Proper Model Research Articles

Domain generalization through meta-learning: a survey

Deep neural networks (DNNs) have revolutionized artificial intelligence but often lack performance when faced with out-of-distribution data, a common scenario due to the inevitable domain shifts in real-world applications. This limitation stems from the common assumption that training and testing data share the same distribution-an assumption frequently violated in practice. Despite their effectiveness with large amounts of data and computational power, DNNs struggle with distributional shifts and limited labeled data, leading to overfitting and poor generalization across various tasks and domains. Meta-learning presents a promising approach by employing algorithms that acquire transferable knowledge across various tasks for fast adaptation, eliminating the need to learn each task from scratch. This survey paper delves into the realm of meta-learning with a focus on its contribution to domain generalization. We first clarify the concept of meta-learning for domain generalization and introduce a novel taxonomy based on the feature extraction strategy and the classifier learning methodology, offering a granular view of methodologies. Additionally, we present a decision graph to assist readers in navigating the taxonomy based on data availability and domain shifts, enabling them to select and develop a proper model tailored to their specific problem requirements. Through an exhaustive review of existing methods and underlying theories, we map out the fundamentals of the field. Our survey provides practical insights and an informed discussion on promising research directions.

Controlling coating thickness distribution for a complex geometry with the help of simulation

This paper aims to develop a proper and valid simulation model for electroplating complex geometries. Since many variables influence the quality of the deposited coating and its thickness distribution, it is challenging to conduct efficient research only through experiments. In contrast, simulation can be an efficient way to optimize the electroplating experiments. Despite its potential, simulation has seen limited commercial use in the electroplating industry due to its inherent complexity and difficulty in achieving accurate precision for intricate geometries. The present study addresses the aspects that can enhance the electroplating simulation’s accuracy, which has been typically overlooked in the literature, such as the effect of current efficiency and its dependency on the current density, the input data for the electrode kinetics, the surface topology changes, and the differences between 2 and 3D simulations. The simulation model was validated by experimental results related to the coating thickness of Ni plating on a T-joint geometry. The results showed good agreement with the experimental ones, confirming the model’s ability to precisely predict the coating thickness and distribution and promote its broader utilization in the industry. Finally, the developed model was used to determine the optimal current density regime for achieving uniform coating thickness distribution on a T-joint sample.

Development of a reliable adaptive estimation approach for a low‐cost attitude and heading reference system

AbstractA main challenge within inertial navigation systems involves attitude determination, which refers to the process of estimating the angles that define orientations. A new algorithm is presented to estimate a reliable and proper model for a low‐cost attitude and heading reference system (AHRS), while also, including long‐term global navigation satellite system (GNSS) outages with various dynamical manoeuvres. In the proposed approach, the initial AHRS model is continuously refined at each time step through the incorporation of complementary terms using a new prediction method. These complementary terms are determined using attitude information from the accelerometers output and the GNSS as reference systems. The proposed estimation technique undergoes mathematical analysis to ascertain stochastic stability and is further assessed through real‐world aerial experimental tests conducted with hardware‐in‐the‐loop mechanisation. The obtained results demonstrate a significant enhancement in the reliability and accuracy of the AHRS through the proposed estimation algorithm, even under GNSS blockages. The comparative results highlight the higher performance of the suggested data fusion scheme in providing a reliable and accurate model for the AHRS in normal conditions and assisting a powerful neural network‐based algorithm to provide reliable heading information even during long‐term GNSS outages under dynamical manoeuvres.

Structure and Dynamics of Water in Polysaccharide (Alginate) Solutions and Gels Explained by the Core-Shell Model.

