Data-driven casting defect prediction model for sand casting based on random forest classification algorithm

The complex sand-casting process combined with the interactions between process parameters makes it difficult to control the casting quality,resulting in a high scrap rate.A strategy based on a data-driven model was proposed to reduce casting defects and improve production efficiency,which includes the random forest(RF)classification model,the feature importance analysis,and the process parameters optimization with Monte Carlo simulation.The collected data includes four types of defects and corresponding process parameters were used to construct the RF model.Classification results show a recall rate above 90% for all categories.The Gini Index was used to assess the importance of the process parameters in the formation of various defects in the RF model.Finally,the classification model was applied to different production conditions for quality prediction.In the case of process parameters optimization for gas porosity defects,this model serves as an experimental process in the Monte Carlo method to estimate a better temperature distribution.The prediction model,when applied to the factory,greatly improved the efficiency of defect detection.Results show that the scrap rate decreased from 10.16% to 6.68%.

Background removal from global auroral images:Data-driven dayglow modeling

Global images of auroras obtained by cameras on spacecraft are a key tool for studying the near-Earth environment.However,the cameras are sensitive not only to auroral emissions produced by precipitating particles,but also to dayglow emissions produced by photoelectrons induced by sunlight.Nightglow emissions and scattered sunlight can contribute to the background signal.To fully utilize such images in space science,background contamination must be removed to isolate the auroral signal.Here we outline a data-driven approach to modeling the background intensity in multiple images by formulating linear inverse problems based on B-splines and spherical harmonics.The approach is robust,flexible,and iteratively deselects outliers,such as auroral emissions.The final model is smooth across the terminator and accounts for slow temporal variations and large-scale asymmetries in the dayglow.We demonstrate the model by using the three far ultraviolet cameras on the Imager for Magnetopause-to-Aurora Global Exploration(IMAGE)mission.The method can be applied to historical missions and is relevant for upcoming missions,such as the Solar wind Magnetosphere Ionosphere Link Explorer(SMILE)mission.

Artificial intelligence-driven radiomics study in cancer:the role of feature engineering and modeling

Modern medicine is reliant on various medical imaging technologies for non-invasively observing patients’anatomy.However,the interpretation of medical images can be highly subjective and dependent on the expertise of clinicians.Moreover,some potentially useful quantitative information in medical images,especially that which is not visible to the naked eye,is often ignored during clinical practice.In contrast,radiomics performs high-throughput feature extraction from medical images,which enables quantitative analysis of medical images and prediction of various clinical endpoints.Studies have reported that radiomics exhibits promising performance in diagnosis and predicting treatment responses and prognosis,demonstrating its potential to be a non-invasive auxiliary tool for personalized medicine.However,radiomics remains in a developmental phase as numerous technical challenges have yet to be solved,especially in feature engineering and statistical modeling.In this review,we introduce the current utility of radiomics by summarizing research on its application in the diagnosis,prognosis,and prediction of treatment responses in patients with cancer.We focus on machine learning approaches,for feature extraction and selection during feature engineering and for imbalanced datasets and multi-modality fusion during statistical modeling.Furthermore,we introduce the stability,reproducibility,and interpretability of features,and the generalizability and interpretability of models.Finally,we offer possible solutions to current challenges in radiomics research.

Data-driven modeling on anisotropic mechanical behavior of brain tissue with internal pressure

