Updated Model Research Articles

Resource-Efficient Clustered Federated Learning Framework for Industry 4.0 Edge Devices

Industry 4.0 is an aggregate of recent technologies including artificial intelligence, big data, edge computing, and the Internet of Things (IoT) to enhance efficiency and real-time decision-making. Industry 4.0 data analytics demands a privacy-focused approach, and federated learning offers a viable solution for such scenarios. It allows each edge device to train the model locally using its own collected data and shares only the model updates with the server without the need to share real collected data. However, communication and computational costs for sharing model updates and performance are major bottlenecks for resource-constrained edge devices. This study introduces a representative-based parameter-sharing framework that aims to enhance the efficiency of federated learning in the Industry 4.0 environment. The framework begins with a server by distributing an initial model to edge devices, which then train it locally and send updated parameters back to the server for aggregation. To reduce communication and computational costs, the framework identifies groups of devices with similar parameter distributions and only sends updates from the resourceful and better-performing device, termed the cluster head, to the server. A backup cluster head is also elected to ensure reliability. Clustering is performed based on the device’s parameter distributions and data characteristics. Moreover, the server incorporates randomly selected past aggregated parameters into the current aggregation process through weighted averaging where more recent parameters are given greater weight to enhance model performance. Comparative experimental evaluation with the state of the art using a testbed dataset demonstrates promising results by minimizing computational cost while preserving prediction performance, which ultimately enhances data analytics on edge devices in industrial environments.

COCOMO2: A Coarse-Grained Model for Interacting Folded and Disordered Proteins.

Biomolecular interactions are essential in many biological processes, including complex formation and phase separation processes. Coarse-grained computational models are especially valuable for studying such processes via simulation. Here, we present COCOMO2, an updated residue-based coarse-grained model that extends its applicability from intrinsically disordered peptides to folded proteins. This is accomplished with the introduction of a surface exposure scaling factor, which adjusts interaction strengths based on solvent accessibility, to enable the more realistic modeling of interactions involving folded domains without additional computational costs. COCOMO2 was parametrized directly with solubility and phase separation data to improve its performance on predicting concentration-dependent phase separation for a broader range of biomolecular systems compared to the original version. COCOMO2 enables new applications including the study of condensates that involve IDPs together with folded domains and the study of complex assembly processes. COCOMO2 also provides an expanded foundation for the development of multiscale approaches for modeling biomolecular interactions that span from residue-level to atomistic resolution.

Thermoelastic Properties of Iron‐Rich Ringwoodite and the Deep Mantle Aerotherm of Mars

AbstractThe Martian mantle is considered to have a higher Fe/Mg ratio than the Earth's mantle. Ringwoodite, γ‐(Mg,Fe)2SiO4, is likely the dominant polymorph of olivine in the core‐mantle boundary (CMB) region of Mars. We synthesized anhydrous iron‐rich ringwoodite with molar Mg/(Mg + Fe) = 0.44 and determined its thermal equation of state up to 35 GPa and 750 K by synchrotron X‐ray diffraction. Using a third order Birch‐Murnaghan equation of state, we obtain KT0 = 182 (3) GPa, K′ = 4.6 (2), and α0 = 3.18 (6) × 10−5 K−1. Using these results and an updated mineralogical model with an iron‐rich composition of Mg/(Mg + Fe) = 0.75 for the Martian mantle, we estimate ∼1900 K for the temperature of the D1000 seismic discontinuity inside Mars. The resulting adiabat predicts a warm aerotherm, which could explain the presence of partial melt at the CMB of Mars recently detected with seismic data from the 2019 InSight mission.