In both biological and engineered systems, polysaccharides offer a means of establishing structural stiffness without altering the availability of water. Notable examples include the extracellular matrix of prokaryotes and eukaryotes, artificial skin grafts, drug delivery materials, and gels for water harvesting. Proper design and modeling of these systems require detailed understanding of the behavior of water confined in pores narrower than about 1 nm. We use molecular dynamics simulations to investigate the properties of water in solutions and gels of the polysaccharide alginate as a function of the water content and polymer cross-linking. We find that a detailed understanding of the nanoscale dynamics of water in alginate solutions and gels requires consideration of the discrete nature of water. However, we also find that the trends in tortuosity, permeability, dielectric constant, and shear viscosity can be adequately represented using the "core-shell" conceptual model that considers the confined fluid as a continuum.

Wavelet-based generation of fully non-stationary random processes with application to seismic ground motions

Assessing the performance of earthquake-resistant structural and geotechnical systems is crucial for achieving a desired reliability level and enhancing the resilience of the built environment in seismic-prone regions. Nonlinear dynamic analyses are widely used to quantify structural and geotechnical performance. Still, they require an accurate representation of nonlinear behaviours and proper modelling of the expected seismic events. Stochastic approaches are popular strategies for modelling dynamic actions to account for the uncertain nature of ground shaking. In this framework, joint time–frequency signal representations of seismic records are powerful tools to analyze signals’ time-varying amplitude and frequency content. This paper presents a new method for the stochastic generation of artificial accelerograms using the circular harmonic wavelet transform, which possesses joint time–frequency localization capabilities and offers the engineers a clear and transparent interpretation of the results. The proposed approach adopts a new exponential auto-correlation structure for generating the random phases in the “child (generated) signals” starting from the deterministic ones in the “parent record”. The effects of the correlation structure and different subdivisions of earthquake records in frequency bands are investigated and discussed, leading to practical considerations for identifying an effective trade-off between localization in time and frequency domains. The method can be used for seismic assessment and design purposes, and numerical applications illustrate its potency.

Deep Learning Model for Detection and Severity Analysis of Car Accidents

Abstract On-road car accidents are immensely unfortunate but quite common occurrences worldwide. Instant data-centric and informed decisions of crisis management are rarely experienced due to the absence of real-time car accident detection and severity analysis mechanisms. On this background, the current paper presents a deep learning model for car accident detection and analysis of its severity so that the crisis management activities might follow without any delay saving invaluable human lives. The existing works lack in using time-series data, the proper learning model for accurate prediction, and minimizing the time taken in post-accident scenarios for the victims to receive immediate medical help. This paper introduces the Long Short Term Memory (LSTM) model in conjunction with the Gradient Boosted Regression Trees (GBRT) technique for the determination of car accidents with different levels of severity. The proposed model works with the accelerometer and gyroscopic data collected through an application installed in the smartphones of the users inside the car. The LSTM-GBRT hybrid model is proposed to achieve higher accuracy than LSTM which deals with time-variant data. The satisfactory performance of the proposed technique has been reported and the results are extensively investigated in comparison with another hybrid technique such as LSTM with Random Forest (RF) as well. The statistics confirm the superiority of the proposed model over other parallel models in terms of several performance metrics, like Accuracy, Precision, etc.

Usefulness of insights into the kinetic compensation effects in the kinetic analysis of the coal gasification process

AbstractIn this paper, an attempt has been made to understand the kinetic compensation effects and their usefulness in the kinetic analysis of a lab‐based gasification study. The gasification experiment was carried out in two different gasifying environments, that is, 0.4%O₂ + 8%H₂O‐Ar and 8%H₂O‐Ar, for two different particle sizes of Loy Yang brown coal. Analysis of kinetic values with the change in particle size and gasifying environment was investigated. This provides information on the path of product gas formation and how the overall controlling factor affects the path of char gasification, including the rate‐limiting step. Furthermore, the results indicate that having multiple sets of kinetic parameters caused by the inclusion of the change in char properties into kinetics during solid–gas heterogeneous reactions opens up the scope for wider applications in chemical reaction engineering. This includes the design of a reactor with a proper kinetic model, optimization of feedstock, and process parameters with the identification of pathways for product gas formation, which ultimately plays a key role in scaling up technology from bench‐scale to plant‐scale. In contrast, the study of kinetics with having a single set of kinetic data based on the initial change in char properties limits its applications.