Brain tissue is one of the softest parts of the human body,composed of white matter and grey matter.The mechanical behavior of the brain tissue plays an essential role in regulating brain morphology and brain function.Besides,traumatic brain injury(TBI)and various brain diseases are also greatly influenced by the brain's mechanical properties.Whether white matter or grey matter,brain tissue contains multiscale structures composed of neurons,glial cells,fibers,blood vessels,etc.,each with different mechanical properties.As such,brain tissue exhibits complex mechanical behavior,usually with strong nonlinearity,heterogeneity,and directional dependence.Building a constitutive law for multiscale brain tissue using traditional function-based approaches can be very challenging.Instead,this paper proposes a data-driven approach to establish the desired mechanical model of brain tissue.We focus on blood vessels with internal pressure embedded in a white or grey matter matrix material to demonstrate our approach.The matrix is described by an isotropic or anisotropic nonlinear elastic model.A representative unit cell(RUC)with blood vessels is built,which is used to generate the stress-strain data under different internal blood pressure and various proportional displacement loading paths.The generated stress-strain data is then used to train a mechanical law using artificial neural networks to predict the macroscopic mechanical response of brain tissue under different internal pressures.Finally,the trained material model is implemented into finite element software to predict the mechanical behavior of a whole brain under intracranial pressure and distributed body forces.Compared with a direct numerical simulation that employs a reference material model,our proposed approach greatly reduces the computational cost and improves modeling efficiency.The predictions made by our trained model demonstrate sufficient accuracy.Specifically,we find that the level of internal blood pressure can greatly influence stress distribution and determine the possible related damage behaviors.

Data-model coupling driven stress field measurements

This paper presents a method for measuring stress fields within the framework of coupled data models,aimed at determining stress fields in isotropic material structures exhibiting localized deterioration behavior without relying on constitutive equations in the deteriorated region.This approach contributes to advancing the field of intrinsic equation-free mechanics.The methodology combines measured strain fields with data-model coupling driven algorithms.The gradient and Canny operators are utilized to process the strain field data,enabling the determination of the deterioration region's location.Meanwhile,an adaptive model building method is proposed for constructing coupling driven models.To address the issue of unknown datasets during computation,a dataset updating strategy based on a differential evolutionary algorithm is introduced.The resulting optimal dataset is then used to generate stress field results.Validation against finite element method calculations demonstrates the accuracy of the proposed method in obtaining full-field stresses in specimens with local degradation behavior.

Parameter-driven Level of Detail Derivation Method for Semantic Building Facade Model

The relentless progress in the research of geographic spatial data models and their application scenarios is propelling an unprecedented rich Level of Detail(LoD)in realistic 3D representation and smart cities.This pursuit of rich details not only adds complexity to entity models but also poses significant computational challenges for model visualization and 3D GIS.This paper introduces a novel method for deriving multi-LOD models,which can enhance the efficiency of spatial computing in complex 3D building models.Firstly,we extract multiple facades from a 3D building model(LoD3)and convert them into individual semantic facade models.Through the utilization of the developed facade layout graph,each semantic facade model is then transformed into a parametric model.Furthermore,we explore the specification of geometric and semantic details in building facades and define three different LODs for facades,offering a unique expression.Finally,an innovative heuristic method is introduced to simplify the parameterized facade.Through rigorous experimentation and evaluation,the effectiveness of the proposed parameterization methodology in capturing complex geometric details,semantic richness,and topological relationships of 3D building models is demonstrated.

Phase diagram and quench dynamics of a periodically drivenHaldane model

We investigate a periodically driven Haldane model subjected to a two-stage driving scheme in the form of a step function.By using the Floquet theory,we obtain the topological phase diagram of the system.We also find that anomalous Floquet topological phases exist in the system.Focusing on examining the quench dynamics among topological phases,we analyze the site distribution of the 0-mode and p-mode edge states in long-period evolution after a quench.The results demonstrate that,under certain conditions,the site distribution of the 0-mode can be confined at the edge even in long-period evolution.Additionally,both the 0-mode and p-mode can recover and become confined at the edge in long-period evolution when the post-quench parameters(T,M_(2) /M_(1))in the phase diagram cross away from the phase boundary (M_(2)/ M_(1))=(6√3t2)/ M_(1)−1.Furthermore,we conclude that whether the edge state is confined at the edge in the long-period evolution after a quench depends on the similarity of the edge states before and after the quench.Our findings reveal some new characteristics of quench dynamics in a periodically driven system.

