of each species, respectively. We further characterised the extent of the molecular emission, which differs among molecules, through the determination of important physical and chemical timescales at the location of the dust trap. We also investigated whether the anticyclonic motion of the potential vortex influences the observable temperature structure of the gas. Through a pixel-by-pixel rotational diagram analysis, we created maps of the rotational temperatures and column densities of SO_2 and CH_3OH . To determine the temperature structure of H_2CO we have used line ratios of the various transitions in combination with non-local thermal equilibrium (LTE) RADEX calculations. The timescales for freeze-out, desorption, photodissociation, and turbulent mixing at the location of the dust trap were determined using an existing thermochemical model. Our rotational diagram analysis yields temperatures of $T$=54.8pm 1.4 K ( SO_2 ) and $T$=125.5$^ $ K ( CH_3OH ) at the emission peak positions of the respective lines. As the SO_2 rotational diagram is well characterised and points towards thermalised emission, the emission must originate from a layer close to the midplane where the gas densities are high enough. The rotational diagram of CH_3OH is, in contrast, dominated by scatter and subsequent non-LTE RADEX calculations suggest that both CH_3OH and H_2CO must be sub-thermally excited higher up in the disk ($z/r For H_2CO the derived line ratios suggest temperatures in the range of $T K. The SO_2 temperature map hints at a potential radial temperature gradient, whereas that of CH_3OH is nearly uniform and that of H_2CO peaks in the central regions. We, however, do not find any hints of the vortex influencing the temperature structure across the dust trap. The longer turbulent mixing timescale, compared to that of photodissociation, does provide an explanation for the expected vertical emitting heights of the observed molecules. On the other hand, the short photodissociation timescales are able to explain the wider azimuthal molecular extent of SO_2 compared to CH_3OH . The short timescales are, however, not able to explain the wider azimtuhal extent of the H_2CO emission. Instead, it can be explained by a secondary reservoir that is produced through the gas-phase formation routes, which are sustained by the photodissociation products of, for example CH_3OH and H_2O Based on our derived temperatures, we expect SO_2 to originate from deep inside the disk, whereas CO comes from a higher layer and both CH_3OH and H_2CO emit from the highest emitting layer. The sub-thermal excitation of CH_3OH and H_2CO suggests that our derived (rotational) temperatures underestimate the kinetic temperature. Given the non-thermal excitation of important species, such as H_2CO and CH_3OH it is important to use non-LTE approaches when characterising low-mass disks, such as that of IRS 48. Furthermore, for the H_2CO emission to be optically thick, as expected from an earlier derived isotopic ratio, we suggest that the emission must originate from a small radial `sliver' with a width of sim 10 au, located at the inner edge of the dust trap.

A Study of New Approaches to Statistical Analysis of Research data

Statistical analysis has been at the heart of scientific research, providing critical tools for data interpretation, decision-making, and hypothesis testing. Some of the ancient techniques used were hypothesis testing, regression analysis, and time series analysis, among others. Such methods have proved to be good tools for researchers dealing with smaller and more structured datasets. However, the large size and complexity of the dataset exposed the weaknesses in such classical approaches, especially in handling large, unstructured, or non-linear datasets. Increased computing power and the development of machine learning algorithms have increased flexibility, nudging the statistical approach towards more flexible, data-driven methods. This paper reviews emerging approaches such as machine learning, deep learning, Bayesian methods, and network analysis and places an emphasis on how these approaches can be applied over a range of fields where they could vastly transform statistical analysis. Based on this review, comparing traditional and modern methodologies, it thereby demonstrates how innovations better complement rather than replacing the capabilities of statistical analysis, thus shaping the future of research in this changing environment.

The Influence of Extra-Ribosomal Functions of Eukaryotic Ribosomal Proteins on Viral Infection