Kinetic insights into heavy crude oil upgrading in supercritical water

Heavy crude oil upgrading in a supercritical water environment is a promising technology owing to diverse advantages such as heteroatom removal, high yields of low molecular weight fractions, and inhibition of coke production. In order to increase the performance of this process, some challenges need to be overcome by researching the reaction mechanisms and developing proper numerical models. Therefore, the kinetic studies reported in the literature for hydrothermal upgrading of heavy crude oil under supercritical water conditions and their results were systematically analyzed and discussed to achieve a better understanding of the diverse approaches considered to determine the distribution of reaction products and limitations. Additionally, fundamental aspects of the kinetic models, including experimental considerations and numerical methodologies applied for kinetic parameter estimation, are reviewed to obtain representative information about the reactant system that can be subsequently integrated into reactor design models or reservoir simulations.

A quantitative analysis of acoustic field directionality from a seismic airgun array

Marine seismic reflection surveys generate high amplitude impulsive acoustic events using airgun arrays to study geophysical characteristics of the seabed. These data can be used beyond seismic imaging – modeling short range propagation, considering impacts on marine mammals, extracting seabed properties and their effect on the acoustic field, etc. Knowledge of the source characteristics is necessary to utilize these data and proper modeling of sound propagation from these arrays requires characterization of the array beam pattern. Complex simulations of airgun arrays have been used in the past to model airgun array spectra, but beam patterns have not been thoroughly considered in the literature. Delay-and-sum combinations of these airgun signatures provide a simple beam pattern estimate, but this approach ignores variability in airgun position, timing, amplitude, interactions between airguns, etc. The use of more complex notional airgun signatures can yield more accurate estimates, but these are more challenging to model and still ignores shot-to-shot variability. Experimentally-determined beam patterns are evaluated here and the variability in the results considered, showing both similarities and notable differences from simulated results. The experimental results indicate that the source array depth impacts the ghost-free array beam pattern and that variability between shots is enough to significantly alter beam patterns. Overall, the observations suggest that accurate simulation of array beam patterns may require more complexity than is currently considered and that inclusion of uncertainty due to environmental and airgun shot variability is essential.

Exploring the Efficacy of Workshops as a Professional Teacher Development Strategy for Inclusive Education in South African Schools: Teachers’ Perspectives

For inclusive education to be a success, in-service teachers need to be thoroughly trained. learning and professional development programmes have been initiated across the globe including South Africa to prepare teachers. Workshops are well-known training methods in inclusive education. Surprisingly, despite their existence, literature points out that there’s still an inadequacy of teacher training. This complaint makes one question the efficacy of the workshops. To this end, this study explored the efficacy of workshops on Inclusive Education (IE) for Foundation Phase (FP) teachers in the Capricorn District of Limpopo Province. To achieve this, the case-study design in qualitative research was used. A purposive sampling strategy was used to sample 11 FP teachers from three ordinary schools. Data was collected through document reviews, in-depth interviews, and a focus group interview. The data was analysed through the thematic analysis method. The analysis revealed that according to teachers the workshops are not effective, for the following reasons: (i) the workshops only provide a theory that does not match the reality of the schools; (ii) the workshops are too short and not sufficiently informative; (iii) through teacher agency, teachers improvise to achieve IE. These findings suggest that IE workshops do get implemented in South African schools, however, their efficacy is questionable. Therefore, a proper model for the workshops that will be considerate of the schools’ context should be devised. This study may inform policy makers workshop facilitators, and researchers about what is working and what is not working regarding inclusion workshops. Keywords: Inclusive Education, Learning and Professional Development Programmes, Workshops, Teachers

An Exploration of Spatial Visualization Skills: Investigating Students' Use of 3D Models in Science Problems during Think-Aloud Interviews.