Bayesian model updating method based on multi-sampling strategy integration

The Bayesian model updating method is widely applied in structural health monitoring. Traditional Bayesian model updating methods suffer from limitations such as dimensional constraints, slow convergence, and low computational efficiency. To enhance the convergence speed and computational efficiency of Bayesian model updating methods, this paper proposes a multi-sampling strategy integrated Bayesian model updating method based on the Differential Evolution Adaptive Metropolis (DREAM) algorithm. First, the Kalman-inspired distribution and sampling difference vectors from past states are introduced into the DREAM algorithm, addressing the issue of parallel chain limitations in DREAM and improving the exploration efficiency of the posterior distribution. Second, drawing inspiration from clustering algorithms that use centroids for data point clustering, a novel centroid update sampling strategy is proposed and integrated with other sampling strategies to increase sampling diversity among different chains, thereby avoiding local optima and accelerating the convergence process. Finally, the proposed method’s effectiveness is validated through numerical examples of simply supported beams and experimental examples of a three-story frame structure. The results demonstrate that the proposed Bayesian model updating method based on multi-sampling strategy integration achieves high updating accuracy and faster convergence. The updated model shows improved ability to simulate the inherent properties and response characteristics of actual structures, making it suitable for use as a benchmark model in structural health monitoring.

Wave-Equation Moment Tensor Inversion of Microseismic Events Using Multi-Well DAS Recording

The directional nature of Distributed Acoustic Sensing (DAS) measurements can limit their usefulness for moment tensor estimation, which is traditionally inverted using multi-component sensors. However, we show that moment tensor inversion using DAS data acquired in a dual-well microseismic monitoring scenario can yield reliable estimations of all moment tensor components. Our approach is based on the adaptation of a linear waveform inversion method, utilizing 3-D anisotropic viscoelastic Green's functions, to DAS data. We estimate the moment tensors of 20 microseismic events recorded at the Hydraulic Fracturing Test Site II (HFTS2) site after relocation and velocity model updates. We observe a very good match between field and synthetic data generated using the inverted moment tensors. Accurate event locations and reasonable velocity models are important for successful moment tensor inversion. Resolution analysis of each moment tensor component, taking into account the influence of the array and the noise-dependent uncertainty, shows that the unidirectional recording of DAS is not a limiting factor for this acquisition geometry. Noise unevenly influences the uncertainty of the different moment tensor components, but most of them remain resolvable for most events even with low signal-to-noise ratio. We show that the horizontal parts of the array contribute mostly to Mxx, Myy, and Mxy, whereas the vertical parts are important for Mzz, Myz, and Mxz. The inversion and resolution analysis methodology is scale-independent and can be used to study the moment tensor resolution of alternative acquisition geometries in microseismic monitoring configurations or seismological studies.

Study on disaster-causing probability evaluation of gas pipeline in karst area.

In this paper, a disaster-causing probability evaluation method of gas pipelines in karst areas based on disaster system theory and vulnerability theory is proposed. The hazard evaluation index system of gas pipeline disaster events in karst areas is established from three aspects: the activity of disaster-prone environment, the risk of disaster factors and the vulnerability of disaster-bearing bodies. Combined with the advantages of information transmission and updating of the Bayesian network model, the hazard probability of disaster events is gradually calculated from a multi-level perspective. Based on the structural reliability theory, the vulnerability evaluation function of gas pipelines in karst areas is established by considering the interaction between disaster hazard intensity and disaster resistance ability. Meanwhile, the design checkpoint method and finite element simulation are used to calculate the vulnerability probability. Finally, a new disaster-causing probability evaluation approach for gas pipelines in karst areas is formed, which comprehensively considers the event hazard and the pipeline vulnerability. The work presented in this paper can provide a reference for the safety management and accident prevention of gas pipelines crossing the karst areas.

FLDQN: Cooperative Multi-Agent Federated Reinforcement Learning for Solving Travel Time Minimization Problems in Dynamic Environments Using SUMO Simulation

The increasing volume of traffic has led to severe challenges, including traffic congestion, heightened energy consumption, increased air pollution, and prolonged travel times. Addressing these issues requires innovative approaches for optimizing road network utilization. While Deep Reinforcement Learning (DRL)-based methods have shown remarkable effectiveness in dynamic scenarios like traffic management, their primary focus has been on single-agent setups, limiting their applicability to real-world multi-agent systems. Managing agents and fostering collaboration in a multi-agent reinforcement learning scenario remains a challenging task. This paper introduces a cooperative multi-agent federated reinforcement learning algorithm named FLDQN to address the challenge of agent cooperation by solving travel time minimization challenges in dynamic multi-agent reinforcement learning (MARL) scenarios. FLDQN leverages federated learning to facilitate collaboration and knowledge sharing among intelligent agents, optimizing vehicle routing and reducing congestion in dynamic traffic environments. Using the SUMO simulator, multiple agents equipped with deep Q-learning models interact with their local environments, share model updates via a federated server, and collectively enhance their policies using unique local observations while benefiting from the collective experiences of other agents. Experimental evaluations demonstrate that FLDQN achieves a significant average reduction of over 34.6% in travel time compared to non-cooperative methods while simultaneously lowering the computational overhead through distributed learning. FLDQN underscores the vital impact of agent cooperation and provides an innovative solution for enabling agent cooperation in a multi-agent environment.

