Model Principle Research Articles

Enhanced support vector machine-based moving regression strategy for response prediction and reliability estimation of complex structure

For predicting response property and estimating reliability level of complex structure, enhanced support vector machine-based moving regression (MR-ESVM) strategy is proposed based on support vector machine (SVM), heuristic algorithm and moving least square (MLS) technique. Under this strategy, we develop four different SVM models including SVM-based moving regression (MR-SVM), SVM-based improved moving regression (IMR-SVM), improved SVM-based moving regression (MR-ISVM) and bi-optimized SVM-based moving regression (BiOMR-SVM) methods. In these developed MR-ESVM approaches, the MR-SVM method is explored by introducing the MLS technique into the SVM model; the IMR-SVM method is discussed by fusing the MR-SVM method and artificial rabbits optimization (ARO), and the ARO is used to search the optimal radius of compact region; the MR-ISVM method is raised by integrating the ARO into the MR-SVM, and the ARO is applied to find the optimal values in the SVM model; The BiOMR-SVM method is emerged by merging the IMR-SVM and MR-ISVM methods. To verify the effectiveness of these developed MR-ESVM strategies, a multivariate nonlinear function approximation is implemented to illustrate the advantages from the mathematics perspective, an aeroengine turbine blisk radial deformation reliability analysis and an aircraft hydraulic system low pressure reliability analysis are derived to demonstrate the applicability in engineering practice. The analytical results show that these four MR-ESVM approaches hold excellent merits in modeling features and simulation characteristics. The efforts of this work provide a novel idea for the response prediction of complex structure, and enrich the reliability estimation principle of surrogate models of complex structure.

Application of Stable Diffusion and LoRA Models in AI Drawing

Abstract. This paper explores the application of Stable Diffusion model and LoRA (Low-Rank Adaptation) model in AI-generated artwork. The authors introduce the foundational principles of Stable Diffusion model and LoRA, as well as their application in high-quality image generation. Using three popular datasets ImageNet, COCO, and CelebA we apply various image quality assessment metrics, including PSNR (Peak Signal-to-Noise Ratio), IS (Inception Score), and FID (Frchet Inception Distance), and further validate their potential in artistic creation through subjective evaluations. By comparing the performance of these two models across different datasets, we examine their strengths, weaknesses, areas for improvement in image generation tasks, along with user experience considerations. The experimental results show that the Stable Diffusion model excels in terms of image quality and diversity, while the LoRA model offers significant benefits in computational efficiency and resource usage. Through comprehensive experimental evaluations, this paper provides a scientific basis for model selection in AI art creation and offers insights for the future development of hybrid models.

Applications of Denoising Diffusion Probability Model in Image Processing

Abstract. In 2020, Jonathan Ho et al. from the University of Berkeley proposed the Denoising Diffusion Probabilistic Models (DDPM), which improved on the shortcomings of the previous-generation Deformable Part Models (DPM). At the same time, they also surpassed previous generative models in image synthesis effects by using noise prediction, such as Generative Adversarial Networks (GAN), Variational Auto-Encoders (VAE), flow-based models and energy-based models, etc. After that, the denoising diffusion probabilistic models gradually received more discussions and research. In 2022, OpenAI launched the DALL-E 2, realizing Text-to-Image generation combined with the denoising diffusion probabilistic model. Two years later, with the release of Sora, the first Text-to-Video large models based on DALL-E, the denoising diffusion probabilistic models have received unprecedented attention in the multi-modal field. This article first provides the background and development process of the denoising diffusion probabilistic models. Secondly, it introduces the algorithmic principles of the diffusion models. Following that, it lists the applications of the models in the area of images. In the last section, it expounds on the advantages, disadvantages and future trends of the denoising diffusion probabilistic models.