The eukaryotic ribosome is a large ribonucleoprotein complex consisting of four types of ribosomal RNA (rRNA) and approximately 80 ribosomal proteins (RPs), forming the 40S and 60S subunits. In all living cells, its primary function is to produce proteins by converting messenger RNA (mRNA) into polypeptides. In addition to their canonical role in protein synthesis, RPs are crucial in controlling vital cellular processes such as cell cycle progression, cellular proliferation, differentiation, DNA damage repair, genome structure maintenance, and the cellular stress response. Viruses, as obligate intracellular parasites, depend completely on the machinery of the host cell for their replication and survival. During viral infection, RPs have been demonstrated to perform a variety of extra-ribosomal activities, which are especially important in viral disease processes. These functions cover a wide range of activities, ranging from controlling inflammatory responses and antiviral immunity to promoting viral replication and increasing viral pathogenicity. Deciphering the regulatory mechanisms used by RPs in response to viral infections has greatly expanded our understanding of their functions outside of the ribosome. Furthermore, these findings highlight the promising role of RPs as targets for the advancement of antiviral therapies and the development of novel antiviral approaches. This review comprehensively examines the many functions of RPs outside of the ribosome during viral infections and provides a foundation for future research on the host–virus interaction.

Vigabatrin-associated brain magnetic resonance imaging abnormalities and clinical symptoms in infants with tuberous sclerosis complex.

Previous retrospective studies have reported vigabatrin-associated brain abnormalities on magnetic resonance imaging (VABAM), although clinical impact is unknown. We evaluated the association between vigabatrin and predefined brain magnetic resonance imaging (MRI) changes in a large homogenous tuberous sclerosis complex (TSC) cohort and assessed to what extent VABAM-related symptoms were reported in TSC infants. The Dutch TSC Registry and the EPISTOP cohort provided retrospective and prospective data from 80 TSC patients treated with vigabatrin (VGB) before the age of 2 years and 23 TSC patients without VGB. Twenty-nine age-matched non-TSC epilepsy patients not receiving VGB were included as controls. VABAM, specified as T2/fluid-attenuated inversion recovery hyperintensity or diffusion restriction in predefined brain areas, were examined on brain MRI before, during, and after VGB, and once in the controls (at approximately age 2 years). Additionally, the presence of VABAM accompanying symptoms was evaluated. Prevalence of VABAM in VGB-treated TSC patients was 35.5%. VABAM-like abnormalities were observed in 13.5% of all patients without VGB. VGB was significantly associated with VABAM (risk ratio [RR] = 3.57, 95% confidence interval [CI] = 1.43-6.39), whereas TSC and refractory epilepsy were not. In all 13 VGB-treated patients with VABAM for whom posttreatment MRIs were available, VABAM entirely resolved after VGB discontinuation. The prevalence of symptoms was 11.7% in patients with VABAM or VABAM-like MRI abnormalities and 4.3% in those without, implicating no significant association (RR = 2.76, 95% CI = .68-8.77). VABAM are common in VGB-treated TSC infants; however, VABAM-like abnormalities also occurred in children without either VGB or TSC. The cause of these MRI changes is unknown. Possible contributing factors are abnormal myelination, underlying etiology, recurrent seizures, and other antiseizure medication. Furthermore, the presence of VABAM (or VABAM-like abnormalities) did not appear to be associated with clinical symptoms. This study confirms that the well-known antiseizure effects of VGB outweigh the risk of VABAM and related symptoms.

Integrins in the kidney - beyond the matrix.

The development and proper functioning of the kidney is dependent on the interaction of kidney cells with the surrounding extracellular matrix (ECM). These interactions are mediated by heterodimeric membrane-bound receptors called integrins, which bind to the ECM via their extracellular domain and via their cytoplasmic tail to intracellular adaptor proteins, to assemble large macromolecular adhesion complexes. These interactions enable integrins to control cellular functions such as intracellular signalling and organization of the actin cytoskeleton and are therefore crucial to organ function. The different nephron segments and the collecting duct system have unique morphologies, functions and ECM environments and are thus equipped with unique sets of integrins with distinct specificities for the ECM with which they interact. These cell-type-specific functions are facilitated by specific intracellular integrin binding proteins, which are critical in determining the integrin activation status, ligand-binding affinity and the type of ECM signals that are relayed to the intracellular structures. The spatiotemporal expression of integrins and their specific interactions with binding partners underlie the proper development, function and repair processes of the kidney. This Review summarizes our current understanding of how integrins, their binding partners and the actin cytoskeleton regulate kidney development, physiology and pathology.