Improving the Reliability of, and Confidence in, DFT Functional Benchmarking through Active Learning.

Validating the performance of exchange-correlation functionals is vital to ensure the reliability of density functional theory (DFT) calculations. Typically, these validations involve benchmarking data sets. Currently, such data sets are usually assembled in an unprincipled manner, suffering from uncontrolled chemical bias, and limiting the transferability of benchmarking results to a broader chemical space. In this work, a data-efficient solution based on active learning is explored to address this issue. Focusing─as a proof of principle─on pericyclic reactions, we start from the BH9 benchmarking data set and design a chemical reaction space around this initial data set by combinatorially combining reaction templates and substituents. Next, a surrogate model is trained to predict the standard deviation of the activation energies computed across a selection of 20 distinct DFT functionals. With this model, the designed chemical reaction space is explored, enabling the identification of challenging regions, i.e., regions with large DFT functional divergence, for which representative reactions are subsequently acquired as additional training points. Remarkably, it turns out that the function mapping the molecular structure to functional divergence is readily learnable; convergence is reached upon the acquisition of fewer than 100 reactions. With our final updated model, a more challenging─and arguably more representative─pericyclic benchmarking data set is curated, and we demonstrate that the functional performance has changed significantly compared to the original BH9 subset.

ELM‐Wet: Inclusion of a Wet‐Landunit With Sub‐Grid Representation of Eco‐Hydrological Patches and Hydrological Forcing Improves Methane Emission Estimations in the E3SM Land Model (ELM)

AbstractWetlands are the largest emitters of biogenic methane (CH4) and represent the highest source of uncertainty in global CH4 budgets. Here, we aim to improve the realism of wetland representation in the U.S. Department of Energy's Exascale Earth System Model land surface model, ELM, thereby reducing uncertainty of CH4 flux predictions. We develop an updated version, ELM‐Wet, where we activate a separate landunit for wetlands that handles multiple wetland‐specific eco‐hydrological patch functional types. We introduce more realistic hydrological forcing through prescribing site‐level constraints on surface water elevation, which allows resolving different sustained inundation depth for different patches, and if data exists, prescribing inundation depth. We modified the calculation of aerenchyma transport diffusivity based on observed conductance per leaf area for different vegetation types. We use Bayesian Optimization to parameterize CO2 and CH4 fluxes in the developed wet‐landunit. Site‐level simulations of a coastal non‐tidal freshwater wetland in Louisiana were performed with the updated model. Eddy covariance observations of CO2 and CH4 fluxes from 2012 to 2013 were used to train the model and data from 2021 were used for validation. Patch‐specific chamber flux observations and observations of CH4 concentration profiles in the soil porewater from 2021 were used for evaluation of the model performance. Our results show that ELM‐Wet reduces the model's CH4 emission root mean squared error by up to 33% and is able to represent inter‐daily CO2 and CH4 flux variability across the wetland's eco‐hydrological patches, including during periods of extreme dry or wet conditions.

Using mode shape residuals for model updating of a nonlinear structure featuring 1:1 internal resonance