Strategy to measure tau g−2 via photon fusion in LHC proton collisions

Measuring the tau-lepton (τ) anomalous magnetic moment aτ=(gτ−2)/2 in photon fusion production (γγ→ττ) tests foundational Standard Model principles. However, γγ→ττ eludes observation in LHC proton collisions (pp) despite enhanced new physics sensitivity from higher-mass reach than existing probes. We propose a strategy to measure pp→p(γγ→ττ)p by introducing the overlooked electron-muon signature with vertex isolation for signal extraction. Applying the effective field theory of dipole moments, we estimate 95% confidence level sensitivity of −0.0092<aτ<0.011 assuming 300 fb−1 luminosity and 5% systematics. This fourfold improvement beyond existing constraints opens a crucial path to unveiling new physics imprinted in tau-lepton dipoles. Published by the American Physical Society 2024

Construction and SHAP interpretability analysis of a risk prediction model for feeding intolerance in preterm newborns based on machine learning

ObjectiveTo construct a highly accurate and interpretable feeding intolerance (FI) risk prediction model for preterm newborns based on machine learning (ML) to assist medical staff in clinical diagnosis.MethodsIn this study, a sample of 350 hospitalized preterm newborns were retrospectively analysed. First, dual feature selection was conducted to identify important feature variables for model construction. Second, ML models were constructed based on the logistic regression (LR), decision tree (DT), support vector machine (SVM) and eXtreme Gradient Boosting (XGBoost) algorithms, after which random sampling and tenfold cross-validation were separately used to evaluate and compare these models and identify the optimal model. Finally, we apply the SHapley Additive exPlanation (SHAP) interpretable framework to analyse the decision-making principles of the optimal model and expound upon the important factors affecting FI in preterm newborns and their modes of action.ResultsThe accuracy of XGBoost was 87.62%, and the area under the curve (AUC) was 92.2%. After the application of tenfold cross-validation, the accuracy was 83.43%, and the AUC was 89.45%, which was significantly better than those of the other models. Analysis of the XGBoost model with the SHAP interpretable framework showed that a history of resuscitation, use of probiotics, milk opening time, interval between two stools and gestational age were the main factors affecting the occurrence of FI in preterm newborns, yielding importance scores of 0.632, 0.407, 0.313, 0.313, and 0.258, respectively. A history of resuscitation, first milk opening time ≥ 24 h and interval between stools ≥ 3 days were risk factors for FI, while the use of probiotics and gestational age ≥ 34 weeks were protective factors against FI in preterm newborns.ConclusionsIn practice, we should improve perinatal care and obstetrics with the aim of reducing the occurrence of hypoxia and preterm delivery. When feeding, early milk opening, the use of probiotics, the stimulation of defecation and other measures should be implemented with the aim of reducing the occurrence of FI.

Molecular profiles, sources and lineage restrictions of stem cells in an annelid regeneration model

Regeneration of missing body parts can be observed in diverse animal phyla, but it remains unclear to which extent these capacities rely on shared or divergent principles. Research into this question requires detailed knowledge about the involved molecular and cellular principles in suitable reference models. By combining single-cell RNA sequencing and mosaic transgenesis in the marine annelid Platynereis dumerilii, we map cellular profiles and lineage restrictions during posterior regeneration. Our data reveal cell-type specific injury responses, re-expression of positional identity factors, and the re-emergence of stem cell signatures in multiple cell populations. Epidermis and mesodermal coelomic tissue produce distinct putative posterior stem cells (PSCs) in the emerging blastema. A novel mosaic transgenesis strategy reveals both developmental compartments and lineage restrictions during regenerative growth. Our work supports the notion that posterior regeneration involves dedifferentiation, and reveals molecular and mechanistic parallels between annelid and vertebrate regeneration.

CNN-Based Cancer Image Diagnosis: Current Progress and Future Directions

Abstract. With the development of deep learning technology, CNN (Convolutional Neural Network) models have shown great value in medical image analysis, especially in diagnosing early lung cancer, breast cancer, and brain tumors. In this study, we recall and organize the application and progress of CNN models in the field of cancer and tumor diagnosis in the past five years to provide a theoretical basis and reference for related researchers. This article introduces the principles of different CNN cancer diagnostic models and compares and analyzes their results, and ultimately finds that these models have significant advantages in improving the accuracy and efficiency of cancer diagnosis, but at the same time, there are also problems such as too much reliance on large datasets, high model complexity, and poor generalization ability. In the future, we can consider optimizing the performance of CNN network models, enhancing the generalization ability of the models, and developing data enhancement techniques on this basis, so that CNN models can be better applied in the field of cancer diagnosis.