Sea ice perturbations in aquaplanet simulations: isolating the physical climate responses from model interventions

Abstract Comprehensive climate model simulations with perturbed sea ice covers have been extensively used to assess the impact of future sea ice loss, suggesting substantial climate changes both in the high latitudes and beyond. However, previous work using an idealized energy balance model calls into question the methods that are used to perturb sea ice cover, demonstrating a consistent overestimate of the surface warming due to sea ice loss, while the large complexity gap between the idealized and comprehensive models makes the implications of this result unclear. To bridge this gap we have performed simulations with a new implementation of the CESM2 model in a slab-ocean aquaplanet configuration coupled with thermodynamic sea ice, which is able to capture the realistic seasonal characteristics of polar climate change. Using this model setup, we perform a suite of experiments to systematically quantify the spurious climate responses associated with melting sea ice without a CO2 forcing. We find that using the sea ice ghost flux method overestimates many aspects of the climate response by ~10-20%, including the polar warming, the mini global warming signal and the increase in both precipitation and evaporation. The location of the latitudinal band of heating applied to melt the sea ice relative to the midlatitude jet is important for determining where the midlatitude circulation response is overestimated. This work advances our ability to isolate the true climate response to sea ice loss, and provides a framework for conducting coupled sea ice loss simulations absent the spurious impacts from the addition of artificial heating.

A quantitative, label-free visual interference colour assay platform for protein targeting and binding assays

The vast array of immunoassay technologies used to assess protein interactions is costly or platform-specific. We present a label-free visual interference colour assay (VICA) that quantifies peptide and protein interactions by creating an iridescent surface allowing direct visualisation without spectrophotometric optics or microfluidics. A nanoporous aluminium oxide surface is tuned to match the refractive indices of the overlying protein layers to generate visual interference colours. To functionalise the surface, we created an affinity-capture system using a protein A-carboxyglutamic (GLA) construct that orients antibodies to enhance the signal. Using off-the-shelf antibodies, the platform can isolate analytes in buffer, whole blood, or serum. This surface generates a discernible colour change at concentrations as low as 50 femtomoles/mm2 and can monitor oligomer formation in sequential steps on the same slide. VICA provides comparable kinetic parameters to biolayer interferometry and traditional immunoassays while also allowing characterisation of proteins in large macromolecular complexes.

Local Wave Function Embedding: Correlation Regions in PNO-LCCSD(T)-F12 Calculations.

Many chemical reactions affect only a rather small number of bonds, leaving the largest part of the chemical and geometrical structure of the molecules nearly unchanged. In this work we extended the previously proposed region method [J. Chem. Phys. 128, 144106 (2008)] to PNO-LCCSD(T)-F12. Using this method, we investigate whether accurate reaction energies for larger systems can be obtained by correlating only the electrons in a region of localized molecular orbitals close to the reaction center at high-level (PNO-LCCSD(T)-F12). The remainder is either treated at lower level (PNO-LMP2-F12) or left uncorrelated (Hartree-Fock frozen core). It is demonstrated that indeed the computed reaction energies converge rather quickly with the size of the correlation regions toward the results of the full calculations. Typically, 2-3 bonds from the reacting atoms need to be included to reproduce the results of the full calculations to within ±0.2 kcal/mol. We also computed spin-state energy differences in a large transition metal complex, where a factor of 15 in computation time could be saved, still yielding a result that is within ±0.1 kcal/mol of the one obtained in a full PNO-LCCSD(T)-F12 calculation.

Molecular basis of mucopolysaccharidosis type II (Hunter syndrome): first review and classification of published IDS gene variants

PurposeMucopolysaccharidosis type II (MPS II) is a rare X-linked lysosomal storage disorder caused by genetic alterations in the iduronate 2-sulfatase (IDS) gene. A wide range of variants has been reported for different countries and ethnic groups. We collected, analyzed and uniformly summarized all published IDS gene variants reported in literature up to June 2023, here providing the first worldwide review and classification.MethodsData was obtained from a literature search, conducted in PubMed and Google. All data was analyzed to define the most common alleles, geographic distribution and genotype-phenotype correlation. Moreover, point variants were classified according to their pathogenicity, based on the ACMG guidelines.ResultsSeveral types of variants have been described in the IDS gene, including intrachromosomal homologous recombination occurring between the homologous regions of IDS gene and its pseudogene IDSP1. Overall, we collected 2852 individuals from 2798 families, including 24 female patients. Most families carried missense variants, followed by large deletions-insertions and complex rearrangements, small frameshift deletions/insertions and nonsense variants. Based on ACMG guidelines, 62.9% of the 779 point variants were classified as “pathogenic”, 35.4% as “likely pathogenic”, and the remaining 13 variants as having “uncertain significance”.ConclusionData from this study confirmed that MPS II is a genetically very heterogeneous disorder, making genotype-phenotype correlation very challenging and in most cases merely unfeasible. Mutation updates are essential for the correct molecular diagnosis, genetic counseling, prenatal and preimplantation diagnosis, and disease management.