Internal resonance isa unique phenomenon in nonlinear structures. It can lead to pronounced energy transfer between vibrating modes, causing unexpected stress concentrations in local areas and posing a risk to structural integrity. This paper presents a general model updating procedure for nonlinear structures with such internal resonances using experimentally measured data. The test involves using multiple amplitudes of sinusoidal excitations to drive the structure to several required vibration amplitudes, enabling the underlying linear dynamics, uncoupled nonlinear dynamics, and internal resonances to be activated individually. The underlying linear system, uncoupled nonlinear terms, and cross-coupling terms in the dynamic equation are updated sequentially, with updating the cross-coupling terms being the most challenging task. Unlike conventional practices that heuristically pre-assume a model for the cross-coupling force terms, this paper minimises their number by using the coupling potential energies and identifies their type via an interaction map. A novel mode shape residual is then proposed to tune the coefficients of the cross-coupling terms, which allows energy transfer between coupled modes to be captured quantitatively. A diesis-like structure featuring 1:1 bending-torsion coupling was experimentally investigated using multi-level stepped-sine excitations to demonstrate the proposed model updating procedure. The vibrations of its first resonance veered from a bending-dominant motion to a bending-torsion coexisting motion as the driving forces increased, indicating strong amplitude-dependent mode shapes on the occurrence of internal resonance. The experimental data also showed that the softening and hardening trend of the torsional mode was dependent on the vibration amplitude of the bending mode, a peculiar vibration phenomenon seldom reported for realistic structures in the literature. Using the proposed model updating procedure, a two-mode coupled nonlinear model was identified from the frequency response functions and validated by comparing the backbone curves and amplitude-dependent mode shapes. Results showed that the updated model could predict the resonant vibration patterns of the structure away from and around its internal resonance with sufficient accuracy. It was also revealed that the softening and hardening behaviours of the torsional mode were attributed to different branches of Nonlinear Normal Modes (NNMs).

Evolving the Control-Chaos Continuum: Part 1 - Translating Knowledge to Enhance On-Pitch Rehabilitation.

BACKGROUND: On-pitch rehabilitation is a crucial part of returning to sport after injury in elite soccer. The control-chaos continuum (CCC) initially offered a framework for practitioners to plan on-pitch rehabilitation, focusing on physical preparation and sport specificity. However, our experiences with the CCC, combined with recent research in injury neurophysiology, point to a need for an updated model that integrates practice design and physical-cognitive interactions. CLINICAL QUESTION: What are the insights from injury neurophysiology, soccer performance, and coaching science needed to update the CCC and improve the planning, delivery, and progression of on-pitch rehabilitation in elite soccer? KEY RESULTS: Drawing on extensive experience in elite sport, we explain how recent research on neurophysiological recovery from injury, game models, and practice design has been applied to update the CCC and evolve the existing framework. CLINICAL APPLICATION: The evolution of the CCC expands on the original model to enhance planning, delivery, and progression of on-pitch rehabilitation. The updated framework incorporates elements of visual cognition, attentional challenges, decision-making, and progressive representation of the game model to enhance sport-specific preparation for returning to sport. J Orthop Sports Phys Ther 2025;55(2):1-11. Epub 3 January 2025. doi:10.2519/jospt.2025.13158.

Extended two-tailed Lindley distribution: An updated model based on the Lindley distribution

Extended two-tailed Lindley distribution: An updated model based on the Lindley distribution

Zonotopic set-membership state estimation for nonlinear systems based on the deep Koopman operator

In this study, a novel nonlinear zonotopic set-membership estimation (SME) approach is proposed based on the Koopman operator and the deep neural networks (DNNs). The core concept involves transforming the original nonlinear system into a higher-dimensional linear system, enabling the development of linear SME algorithms. To address noise in real-world data, we propose a three-step offline training strategy for computing the linear approximation of nonlinear dynamics. Building on the deep-Koopman-based linearized system, we construct a reduced-order filter structure that directly estimates the original states and then computes the lifted states for improved observability. We derive two optimal gains based on different size criteria and present a data-driven nonlinear mapping approach that ensures a closed-form solution for the algorithm. Additionally, we introduce a local model update method that refines estimation accuracy using the set of state estimates obtained from SME. The effectiveness and superiority of our proposed methods is demonstrated through two illustrative examples.