From practice to theory: characterizing the gap in surgical simulation

PurposeSimulation curricula continue to struggle with adequately preparing trainees for the operating room. One reason for this phenomenon may be the lack of application and enactment of learning and instructional theories into simulation curricular design and practice. Few educators have taken a reflective approach to understand how surgical simulation succeeds and fails to incorporate best practices for learning based on theory. As such, this study aims to examine simulation sessions to identify gaps in practice by nesting two key frameworks from general education into surgical simulation: learning integration and the cognitive apprenticeship model.MethodsWe conducted an observational qualitative study in which we recorded simulation sessions with fifteen trainees and surgeons and deductively applied components of the above frameworks to transcripts. Subsequently, we analyzed gaps in the transcripts with regard to the application of these frameworks as theoretical concepts informing the analysis and interpretation.ResultsWe organized results around the four fundamental tenets of learning integration, with principles of the cognitive apprenticeship model explored to provide further units of analysis. In doing so, we identified that simulation instructors adequately modeled, coached, and scaffolded to enable early phases of learning integration. However, instructors less aptly enabled reflection and self-guided exploration, which are critical components of learning integration.ConclusionsWe found areas in which instruction diverged from ideal standards as informed by our theoretical frameworks, thus highlighting the importance of regular simulation review to ensure that well-designed and intentioned simulation curricula continue to reflect the best educational principles when enacted in practice.

Application of deep learning algorithms for identifying deterioration in the ushnisha (Head Bun) of the Leshan Giant Buddha

The Leshan Giant Buddha’s ushnisha (Head Bun) has suffered from loss of lime plaster, cracks, and biological damage, compromising its structural integrity and reducing the effectiveness of the drainage system in the Buddha's head. The infiltration of moisture has led to water damage within the statue, significantly accelerating its weathering. This situation urgently requires protection and reinforcement measures. Detecting deterioration in the ushnisha is a crucial step in the preservation process. In this study, we utilized two deep learning models for pixel-level semantic segmentation of the damage. Due to the small size of the cracks, a weighted loss function was applied to improve both the training speed of the model and the efficiency of crack identification. This weighting strategy proved effective for both models. The weighted K-Net model achieved a mean accuracy (mAcc) of 90.23% and a mean intersection-over-union (mIoU) of 69.55%, with a damage segmentation speed of 7 images per second, which is 1309 times faster than manual segmentation. By applying the trained deep learning models to re-examine the ushnisha, we successfully identified damage that had been overlooked during manual annotation. Using the model’s enhanced results, we conducted a comprehensive quantification of the damage across all ushnisha and identified the most severely affected areas. Additionally, we performed a model interpretability analysis to explain the decision-making process and principles of the deep learning models. This research provides significant practical value for detecting and quantifying damage in the Leshan Giant Buddha.

Nowhere else in the world? The Korean Safe Rates System in global context

Abstract The Korean ‘Safe Trucking Freight Rates System’ (‘Safe Rates System’) was an important effort to address road safety risks by regulating road transport supply chains. In effect between 2020 and 2022, the system set minimum standards for truck driver pay and placed obligations on road transport supply chain parties to comply with these standards. In this article, we explain how the system developed in response to the deregulation and restructuring of the road freight market in the late 1980s and 1990s. We also trace the influence on the system of regulatory development in Australia and debates at the International Labour Organization (ILO). In 2019, workers, employers, and government representatives at the ILO reached an agreement on the main principles of the Safe Rates regulatory model through the adoption of the Guidelines on the promotion of decent work and road safety in the transport sector. We use these principles to explain and evaluate the Korean system. We also summarise assessments of the system’s impact, arguing that the results of the few studies that exist, justify the continuation of the system. By locating Korean Safe Rates as part of a broader global trend, we respond to opponents’ claims that the system is without international precedent and make the system eligible for a global audience. In so doing, we seek to contribute to the ongoing debate about the reintroduction of Safe Rates in Korea and draw lessons from the Korean experience that may be used in other countries.