Smaller & Smarter: Score-Driven Network Chaining of Smaller Language Models

With the continuous evolution and expanding applications of Large Language Models (LLMs), there has been a noticeable surge in the size of the emerging models. It is not solely the growth in model size, primarily measured by the number of parameters, but also the subsequent escalation in computational demands, hardware and software prerequisites for training, all culminating in a substantial financial investment as well. In this paper, we present novel techniques like supervision, parallelization, and scoring functions to get better results out of chains of smaller language models, rather than relying solely on scaling up model size. Firstly, we propose an approach to quantify the performance of a Smaller Language Models (SLM) by introducing a corresponding supervisor model that incrementally corrects the encountered errors. Secondly, we propose an approach to utilize two smaller language models (in a network) performing the same task and retrieving the best relevant output from the two, ensuring peak performance for a specific task. Experimental evaluations establish the quantitative accuracy improvements on financial reasoning and arithmetic calculation tasks from utilizing techniques like supervisor models (in a network of model scenario), threshold scoring and parallel processing over a baseline study.

Analytical Solutions for Testing of Baroclinic and Wind-Driven Three-Dimensional Hydrodynamic Models

It is always a challenge for a model developer to verify a three-dimensional hydrodynamic model, especially for the baroclinic term over variable topography, due to a lack of observational data sets or suitable analytical solutions. In this paper, exact solutions for the periodic forcing by surface heat flux and wind stress are given by solving the linearized equations of motion neglecting the rotation, advection and horizontal diffusion terms. The temperature at the bottom is set to a prescribed periodic value and a slip condition on flow is enforced at the bottom. The geometry of the quarter annulus, which has been extensively studied for two- and three-dimensional analytical solutions of unstratified water bodies, is used with a general power law variation of the bottom slope in the radial direction and is constant in the azimuthal direction. The analytical solutions are derived in a cylindrical coordinate system, which describes the three-dimensional fluid field in a Cartesian coordinate system. The results presented in this paper should provide a foundation for studying and verifying the baroclinic term over a varied topography in a three-dimensional numerical model.

A data and physical model dual-driven based trajectory estimator for long-term navigation

Long-term navigation ability based on consumer-level wearable inertial sensors plays an essential role towards various emerging fields, for instance, smart healthcare, emergency rescue, soldier positioning et al. The performance of existing long-term navigation algorithm is limited by the cumulative error of inertial sensors, disturbed local magnetic field, and complex motion modes of the pedestrian. This paper develops a robust data and physical model dual-driven based trajectory estimation(DPDD-TE) framework, which can be applied for long-term navigation tasks. A Bi-directional Long Short-Term Memory(Bi-LSTM) based quasi-static magnetic field(QSMF) detection algorithm is developed for extracting useful magnetic observation for heading calibration, and another Bi-LSTM is adopted for walking speed estimation by considering hybrid human motion information under a specific time period. In addition, a data and physical model dual-driven based multi-source fusion model is proposed to integrate basic INS mechanization and multi-level constraint and observations for maintaining accuracy under long-term navigation tasks, and enhanced by the magnetic and trajectory features assisted loop detection algorithm. Real-world experiments indicate that the proposed DPDD-TE outperforms than existing algorithms, and final estimated heading and positioning accuracy indexes reaches 5° and less than 2 m under the time period of 30 min, respectively.