Effective spatial visualization and reasoning skills are often credited for students' success in science and engineering courses. However, students enrolled in these science courses are not always exposed to or trained properly on the best ways to utilize models to aid in their learning. Improving spatial visualization techniques with 3D models, such as molecular and DNA modeling kits, is often suggested to facilitate students' ability to conceptualize compounds in two and three dimensions. Here, we investigate what techniques students use to conceptualize 2D representations of various biomolecules with the use of 3D models by interviewing undergraduate students from various natural science and engineering disciplines in task-based, think-aloud sessions. After scoring and analyzing the participant data we explored some of the techniques used among successful scoring participants, including the use of informal models to transition between 2D and 3D. Additional techniques used by students who were able to successfully conceptualize 3D images included starting with smaller, granular details to inductively make conclusions when thinking between two and three dimensions. We find that (1) students who anchor their thinking in 3D models show a deeper level of understanding in initially solving science problems successfully, and (2) proper 3D model use and spatial visualization techniques may improve students' abilities to accurately visualize 2D and 3D representations of molecules in science courses. Our results demonstrate that implementing spatial visualization training to teach students how to effectively use 3D models may improve students' problem-solving techniques in science curricula.

Dynamics Parameter Identification of Articulated Robot

Dynamics parameter identification in the establishment of a multiple degree-of-freedom (DOF) robot’s dynamics model poses significant challenges. This study employs a non-symbolic numerical method to establish a dynamics model based on the Newton–Euler formula and then derives a proper dynamics model through decoupling. Initially, a minimum inertial parameter set is acquired by using QR decomposition, with the inclusion of a friction model in the robot dynamics model. Subsequently, the least squares method is employed to solve for the minimum inertial parameters, forming the basis for a comprehensive robot dynamics parameter identification system. Then, after the optimization of the genetic algorithm, the Fourier series trajectory function is used to derive the trajectory function for parameter identification. Validation of the robot’s dynamics parameter identification is performed through simulation and experimentation on a 6-DOF robot, leading to a precise identification value of the robot’s inertial parameters. Furthermore, two methods are employed to verify the inertia parameters, with analysis of experimental errors demonstrating the effectiveness of the robot dynamics parameter identification method. Overall, the effectiveness of the entire calibration system is verified by experiments, which can provide valuable insights for practical engineering applications, and a complete and effective robot dynamics parameter identification scheme for a 6-DOF robot is established and improved.

Accuracy of Green Bond Issuance Predictor

Climate change is affecting the development of many industries in different aspects. These impacted enterprises transform into sustainable enterprises to avoid the risks, and by doing so they enter into the green bond market. The current literature provides effective reference indicators for participants in the green bond market. These indicators illustrate the funding size of the green bonds in different dimensions to the participants. As for the improvement of the policies about environmental protection there also emerge some new indicators such as ESG score. Besides, machine learning is an accurate and effective tool in many fields, and some researchers have established a model for predicting the issuance of green bonds but have not involved the new indicators in the past. In this paper, on the one hand, we discuss the new indicator, ESG scores, and how it affects the funding size of the green bonds, on the other hand, we add this new indicator and the old indicators into four machine learning models to compare the accuracy of predicting the issuance of green bonds of these four models. In these four models, the Random Forest Regressor and LGBM Regressor are the best models on average. The former has the best performance of accuracy but needs much more time than the latter. On the opposite, the latter is the most efficient model among all but is the second most accurate. Besides, other models have the best numerical measurements in different dimensions which means we could use different models depending on different situations. Choosing the proper model for the specific situation can optimize the benefit of the green bond market participant.

Performance evaluation of an appointment-based medical service system with no-shows and emergency arrivals

In an appointment-based service system, patients usually experience two types of delays: appointment delay (virtual wait) and service delay (physical wait). Managers want to give patients more opportunities to make appointments and also want to increase the number of patients served. For this purpose, schedulers allow as many patients as possible to make appointments to wait virtually. Meanwhile, more patients are permitted to wait physically in the waiting room. Do these methods work? We define a discrete time tandem queueing system to analyse the performance measures (patients’ opportunities to make appointments, number of patients served, two waiting times, and utilization of the medical resources) in such a system. These performance measures are obtained by combining an absorbing Markov Chain and two embedded Markov Chains. The parameter sensitivity analysis suggests that increasing the number of patients waiting in the appointment queue and service queue cannot increase their opportunities to make appointments and to be served. Increasing the number of patients is detrimental to the service system. The comparative analysis is carried out to determine the proper model parameters for different management targets. A proper time between consecutive appointments can improve the utilization of medical resources, control patient flow, and increase patients’ access to services.