Modified REL-Based Piecewise Path Loss Modeling Approach for Shore-to-Ship Communication at 5.6 GHz

The need for reliable and uninterrupted communication systems in the marine environment has become critically important with increasing maritime activities, environmental monitoring, and the spread of autonomous systems. However, the complex structure of electromagnetic wave propagation in a sea environment limits the accuracy of the existing propagation models. Thus, the Modified Round Earth Loss (REL) model was first developed in this study to estimate the path loss more accurately in shore-to-ship communication. Subsequently, a piecewise modeling approach based on the principle of two-segment data modeling was proposed. In the Modified REL model, unlike the traditional REL model, the paths and gains of the direct and reflected waves were not considered equal in the calculations. Moreover, unlike in the classical approach, the receiver height was not taken as a fixed value; the estimated best receiver height value for each measurement was included in the calculations as a representation of the effect of roughness in the sea environment. Thus, the model is better adapted to various environmental conditions. In addition, the proposed piecewise model divides the propagation medium into two regions using a break point calculated by Fresnel zone theory. The Modified REL model was used for the first region and the log-distance model was used for the second region. This method allows for more accurate modeling of signal behaviors, especially at different distances. Experimental measurements and performance evaluations conducted using four different shore-to-ship communication scenarios show that the Modified REL model shows an average improvement of up to 3% in Root Mean Square Error (RMSE) values compared to the classical REL model. Additionally, the proposed piecewise model improves the fitting error of the Modified REL model, which models the data as a single whole, by an average of 22.25%. These findings emphasize the necessity of propagation models that are sensitive and adaptable to environmental changes for maritime communication.

Contribution of transient heat transfer to overpressure protection in a passive Boiling Water Reactor

The BWRX-300 reactor is a small modular reactor with 300 MW nominal electric power output. As the abbreviation indicates, the reactor is a boiling type reactor using water as the coolant. The main operating principle of the reactor model is the use of natural circulation of the coolant for both steady-state operation and emergency cooling of the reactor. Isolation Condenser Systems (ICS) have been used in early BWRs and were reintroduced in the 1990s to implement passive safety.Modern technologies enable effective simulation of objects of various complexity. In this study, the thermal-hydraulic system code TRACE v.5 is used to simulate BWRX-300 reactor plant operation. The software enables calculation of a system consisting of the reactor pressure vessel and ICS in both steady-state and transient operation.The ICS is an emergency cooling system borrowed from the previous generation of GE Hitachi BWRs, the ESBWRs, and it consists of a vertical tube bundle heat exchanger immersed in a water tank. Although this system design remains unchanged, its purpose is different. The BWRX-300 design completely eliminates safety and relief valves. This is why the ICS must perform both overpressure protection and long term heat dissipation functions.Three reactor plant operation states were simulated: normal operation with nominal parameters; reactor isolation; and the transient process into the cooling down mode by reactor shutdown and ICS activation. The results demonstrate that the ICS is capable to mitigate overpressure transients. Upon ICS startup, its metallic structures absorb heat well in excess of its nominal steady-state rating. Initial thermal inertia of the system is important for overpressure mitigation in transients.

Experimental and numerical analyses of grid couplings in quasi-static and dynamic conditions

Grid couplings typically comprise flexible, spring-like elements connecting two toothed hubs. They are used in heavy machinery to transmit power from prime movers to driven parts, even in the presence of positioning errors and deviations. The main objective of this paper is to introduce a comprehensive three-dimensional model, which can be used to predict coupling stiffness characteristics and load distributions at the spring/hub tooth contacts. The modelling principles are presented with emphasis being placed on the spring/hub contact simulation based on a combination of finite elements and Winkler elastic foundations. A series of comparisons with experimental evidence from a specific test rig are shown, which prove the validity of the model and its applicability to industrial couplings.

Building inclusive Learning Development spaces ‘…but we don’t have time for universal design’