Particle Filter Tracking System Based on Digital Zoom and Regional Image Measure

To address the challenges of low accuracy and the difficulty in balancing a large field of view and long distance when tracking high-speed moving targets with a single sensor, an ROI adaptive digital zoom tracking method is proposed. In this paper, we discuss the impact of ROI on image processing and describe the design of the ROI adaptive digital zoom tracking system. Additionally, we construct an adaptive ROI update model based on normalized target information. To capture target changes effectively, we introduce the multi-scale regional measure and propose an improved particle filter algorithm, referred to as the improved multi-scale regional measure resampling particle filter (IMR-PF). This method enables high temporal resolution processing efficiency within a high-resolution large field of view, which is particularly beneficial for high-resolution videos. The IMR-PF can maintain high temporal resolution within a wide field of view with high resolution. Simulation results demonstrate that the improved target tracking method effectively improves tracking robustness to target motion changes and reduces the tracking center error by 20%, as compared to other state-of-the-art methods. The IMR-PF still maintains good performance even when confronted with various interference factors and in real-world scenario applications.

Assessing fetal radiation dose from iodine-125 seeds in pregnant breast cancer patients: an updated model

Objective.The treatment of breast cancer during pregnancy requires careful consideration of consequences for both maternal and fetal health. In non-pregnant patients, the use of radioactive iodine-125 (125I)-seeds is standard practice for localising non-palpable breast tumors before breast-conserving surgery. However, the use of125I-seeds in pregnant patients has been avoided due to concerns about fetal radiation exposure.Approach.In this study a mathematical model was developed to estimate the fetal absorbed dose based on several factors: the radioactivity of the125I-seed, the duration of implantation, and the distance between the125I-seed and fetus as a function of maternal anatomy, gestational age, and fetal development. Three scenarios, representing a range of maternal and fetal anatomy, were evaluated, including a worst-case scenario from a radiation safety perspective.Main results.The results show that the fetal absorbed dose varies across the three scenarios, with ranges of 0.0-0.4 mGy, 0.0-1.0 mGy, and 0.0-1.6 mGy, depending on when the125I-seed was implanted and when it was removed. These dose ranges are similar to conventional diagnostic x-ray scans. The maximum calculated absorbed dose (1.6 mGy) is unlikely to be reached in practice and is well below the 100 mGy threshold associated with possible fetal malformations. The associated theoretical cancer risk increase (0.016%) is minimal.Significance.The use of125I-seeds as localisation method of breast tumors in pregnant patients results in low fetal radiation doses and should not be avoided due to dose concerns.

Causes of Stability in Dynamic Coalition Formation

We study the formation of stable outcomes via simple dynamics in cardinal hedonic games, where the valuations of agents change over time depending on the history of the coalition formation process. Specifically, we analyze situations where members of a coalition decrease their valuation for a leaving agent ( resentment ) or increase their valuation for a joining agent ( appreciation ). We show a series of convergence results for dynamics for resentful or appreciative agents which do not hold for classic dynamics. In particular, resentment turns out to be a strong stability-driving force. We complement our theoretical analysis with simulations that shed some light on the average running time of the dynamics and on the structure of the produced outcomes. From an algorithmic perspective, we obtain general hardness results for determining the fastest convergence time, results which also carry over to classic dynamics under static valuation functions. Finally, we explore a related model for preference updates where resentment is expressed by the deviator. We find a more nuanced picture, but still a broad possibility of convergence.

Federated Learning for Privacy-Friendly Health Apps: A Case Study on Ovulation Tracking

In an era of increasing reliance on digital health solutions, safeguarding user privacy has emerged as a paramount concern. Health applications often need to balance advanced AI functionalities with sufficient privacy measures to ensure user engagement. This paper presents the architecture of FLORA, a privacy-first ovulation-tracking application that leverages federated learning (FL), privacy-enhancing technologies (PETs), and blockchain to protect user data while delivering accurate and personalized health insights. Unlike conventional centralized systems, FLORA ensures that sensitive information remains on users’ devices, with predictive algorithms powered by local computations. Blockchain technology provides immutable consent tracking and model update transparency, further improving user trust. In addition, FLORA’s design incentivizes participation through a token-based reward system, fostering collaborative data contributions. This work illustrates how the integration of cutting-edge technologies creates a secure, scalable, and user-centric health application, setting a new standard for privacy-preserving digital health platforms.