Analyzing the effect of infill density on the mechanical compression of ASA in additive manufacturing: a FEM perspective

Additive manufacturing (AM) represents a novel method for parts manufacturing, revolutionizing the design principles and processes. Among the different AM methods, fused filament fabrication (FFF) is one of the most widely employed and affordable, with numerous applications across a broad range of fields. Inherently, due to the fundamental physical mechanisms occurring during part building, the material acquires different properties compared to those of bulk material. Simultaneously, parameters such as the infill pattern and infill density significantly affect the overall behavior of the part. An efficient and effective tool to minimize the necessity for experimental investigations and to define the mechanical properties with respect to these parameters (i.e., infill density and pattern) is the finite element method (FEM). In the current study, accurate FEM models were developed and presented, considering the precise geometry of compression specimens for simulating the compression behavior of FFF-printed ASA polymer. More specifically, honeycomb infill patterns with different infill densities were simulated, and the results were validated by direct comparison to respective experimental results. It was deduced that utilizing an appropriate mesh size leads to higher precision and also increases the stability of the numerical simulation, while the FEM models can predict the loads as well as the deformed geometric shapes for different infill densities. As an overall conclusion, it is proved and reasoned that employing FEM and a proper modeling approach is indeed a feasible and efficient way to predict and define the compressive behavior of FFF parts.

Differential scanning calorimetry of proteins and Zimm–Bragg model in water

Differential Scanning Calorimetry (DSC) is a regular and powerful tool to measure the specific heat profile of various materials. Hydrogen bonds play a crucial role in stabilizing the three-dimensional structure of proteins. Naturally, information about the strength of hydrogen bonds is contained in the measured DSC profiles. Despite its obvious importance, there is no approach that would allow the extraction of such information from the heat capacity measurements. In order to connect the measured profile to microscopic properties of a polypeptide chain, a proper model is required to fit. Using recent advances in the Zimm–Bragg (ZB) theory of protein folding in water, we propose a new and efficient algorithm to process the DSC experimental data and to extract the H-bonding energy among other relevant constants. Thus, for the randomly picked set of 33 proteins, we have found a quite narrow distribution of hydrogen bonding energies from 1 to 8 kJ/mol with the average energy of intra-protein hydrogen bonds h¯=4.2±1.5 kJ/mol and the average energy of water–protein bonds as hps¯=3.8±1.5 kJ/mol. This is an important illustration of a tiny disbalance between the water–protein and intraprotein hydrogen bonds. Fitted values of the nucleation parameter σ belong to the range from 0.001 to 0.01, as expected. The reported method can be considered as complementary to the classical two-state approach and together with other parameters provides the protein–water and intraprotein H-bonding energies, not accessible within the two-state paradigm.

MP2-Based Composite Extrapolation Schemes Can Predict Core-Ionization Energies for First-Row Elements with Coupled-Cluster Level Accuracy.

X-ray photoelectron spectroscopy (XPS) measures core-electron binding energies (CEBEs) to reveal element-specific insights into the chemical environment and bonding. Accurate theoretical CEBE prediction aids XPS interpretation but requires proper modeling of orbital relaxation and electron correlation upon core-ionization. This work systematically investigates basis set selection for extrapolation to the complete basis set limit of CEBEs from ΔMP2 and ΔCC energies across 94 K-edges in diverse organic molecules. We demonstrate that an alternative composite scheme using ΔMP2 in a large basis corrected by ΔCC-ΔMP2 difference in a small basis can quantitatively recover optimally extrapolated ΔCC CEBEs within 0.02 eV. Unlike ΔCC, MP2 calculations do not suffer from convergence issues and are computationally cheaper, and thus, the composite ΔMP2/ΔCC scheme balances accuracy and cost, overcoming limitations of solely using either method. We conclude by providing a comprehensive analysis of the choice of small and large basis sets for the composite schemes and provide practical recommendations for highly accurate (within 0.10-0.15 eV MAE) ab initio prediction of XPS data.