This project sets out to convert a Continuous Professional Development (CPD) workshop on Dyslexia awareness into a series of mini asynchronous learning experiences. The aim is that learning and teaching staff will access the short asynchronous activities by scanning a Quick Response (QR) code; these activities will provide a brief insight into relevant examples of neurodiverse experiences, followed by tips on changes to professional practice that can embed inclusive and universal design. Universal Design for Learning (UDL) is an ethos that we all strive for in our various roles in learning development; It embraces the principles of the social model of disability, which the Office of Students (2019) states ‘is not about “fixing” the individual but about restructuring environments and attitudes around them’. As highlighted by Dettmer and Welton (2024, p.62) the learning community seeks to promote the social model of disability when promoting inclusive learning environments. Stripe and Ntonia’s article on implementing change for inclusive teaching (2023) highlights that the challenges faced include lack of time for staff CPD, negative attitudes towards reasonable adjustments, and mindsets that are not conducive to inclusivity. These factors are sector wide; time is precious for many learning and teaching staff, and attending CPD workshops, particularly when it means adding to an existing heavy workload, are not typically at the top of priority lists. However, as proposed, engaging in a 30-60 second asynchronous activity, might lead to some ‘quick wins’ for the adoption of universal design principles. The proposed short asynchronous activities discussed here are based on experiences of neurodiverse learners when faced with non-inclusive environments, and provide strategies that can be immediately implemented. It is hoped that by scanning a QR code and engaging in learning experiences ‘on the go’, this approach will encourage staff to see that inclusive and universal design can start with small changes and have a fundamental impact on students’ learning experiences.

Generative technology for human emotion recognition: A scoping review

Affective computing stands at the forefront of artificial intelligence (AI), seeking to imbue machines with the ability to comprehend and respond to human emotions. Central to this field is emotion recognition, which endeavors to identify and interpret human emotional states from different modalities, such as speech, facial images, text, and physiological signals. In recent years, important progress has been made in generative models, including Autoencoder, Generative Adversarial Network, Diffusion Model, and Large Language Model. These models, with their powerful data generation capabilities, emerge as pivotal tools in advancing emotion recognition. However, up to now, there remains a paucity of systematic efforts that review generative technology for emotion recognition. This survey aims to bridge the gaps in the existing literature by conducting a comprehensive analysis of over 330 research papers until June 2024. Specifically, this survey will firstly introduce the mathematical principles of different generative models and the commonly used datasets. Subsequently, through a taxonomy, it will provide an in-depth analysis of how generative techniques address emotion recognition based on different modalities in several aspects, including data augmentation, feature extraction, semi-supervised learning, cross-domain, etc. Finally, the review will outline future research directions, emphasizing the potential of generative models to advance the field of emotion recognition and enhance the emotional intelligence of AI systems.

Analysis of the Current Status and Means of Studying Gravitational Lensing

Lens in physics is a kind of mirror that deflects light and results in a deflated and distorted image of the light-emitting object. Hence, the phenomenon of a massive celestial object deflecting light to form a distorted, magnified image of a luminous object is well known as gravitational lensing. In this paper, a brief summary of the concept and the status quo of the research and methodology related to it is rendered. This study offers a review of the overall founding history of the concept, and presents the current advances in relevant research. Later, the principles of generation and construction modeling of the concept, basic methodology are demonstrated. Subsequently, the applications of gravitational lensing in dark matter research, celestial body relationship research and validation of research results will be illustrated with examples of recent research. Then, the potential discrepancies in the use of gravitational lensing and future prospects are discussed in short. These results intended to help scholars who lack an understanding of the concept to have a primer on the field of gravitational lensing.

A Review of Comprehensive Guidelines for Computational Fluid Dynamics (CFD) Validation in Solar Chimney Power Plants: Methodology and Manzanares Prototype Case Study

This review provides a comprehensive examination of CFD modeling procedures for SCPP, with an emphasis on the detailed methodologies and a case study of the Manzanares prototype in Spain. The introduction delineates the historical context and physical modeling principles of solar chimneys, while highlighting their potential in industrial applications. The governing equations are meticulously discussed, covering assumptions in both 2D and 3D CFD modeling, the continuity and momentum equations, and the selection and accuracy of turbulence models, particularly the k-ε equations. The review also delves into heat transfer modeling, encompassing the energy equation and radiation modeling. Analytical evaluations of turbine pressure drop ratios and performance metrics for power generation efficiency are critically analyzed. The establishment of boundary conditions in solar chimney applications, including sky temperature assessments and distinctions between 2D and 3D boundary conditions, is extensively explored. Mesh generation techniques for both 2D and 3D CFD models are presented, supported by case studies. Parametric studies and experimental investigations are scrutinized to elucidate their impact on the performance of solar chimneys. The temperature–entropy diagram for an idealized Brayton cycle is introduced as a conceptual framework for efficiency analysis. Validation of the CFD codes, both 2D and 3D, against experimental data is performed to ensure model accuracy. The review further examines energy balance approaches in modeling solar chimneys, presenting state-of-the-art CFD results and discussing their implications in both 2D and 3D contexts. The synthesis of these findings culminates in a comprehensive conclusion, offering insights into the future directions and potential advancements in the CFD modeling of solar chimneys. This work aims to serve as a definitive reference for researchers and practitioners in the field, providing a robust foundation for the development and optimization of SCPP technology.