SVTR-SRNet: A Deep Learning Model for Scene Text Recognition via SVTR Framework and Spatial Reduction Mechanism

Most deep learning models suffer from the problems of large computational complexity and insufficient feature extraction. To achieve a dynamic balance and tradeoff between computational complexity and performance, an enhanced SVTR-based scene text recognition model (SVTR-SRNet) was designed in this paper. In the SVTR-SRNet, we first created a bottom-up jump connection network that increases the number of information transfer pathways between the top and bottom features and improves the accuracy of information extraction. Second, we modified the attention mechanism by adding a new intermediate parameter called SR(Q) (Spatial Reduction (Q)), which finds a suitable compromise between the representational power and computing efficiency. In contrast to the conventional attention mechanism, the novel technique maintains the ability to model the global context while also enhancing efficiency. Ultimately, we developed a novel adaptive hybrid loss function to mitigate the shortcomings of a singular loss function’s inadequate generalization capacity and enhance the model’s resilience in handling a variety of challenging scenarios. Our technique outperforms existing standard models in terms of recognition performance on both the English and Chinese datasets, which deal with a high number of similar characters. As the model possesses great efficiency and outstanding cross-linguistic adaptability, it has a wide range of practical applications.

Multi-scale perceptual modulation network for low-dose computed tomography denoising.

Low-dose computed tomography (LDCT) reduces radiation exposure, but the introduced noise and artifacts impair its diagnostic accuracy. Convolutional neural networks (CNNs) are widely used for LDCT denoising, but they suffer from a limited receptive field. The use of a larger kernel size can enlarge the receptive field and boost model performance; however, the computational cost of the model greatly increases. We aimed to develop a LDCT denoising CNN with a large receptive field and lower computational complexity. We developed a multi-scale perceptual modulation network (MSPMnet) incorporating a powerful multi-head decomposable convolution (MHDC). To address the high computational complexity of large kernel convolutions, we developed a novel MHDC module that can capture multi-scale features and efficiently expand the receptive field. The MHDC module couples maximum-pooling with three depth-wise convolutions of varying kernel sizes via a channel splitting mechanism, where, unlike conventional CNNs, the two large two-dimensional kernels are each decomposed into a set of cascaded orthogonal one-dimensional kernels to remain lightweight. Further, departing from prior methodologies that apply a uniform kernel size throughout the network, we introduced a receptive field-ramp mechanism that adeptly transitions from local to relatively long-range dependency modeling as the network depth increases, thereby achieving superior performance. The proposed MSPMnet was evaluated on a Mayo Clinic data set with a conventional iterative algorithm, two CNN models, and two Transformer models used for comparison. Compared to the competing baseline methods, the MSPMnet exhibited better performance in both the visual and quantitative assessments. Visually, the MSPMnet preserved the structure, edges, and textures with excellent noise and artifact reduction, generating the denoised images closest to normal-dose computed tomography images. Quantitatively, the MSPMnet had the lowest root mean-square error (RMSE) (8.3094±1.9325) and the highest peak signal-to-noise ratio (PSNR) (33.8525±1.8213 dB), structural similarity index (SSIM) (0.9309±0.0272), and feature similarity index (FSIM) (0.9699±0.0113), demonstrating superior denoising performance. The proposed MSPMnet excelled at LDCT denoising, effectively removing noise and artifacts while preserving edges. Compared to the state-of-the-art CNNs and Transformers, the proposed MSPMnet exhibited superior denoising performance both quantitatively and qualitatively.