Data-Driven Modeling for Wind Turbine Blade Loads Based on Deep Neural Network

Blades are essential components of wind turbines.Reducing their fatigue loads during operation helps to extend their lifespan,but it is difficult to quickly and accurately calculate the fatigue loads of blades.To solve this problem,this paper innovatively designs a data-driven blade load modeling method based on a deep learning framework through mechanism analysis,feature selection,and model construction.In the mechanism analysis part,the generation mechanism of blade loads and the load theoretical calculationmethod based on material damage theory are analyzed,and four measurable operating state parameters related to blade loads are screened;in the feature extraction part,15 characteristic indicators of each screened parameter are extracted in the time and frequency domain,and feature selection is completed through correlation analysis with blade loads to determine the input parameters of data-driven modeling;in the model construction part,a deep neural network based on feedforward and feedback propagation is designed to construct the nonlinear coupling relationship between the unit operating parameter characteristics and blade loads.The results show that the proposed method mines the wind turbine operating state characteristics highly correlated with the blade load,such as the standard deviation of wind speed.The model built using these characteristics has reasonable calculation and fitting capabilities for the blade load and shows a better fitting level for untrained out-of-sample data than the traditional scheme.Based on the mean absolute percentage error calculation,the modeling accuracy of the two blade loads can reach more than 90%and 80%,respectively,providing a good foundation for the subsequent optimization control to suppress the blade load.

Static-to-kinematic modeling and experimental validation of tendon-driven quasi continuum manipulators with nonconstant subsegment stiffness

Continuum robots with high flexibility and compliance have the capability to operate in confined and cluttered environments. To enhance the load capacity while maintaining robot dexterity, we propose a novel non-constant subsegment stiffness structure for tendon-driven quasi continuum robots(TDQCRs) comprising rigid-flexible coupling subsegments.Aiming at real-time control applications, we present a novel static-to-kinematic modeling approach to gain a comprehensive understanding of the TDQCR model. The analytical subsegment-based kinematics for the multisection manipulator is derived based on screw theory and product of exponentials formula, and the static model considering gravity loading,actuation loading, and robot constitutive laws is established. Additionally, the effect of tension attenuation caused by routing channel friction is considered in the robot statics, resulting in improved model accuracy. The root-mean-square error between the outputs of the static model and the experimental system is less than 1.63% of the arm length(0.5 m). By employing the proposed static model, a mapping of bending angles between the configuration space and the subsegment space is established. Furthermore, motion control experiments are conducted on our TDQCR system, and the results demonstrate the effectiveness of the static-to-kinematic model.

Robust Platoon Control of Mixed Autonomous and Human-Driven Vehicles for Obstacle Collision Avoidance:A Cooperative Sensing-Based Adaptive Model Predictive Control Approach

Obstacle detection and platoon control for mixed traffic flows,comprising human-driven vehicles(HDVs)and connected and autonomous vehicles(CAVs),face challenges from uncertain disturbances,such as sensor faults,inaccurate driver operations,and mismatched model errors.Furthermore,misleading sensing information or malicious attacks in vehicular wireless networks can jeopardize CAVs’perception and platoon safety.In this paper,we develop a two-dimensional robust control method for a mixed platoon,including a single leading CAV and multiple following HDVs that incorpo-rate robust information sensing and platoon control.To effectively detect and locate unknown obstacles ahead of the leading CAV,we propose a cooperative vehicle-infrastructure sensing scheme and integrate it with an adaptive model predictive control scheme for the leading CAV.This sensing scheme fuses information from multiple nodes while suppressing malicious data from attackers to enhance robustness and attack resilience in a distributed and adaptive manner.Additionally,we propose a distributed car-following control scheme with robustness to guarantee the following HDVs,considering uncertain disturbances.We also provide theoretical proof of the string stability under this control framework.Finally,extensive simulations are conducted to validate our approach.The simulation results demonstrate that our method can effectively filter out misleading sensing information from malicious attackers,significantly reduce the mean-square deviation in obstacle sensing,and approach the theoretical error lower bound.Moreover,the proposed control method successfully achieves obstacle avoidance for the mixed platoon while ensuring stability and robustness in the face of external attacks and uncertain disturbances.