Transport mechanism of DgoT, a bacterial homolog of SLC17 organic anion transporters.

The solute carrier 17 (SLC17) family contains anion transporters that accumulate neurotransmitters in secretory vesicles, remove carboxylated monosaccharides from lysosomes, or extrude organic anions from the kidneys and liver. We combined classical molecular dynamics simulations, Markov state modeling and hybrid first principles quantum mechanical/classical mechanical (QM/MM) simulations with experimental approaches to describe the transport mechanisms of a model bacterial protein, the D-galactonate transporter DgoT, at atomic resolution. We found that protonation of D46 and E133 precedes galactonate binding and that substrate binding induces closure of the extracellular gate, with the conserved R47 coupling substrate binding to transmembrane helix movement. After isomerization to an inward-facing conformation, deprotonation of E133 and subsequent proton transfer from D46 to E133 opens the intracellular gate and permits galactonate dissociation either in its unprotonated form or after proton transfer from E133. After release of the second proton, apo DgoT returns to the outward-facing conformation. Our results provide a framework to understand how various SLC17 transport functions with distinct transport stoichiometries can be attained through subtle variations in proton and substrate binding/unbinding.

Four-parameter punching theory for reinforced concrete flat slabs without transverse reinforcement

Despite extensive scientific research, the underlying physical principles in theoretical models for understanding punching behavior remain a concern. In this paper, a four-parameter punching theory for reinforced concrete flat slabs without transverse reinforcement has been introduced and validated. The kinematic behavior of the slab was characterized by three key parameters, i.e., critical shear crack inclination, slab rotation, and slab translation. Four distinct shear transfer effects from aggregate interlocking, concrete residual tensile stress, reinforcement dowel action, and the shear-compression ring, were considered and quantified using a novel algorithm based on these three parameters, and the fourth parameter, i.e., the direction of the minimum principal stress within the shear-compression ring. The punching capacity was subsequently determined by summing the contributions from these effects. This model was evaluated through extensive comparisons with experimental data, demonstrating that it accurately captures the influence of various parameters. Across the collected 260 test data, the average ratio of experimental to predicted punching capacity was 1.01, with a coefficient of variance of 0.11 and a coefficient of determination of 0.97. The average ratio, coefficient of variation, and coefficient of determination improved by 8 %, 21 %, and 1 %, compared to the optimal results from other design and theoretical models. The developed model can also assess all shear contributions, slab deformation, and critical shear crack inclination. It is ultimately confirmed that the proposed model can elucidate the physical phenomena and underlying mechanisms involved in punching behavior. Slab deformation primarily arises from flexural deformation, whereas shear forces are predominantly transmitted through the shear-compression ring. Punching failure is attributed to the failure of the shear-compression ring under tension-compression-compression triaxial stress.

Dynamic uncertainty evaluation of cylindricity error based on Bayesian deep neural network propagation method

Most of the existing methods for measuring and evaluating product geometric tolerances are for static processes, which use lower dimensions, are more complex to calculate, and have poor traceability and dynamic manageability and control of the measurement process. In this paper, a dynamic evaluation model of cylindricity error uncertainty based on the Bayesian deep neural network propagation method is proposed. Firstly, the input random variable sampling sample containing error uncertainty is generated by proposing the improved Monte Carlo sampling method (IMCS), which has the characteristics of homogeneity and validity; then the principle and construction process of a Bayesian neural network model are introduced to generate the propagation model of multidimensional random error variables; further, the Gaussian mixture model is trained by the Monte Carlo method, and the probability density function of the target response is generated by using the Gaussian mixture model. Through numerical experiments, test analysis of bearing outer rings, and comparison with ISO standard methods, the results show that the model based on the dynamic Bayesian deep neural network propagation method can correctly and effectively assess the uncertainty dynamically, and the whole estimation process is stable.