INDUSTRIAL SNOW AND SELECTED ECONOMIC IMPACTS ON THE OPERATION OF KOSICE AIRPORT

The aim of the paper is to describe industrial snow as a meteorological phenomenon of anthropological origin and the physical principle of its formation in the context of Kosice Airport, with international traffic. The research also focused on selected economic impacts on the operation of Kosice Airport, which is located next to the large industrial complex U.S. Steel Kosice. There are only a few international airports in the world that are affected by industrial snow, which is why we also focus on the economic impact of maintaining the operational areas of the airport to the required extent according to standards. A solution to the problem is also proposed based on the scientific works of experts who focused their scientific activities on the issue of industrial snow production. In addition to the differences in the formation of industrial and natural snow, the article mentions the differences in the structure and properties of crystalline elements and their distribution according to the external meteorological conditions under which they are formed. The paper identifies meteorological situations that are favorable for the formation of industrial snow and due to the mutual location of the airport and the industrial complex, the airport is affected by industrial snow. This article describes the selected economic impacts of industrial snowfall on air traffic at Kosice Airport from 2017–2022. These were expressed after providing data from the airfield maintenance department of the Kosice Airport. At the same time, airport maintenance workers shared their unique experiences with this special meteorological phenomenon. The outcome of this research brings the unique method of industrial snow day occurrence and helps to explain this rare weather phenomenon and its impact on transport and social aspects.

Thoracic Aortic Calcification and Complex Large Aortic Arch Plaque in Patients with Chronic Coronary Syndrome.

Aortic arch plaque (AAP) morphology and complexity can serve as markers of atherosclerotic cardiovascular disease. This study investigated 1) the diagnostic value of the thoracic aortic calcification (TAC) score for detecting AAP and large complex AAP and 2) the prognostic significance of TAC in patients with chronic coronary syndrome (CCS). The predictors of AAPs with large (≥ 4 mm in thickness) and complex (ulcerated or protruded) morphologies were evaluated in 412 symptomatic patients with CCS who underwent coronary computed tomography angiography and simultaneous assessment of the aortic arch. Receiver operating curve analysis was performed to determine whether the TAC score can improve the diagnostic value of complex large AAP compared with clinical parameters. Multivariate logistic regression analyses revealed that the TAC score was an independent predictor of complex large AAP (odds ratio, 1.46; P = 0.0074). The area under the curve (AUC) for the base model comprising clinical parameters (age ≥ 70 years + current smoking status + systolic blood pressure) was 0.707 (P = 0.0177). The addition of a TAC score of ≥ 978 Agatston units (AUC, 0.744) significantly improved the AUC (AUC, 0.751, P < 0.001) for patients aged ≥ 70 years (AUC, 0.638). The multivariable Cox hazard model revealed that TAC of ≥ 432.4 (hazard ratio [HR], 4.312; P = 0.01) and large complex AAP are independently associated with the incidence of major adverse cardiovascular events (HR, 3.108; P = 0.011). This study demonstrated that TAC and AAP morphology serve as markers of major adverse cardiovascular events. The combination of advanced age, TAC score, and obstructive CAD exhibited robust diagnostic accuracy in detecting complex large AAP.

Stabilization Mechanism of Initiator Transfer RNA in the Small Ribosomal Subunit from Coarse-Grained Molecular Simulations.

Proteins play a variety of roles in biological phenomena in cells. Proteins are synthesized by the ribosome, which is a large molecular complex composed of proteins and nucleic acids. Among the many molecules involved in the process of protein synthesis, tRNA is one of the essential molecules. In this study, coarse-grained molecular dynamics simulations were performed to understand how the tRNA molecule is stabilized in the ribosome, and the free energy along the dissociation path of the tRNA was calculated. We found that some ribosomal proteins, which are components of the ribosome, are involved in the stabilization of the tRNA. The positively charged amino acid residues in the C-terminal region of the ribosomal proteins are particularly important for stabilization. These findings contribute to our understanding of the molecular evolution of protein synthesis in terms of the ribosome, which is a universal component of life.