Air-Side Heat Transfer Performance Prediction for Microchannel Heat Exchangers Using Data-Driven Models with Dimensionless Numbers

This study explores the effectiveness of machine learning models in predicting the air-side performance of microchannel heat exchangers.The data were generated by experimentally validated Computational Fluid Dynam-ics(CFD)simulations of air-to-water microchannel heat exchangers.A distinctive aspect of this research is the comparative analysis of four diverse machine learning algorithms:Artificial Neural Networks(ANN),Support Vector Machines(SVM),Random Forest(RF),and Gaussian Process Regression(GPR).These models are adeptly applied to predict air-side heat transfer performance with high precision,with ANN and GPR exhibiting notably superior accuracy.Additionally,this research further delves into the influence of both geometric and operational parameters—including louvered angle,fin height,fin spacing,air inlet temperature,velocity,and tube temperature—on model performance.Moreover,it innovatively incorporates dimensionless numbers such as aspect ratio,fin height-to-spacing ratio,Reynolds number,Nusselt number,normalized air inlet temperature,temperature difference,and louvered angle into the input variables.This strategic inclusion significantly refines the predictive capabilities of the models by establishing a robust analytical framework supported by the CFD-generated database.The results show the enhanced prediction accuracy achieved by integrating dimensionless numbers,highlighting the effectiveness of data-driven approaches in precisely forecasting heat exchanger performance.This advancement is pivotal for the geometric optimization of heat exchangers,illustrating the considerable potential of integrating sophisticated modeling techniques with traditional engineering metrics.

Artificial Intelligence-Driven FVM-ANNModel for Entropy Analysis ofMHD Natural Bioconvection in Nanofluid-Filled Porous Cavities

The research examines fluid behavior in a porous box-shaped enclosure.The fluid contains nanoscale particles and swimming microbes and is subject to magnetic forces at an angle.Natural circulation driven by biological factors is investigated.The analysis combines a traditional numerical approach with machine learning techniques.Mathematical equations describing the system are transformed into a dimensionless form and then solved using computational methods.The artificial neural network(ANN)model,trained with the Levenberg-Marquardt method,accurately predicts(Nu)values,showing high correlation(R=1),low mean squared error(MSE),and minimal error clustering.Parametric analysis reveals significant effects of parameters,length and location of source(B),(D),heat generation/absorption coefficient(Q),and porosity parameter(ε).Increasing the cooling area length(B)reduces streamline intensity and local Nusselt and Sherwood numbers,while decreasing isotherms,isoconcentrations,and micro-rotation.The Bejan number(Be+)decreases with increasing(B),whereas(Be+++),and global entropy(e+++)increase.Variations in(Q)slightly affect streamlines but reduce isotherm intensity and average Nusselt numbers.Higher(D)significantly impacts isotherms,iso-concentrations,andmicro-rotation,altering streamline contours and local Bejan number distribution.Increased(ε)enhances streamline strength and local Nusselt number profiles but has mixed effects on average Nusselt numbers.These findings highlight the complex interactions between cooling area length,fluid flow,and heat transfer properties.By combining finite volume method(FVM)with machine learning technique,this study provides valuable insights into the complex interactions between key parameters and heat transfer,contributing to the development of more efficient designs in applications such as cooling systems,energy storage,and bioengineering.

Comparative analysis of empirical and deep learning models for ionospheric sporadic E layer prediction

Sporadic E(Es)layers in the ionosphere are characterized by intense plasma irregularities in the E region at altitudes of 90-130 km.Because they can significantly influence radio communications and navigation systems,accurate forecasting of Es layers is crucial for ensuring the precision and dependability of navigation satellite systems.In this study,we present Es predictions made by an empirical model and by a deep learning model,and analyze their differences comprehensively by comparing the model predictions to satellite RO measurements and ground-based ionosonde observations.The deep learning model exhibited significantly better performance,as indicated by its high coefficient of correlation(r=0.87)with RO observations and predictions,than did the empirical model(r=0.53).This study highlights the importance of integrating artificial intelligence technology into ionosphere modelling generally,and into predicting Es layer occurrences and characteristics,in particular.