Modelling breakage of steelmaking materials for simulation of mechanical degradation during handling

Mechanical degradation of steelmaking materials can have significant economic and environmental implications. Such degradation typically occurs as result of impacts during transportation and handling from their production site to the steelmaking furnace, in operations that include transfers in chutes, ship loading, stockpiling, reclaiming, wagon loading and discharging, etc. Predicting the extent of such degradation and attempting to prevent it, is worthwhile. Fortunately, mechanical degradation during such operations can now be predicted using different approaches, but a proper breakage model is required. The work presents results from single-particle breakage tests of iron ore lump, sinter, briquette, and pellet samples, besides coke, followed by fitting parameters of the Tavares UFRJ Breakage Model. These are then used to simulate degradation in a hypothetical handling circuit. Predictions demonstrate the high susceptibility of iron ore lump and briquette to fines generation, when compared to the other materials, and the occurrence of conditioning for briquettes and sinter, which is the increase in strength of particles surviving handling. Iron ore pellets and coke remained comparatively unaffected by the handling operations simulated, which is explained by their greater resistance to either body or surface breakage.

Tuning Ceramic Surface to Minimize the Ionic Resistance at the Interface between PEO- and LATP-Based Ceramic Electrolyte.

New battery technologies are currently under development, and among them, all-solid-state batteries should deliver better electrochemical performance and enhanced safety. Composite solid electrolytes, combining a solid polymer electrolyte (SPE) and a ceramic electrolyte (CE), should then provide high ionic conductivity coupled to high mechanical stability. To date, this synergy has not yet been reached due to the complexity of the Li-ion transport within the hybrid solid electrolyte, especially at the SPE/CE interface currently considered the limiting step. Yet, there is no proper kinetic model to elucidate the parameters influencing this interfacial barrier. The limited understanding of the SPE/CE interface can be partly explained by scattered SPE/CE interface resistances reported in the literature as well as the lack of systematic studies. Herein, we propose a systematic study of the effect on the SPE/CE interfacial resistance of chemical and thermal treatments of a model LATP-based ceramic based on a methodology relying on electrochemical impedance spectroscopy (EIS) and X-ray photoemission spectroscopy (XPS). The results provide different levers for the optimization of this interface and valuable insights into experimental precautions needed to obtain more reproducible results.

Tensile constitutive relationship of UHPFRC from crack hinge based inverse analysis of notched beams

The structural applications of ultra-high performance fiber reinforced concrete (UHPFRC) in the construction industry can be promoted if proper constitutive models can be established for design. The various tension test methods for UHPFRC are being adopted to develop a complete stress-crack width or strain response for design. This study aims to understand the fracture behavior of UHPFRC by testing notched beams under a three-point bending configuration. This testing method helps in quality control checks and obtaining complete tension stress-crack opening behavior that can be used for design. The tensile stress-crack width response of UHPFRC is obtained using the load-CMOD response. The variables considered in this study are the volume fraction of fibers (1 % and 2 %), types of steel fibers (straight, combination of straight and hooked ended), and different cross-section dimensions (100×100×500 mm and 75×75×500 mm). A crack hinge-based inverse analysis method is used for UHPFRC to obtain the complete tensile stress-crack opening response. Stress-strain relations from inverse analysis and direct tension stress-crack opening results are compared and found to be similar. A data-driven tension stress-crack opening response UHPFRC is verified by simulating the girder's response using nonlinear finite element analysis (NFEA). Thus, the proposed inverse analysis can be used to obtain the multi-linear tensile stress-crack opening response which is adequate for capturing the structural response.