CASCADE BOILER PLANTS AS A LIFE SUPPORT SYSTEM FOR BUILDINGS AND STRUCTURES

Problem statement. Destruction and damage to the energy infrastructure as a result of military operations lead to the shutdown of life support systems, which affects the lives of the population and the economy. The study of heating systems in modern realities is gaining great relevance and requires scientific substantiation, new strategic solutions, as well as organizational, economic and technological developments to overcome new challenges. The study points to the need to find energy-efficient and easily renewable alternative heating systems. Purpose of the study. The research objective is to identify alternative, individual building life support systems that are more energy-efficient, modern, and quickly recoverable. This includes systems that can be restored and used independently of other power sources. Methods. Analytical studies of the heat supply system load, technological study of the heat supply system of buildings and structures, analysis and synthesis of available data, comparison of the efficiency of cascade boiler houses with other heating systems. Research results. Cascade boiler systems have demonstrated significant potential for fuel savings, reduced emissions, and optimized operational costs. These systems have proven to be versatile, finding applications in various sectors, from residential buildings to large industrial complexes. Notably, cascade boiler systems have shown to be highly resilient in the face of modern challenges, such as natural disasters, economic crises, and conflicts, ensuring a reliable and efficient heat supply. Scientific novelty. The study explores the implementation of cascade boiler systems as an alternative heating solution for buildings in Ukrainian cities affected by the conflict, highlighting their advantages and potential to address challenges associated with the destruction of traditional energy infrastructure. Practical significance. The search for innovative heating solutions for buildings is highly relevant, especially as we approach the completion of construction projects and prepare for the upcoming heating season.

RCT-YOLOv8: A Tuna Detection Model for Distant-Water Fisheries Based on Improved YOLOv8

With the development of distant-water fisheries, ship fishing and fish catch detection are now vital to modern fishing. Existing manual detection methods are prone to issues such as missed detections and false detections. Deep learning has enabled the deployment of detection models on shipboard devices, offering a new solution. However, many existing models are hindered by large parameters and computational complexity, making them unsuitable for shipboard use due to limited resources and costs onboard ships. To address these challenges, we propose the RCT-YOLOv8 model for tuna catch detection in this paper. Specifically, we adopt YOLOv8 as the base model and replace the network backbone with RepVGG network, which employs re-parameterized convolutions to enhance detection accuracy. Additionally, we incorporate coordinate attention at the end of the backbone to better aggregate channel-wise information. In the neck part, we introduce the contextual transformer (CoT) attention and propose the C2F-CoT model, which combines convolutional neural network with Transformer to capture global features, thereby improving detection accuracy and the effectiveness of feature propagation. We test multiple loss functions and select efficient intersection over union, which is more suitable for our algorithm. Furthermore, to adapt to devices with limited computational resources, we utilize the dependency-graph-based pruning method to compress the network model. Compared to the base network, the pruned model achieves a 9.8% increase in detection accuracy while reducing parameters and computational complexity by 40% and 35.8%, respectively. Compared to various algorithms, the pruned model demonstrates the highest detection accuracy, lowest parameter count, and lowest computational complexity, achieving optimal results at all fronts.

Advancing Non-Atom-Centered Basis Methods for More Accurate Interaction Energies: Benchmarks and Large-Scale Applications.

Recent advances in local electron correlation approaches have enabled the relatively routine access to CCSD(T) [that is, coupled cluster (CC) with single, double, and perturbative triple excitations] computations for molecules of a hundred or more atoms. Here, approaching their complete basis set (CBS) limit becomes more challenging due to extensive basis set superposition errors, often necessitating the use of large atomic orbital (AO) basis sets with diffuse functions. Here, we study a potential remedy in the form of non-atom-centered or floating orbitals (FOs). FOs are still rarely employed even for small molecules due to the practical complication of defining their position, number, exponents, etc. The most frequently used FO method thus simply places a single FO center with a large number of FOs toward the middle of noncovalent dimers; however, a single FO center for larger complexes can soon become insufficient. A recent alternative uses a grid of FO centers around the monomers with a single s function per center, which is currently applicable only for H, C, N, and O atoms. Here, we build on the above advantages and mitigate some drawbacks of previous FO approaches by using a layer of FO centers and 4-9 FOs/center for each monomer. Thus, a double layer of FOs is placed between the interacting subsystems. When extending the double-ζ AO basis with this double layer of FOs, the quality of conventional augmented double-ζ or conventional triple-ζ AO bases can be reached or surpassed with less orbitals, leading to few tenths of a kcal/mol basis set errors for medium-sized dimers. This good performance extends to larger molecules (shown here up to 72 atoms), as efficient local natural orbital (LNO) CCSD(T) computations with only double-ζ AO and 4 FOs/center FO bases match our LNO-CCSD(T)/CBS reference within ca. 0.1 kcal/mol. These developments introduce FO methods to the accurate modeling of large molecular complexes without limitations to atom types by further accelerating efficient correlation calculations, like LNO-CCSD(T).