Big Model Strategy for Bridge Structural Health Monitoring Based on Data-Driven, Adaptive Method and Convolutional Neural Network (CNN) Group

This study introduces an innovative“Big Model”strategy to enhance Bridge Structural Health Monitoring(SHM)using a Convolutional Neural Network(CNN),time-frequency analysis,and fine element analysis.Leveraging ensemble methods,collaborative learning,and distributed computing,the approach effectively manages the complexity and scale of large-scale bridge data.The CNN employs transfer learning,fine-tuning,and continuous monitoring to optimize models for adaptive and accurate structural health assessments,focusing on extracting meaningful features through time-frequency analysis.By integrating Finite Element Analysis,time-frequency analysis,and CNNs,the strategy provides a comprehensive understanding of bridge health.Utilizing diverse sensor data,sophisticated feature extraction,and advanced CNN architecture,the model is optimized through rigorous preprocessing and hyperparameter tuning.This approach significantly enhances the ability to make accurate predictions,monitor structural health,and support proactive maintenance practices,thereby ensuring the safety and longevity of critical infrastructure.

Data-Driven Healthcare:The Role of Computational Methods in Medical Innovation

The purpose of this review is to explore the intersection of computational engineering and biomedical science,highlighting the transformative potential this convergence holds for innovation in healthcare and medical research.The review covers key topics such as computational modelling,bioinformatics,machine learning in medical diagnostics,and the integration of wearable technology for real-time health monitoring.Major findings indicate that computational models have significantly enhanced the understanding of complex biological systems,while machine learning algorithms have improved the accuracy of disease prediction and diagnosis.The synergy between bioinformatics and computational techniques has led to breakthroughs in personalized medicine,enabling more precise treatment strategies.Additionally,the integration of wearable devices with advanced computational methods has opened new avenues for continuous health monitoring and early disease detection.The review emphasizes the need for interdisciplinary collaboration to further advance this field.Future research should focus on developing more robust and scalable computational models,enhancing data integration techniques,and addressing ethical considerations related to data privacy and security.By fostering innovation at the intersection of these disciplines,the potential to revolutionize healthcare delivery and outcomes becomes increasingly attainable.

Sustainable-Driven Business Model Innovation in Management

The concept of sustainability has evolved to become an engine of innovation in business models in numerous sectors.Businesses are increasingly aware of the need to incorporate sustainable practices to generate value over time while tackling environmental and social issues.This paper examines how companies use sustainability as a source of innovation in developing new market ventures and profit-increasing strategies.Based on case studies and analyses,the paper demonstrates how sustainable innovation struggling in businesses is evidenced in real-time and suggests principles and difficulties put positive changes in place.

Aquaporin-4-IgG-seropositive neuromyelitis optica spectrum disorders:progress of experimental models based on disease pathogenesis

Neuromyelitis optica spectrum disorders are neuroinflammatory demyelinating disorders that lead to permanent visual loss and motor dysfunction.To date,no effective treatment exists as the exact causative mechanism remains unknown.Therefore,experimental models of neuromyelitis optica spectrum disorders are essential for exploring its pathogenesis and in screening for therapeutic targets.Since most patients with neuromyelitis optica spectrum disorders are seropositive for IgG autoantibodies against aquaporin-4,which is highly expressed on the membrane of astrocyte endfeet,most current experimental models are based on aquaporin-4-IgG that initially targets astrocytes.These experimental models have successfully simulated many pathological features of neuromyelitis optica spectrum disorders,such as aquaporin-4 loss,astrocytopathy,granulocyte and macrophage infiltration,complement activation,demyelination,and neuronal loss;however,they do not fully capture the pathological process of human neuromyelitis optica spectrum disorders.In this review,we summarize the currently known pathogenic mechanisms and the development of associated experimental models in vitro,ex vivo,and in vivo for neuromyelitis optica spectrum disorders,suggest potential pathogenic mechanisms for further investigation,and provide guidance on experimental model choices.In addition,this review summarizes the latest information on pathologies and therapies for neuromyelitis optica spectrum disorders based on experimental models of aquaporin-4-IgG-seropositive neuromyelitis optica spectrum disorders,offering further therapeutic targets and a theoretical basis for clinical trials.