FL-TENB4: A Federated-Learning-Enhanced Tiny EfficientNetB4-Lite Approach for Deepfake Detection in CCTV Environments

The widespread deployment of CCTV systems has significantly enhanced surveillance and public safety across various environments. However, the emergence of deepfake technology poses serious challenges by enabling malicious manipulation of video footage, compromising the reliability of CCTV systems for evidence collection and privacy protection. Existing deepfake detection solutions often suffer from high computational overhead and are unsuitable for real-time deployment on resource-constrained CCTV cameras. This paper proposes FL-TENB4, a Federated-Learning-enhanced Tiny EfficientNetB4-Lite framework for deepfake detection in CCTV environments. The proposed architecture integrates Tiny Machine Learning (TinyML) techniques with EfficientNetB4-Lite, a lightweight convolutional neural network optimized for edge devices, and employs a Federated Learning (FL) approach for collaborative model updates. The TinyML-based local model ensures real-time deepfake detection with minimal latency, while FL enables privacy-preserving training by aggregating model updates without transferring sensitive video data to centralized servers. The effectiveness of the proposed system is validated using the FaceForensics++ dataset under resource-constrained conditions. Experimental results demonstrate that FL-TENB4 achieves high detection accuracy, reduced model size, and low inference latency, making it highly suitable for real-world CCTV environments.

The Derivation of Acute and Chronic Environmental Quality Standards for Nickel in European Surface Waters: A Demonstrable need to follow scientific evidence.

Environmental Quality Standards (EQS) derived under the European Water Framework Directive are legally binding and enshrined in individual European Member State Country national legislation. These EQS are derived following well-established guidance documents. In 2013, EQS for nickel were derived for freshwaters to be protective against long and short-term exposures, at 4 and 34 µg L-1, respectively. The value for long-term exposures uses chronic ecotoxicity data and accounts for bioavailability, whereas the short-term value uses acute data and does not account for bioavailability. In 2022, the European Commission revised these values as part of the on-going legislative process. Despite an increase in available data for both chronic and acute ecotoxicity endpoints, the update and development of chronic and acute Biotic Ligand Models (BLMs) published in peer-reviewed literature, and the accessibility of vastly more monitoring data (used in the European EQS derivation), the values for the nickel EQS were reduced by increasing the assessment factors, to account for increases in apparent uncertainties. The Commission's 2022 derivation failed to consider additional chronic data for more than 20 species as well as the updated and new acute and chronic BLMs. As a result, the derived nickel EQS is limited in its applicability and relevance to European freshwater ecosystems, as illustrated in practice by the observation that monitoring sites can comply with the chronic EQS but fail the acute EQS. Here, we provide an explanation as to why this has occurred and detail what it means for the risk assessment of nickel in European Member State freshwaters. Finally, we outline a path forward, that should be relevant for any risk-based and evidence-driven regulatory framework and acknowledging that political decisions are part of the process, but that these should be separate and after scientific aspects are undertaken.

Theoretical Kinetic Study of NH2 Reactions With Dimethyl Ether and Diethyl Ether: Implications for Kinetic Modeling

ABSTRACTAmmonia (NH3) and hydrogen (H2) have emerged as promising carbon‐free fuels to help mitigate global warming by reducing greenhouse gas emissions. Our ongoing research currently focuses on understanding the combustion characteristics of NH3 blends with oxygenates and hydrocarbons, uncovering the critical role of carbon–nitrogen cross‐reactions in accurately modeling their combustion behavior. Amino (NH2) radicals, which are abundant in ammonia and nitrogen‐rich environments, strongly influence the low‐temperature reactivity of NH3‐hydrocarbon/oxygenate mixtures, affecting overall reactivity and emission characteristics. Recognizing the importance of NH2 radicals, we investigated the reaction kinetics of NH2 with dimethyl ether (DME, CH3OCH3) and diethyl ether (DEE, CH3CH2OCH2CH3) using appropriate high‐level ab initio and statistical rate theory methods. We computed the potential energy profiles at the CCSD(T)/cc‐pV(T, Q)Z//M06‐2X/aug‐cc‐pVTZ level of theory, analyzing the reactivity of NH2 radicals at various C─H sites of these diethers. Incorporating these newly derived rate parameters, our updated kinetic model successfully captures previous experimental data, addressing the modeling challenges encountered in our earlier studies. Our findings, including insights into the impact of NH2 radicals, contribute to an understanding of ammonia combustion and its potential in achieving carbon‐neutral energy systems.