Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Statistical Testing Approach”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 469 records · Page 26
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

The Dark Energy Camera All Data Everywhere cosmic shear project V: Constraints on cosmology and astrophysics from 270 million galaxies across 13,000 deg$^2$ of the sky

We present constraints on models of cosmology and astrophysics using cosmic shear data vectors from three datasets: the northern and southern Galactic cap of the Dark Energy Camera All Data Everywhere (DECADE) project, and the Dark Energy Survey (DES) Year 3. These data vectors combined consist of 270 million galaxies spread across 13,000 ${\rm deg}^2$ of the sky. We first extract constraints for $Λ$CDM cosmology and find $S_8= 0.805^{+0.019}_{-0.019}$ and $Ω_{\rm m} = 0.262^{+0.023}_{-0.036}$, which is consistent within $1.9 σ$ of constraints from the Planck satellite. Extending our analysis to dynamical dark energy models shows that lensing provides some (but still minor) improvements to existing constraints from supernovae and baryon acoustic oscillations. Finally, we study six different models for the impact of baryons on the matter power spectrum. We show the different models provide consistent constraints on baryon suppression, and associated cosmology, once the astrophysical priors are sufficiently wide. Current scale-cut approaches for mitigating baryon contamination result in a residual bias of $\approx 0.3σ$ in the $S_8, Ω_{\rm m}$ posterior. Using all scales with dedicated baryon modeling leads to negligible improvement as the new information is used solely to self-calibrate the baryon model on small scales. Additional non-lensing datasets, and/or calibrations of the baryon model, will be required to access the full statistical power of the lensing measurements. The combined dataset in this work represents the largest lensing dataset to date (most galaxies, largest area) and provides an apt testing ground for analyses of upcoming datasets from Stage IV surveys. The DECADE shear catalogs, data vectors, and likelihoods are made publicly available.

Anbajagane, D. [Chicago U., Astron. Astrophys. Ctr↗
DOE OSTI2023

Predicting rare events using neural networks and short-trajectory data

Estimating the likelihood, timing, and nature of events is a major goal of modeling stochastic dynamical systems. When the event is rare in comparison with the timescales of simulation and/or measurement needed to resolve the elemental dynamics, accurate prediction from direct observations becomes challenging. In such cases a more effective approach is to cast statistics of interest as solutions to Feynman-Kac equations (partial differential equations). Here, we develop an approach to solve Feynman-Kac equations by training neural networks on short-trajectory data. Our approach is based on a Markov approximation but otherwise avoids assumptions about the underlying model and dynamics. This makes it applicable to treating complex computational models and observational data. Additionally, we illustrate the advantages of our method using a low-dimensional model that facilitates visualization, and this analysis motivates an adaptive sampling strategy that allows on-the-fly identification of and addition of data to regions important for predicting the statistics of interest. Finally, we demonstrate that we can compute accurate statistics for a 75-dimensional model of sudden stratospheric warming. This system provides a stringent test bed for our method.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Data-driven Mori–Zwanzig modeling of Lagrangian particle dynamics in turbulent flows

The dynamics of Lagrangian particles in turbulence play a crucial role in mixing, transport, and dispersion in complex flows. Their trajectories exhibit highly nontrivial statistical behavior, motivating the development of surrogate models that can reproduce these trajectories without incurring the high computational cost of direct numerical simulations of the full Eulerian field. This task is particularly challenging because reduced-order models typically lack access to the full set of interactions with the underlying turbulent field. Novel data-driven machine learning techniques can be powerful in capturing and reproducing complex statistics of the reduced-order/surrogate dynamics. In this work, we show how one can learn a surrogate dynamical system that is able to evolve a turbulent Lagrangian trajectory in a way that is point-wise accurate for short-time predictions (with respect to Kolmogorov time) and stable and statistically accurate at long times. This approach is based on the Mori–Zwanzig formalism, which prescribes a mathematical decomposition of the full dynamical system into resolved dynamics that depend on the current state and the past history of a reduced set of observables, and the unresolved orthogonal dynamics due to unresolved degrees of freedom of the initial state. We show how by training this reduced order model on a point-wise error metric on short time-prediction, we are able to correctly learn the dynamics of Lagrangian turbulence, such that also the long-time statistical behavior is stably recovered at test time. This opens up a range of applications, for example, for the control of active Lagrangian agents in turbulence.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

A comparison of model validation approaches for echo state networks using climate model replicates

As global temperatures continue to rise, climate mitigation strategies such as stratospheric aerosol injections (SAI) are increasingly discussed, but the downstream effects of these strategies are not well understood. As such, there is interest in developing statistical methods to quantify the evolution of climate variable relationships during the time period surrounding an SAI. Feature importance applied to echo state network (ESN) models has been proposed as a way to understand the effects of SAI using a data-driven model. This approach depends on the ESN fitting the data well. If not, the feature importance may place importance on features that are not representative of the underlying relationships. Typically, time series prediction models such as ESNs are assessed using out-of-sample performance metrics that divide the times series into separate training and testing sets. However, this model assessment approach is geared towards forecasting applications and not scenarios such as the motivating SAI example where the objective is using a data driven model to capture variable relationships. Here, in this paper, we demonstrate a novel use of climate model replicates to investigate the applicability of the commonly used repeated hold-out model assessment approach for the SAI application. Simulations of an SAI are generated using a simplified climate model, and different initialization conditions are used to provide independent training and testing sets containing the same SAI event. The climate model replicates enable out-of-sample measures of model performance, which are compared to the single time series hold-out validation approach. For our case study, it is found that the repeated hold-out sample performance is comparable, but conservative, to the replicate out-of-sample performance when the training set contains enough time after the aerosol injection.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

OperonSEQer: A set of machine-learning algorithms with threshold voting for detection of operon pairs using short-read RNA-sequencing data

Operon prediction in prokaryotes is critical not only for understanding the regulation of endogenous gene expression, but also for exogenous targeting of genes using newly developed tools such as CRISPR-based gene modulation. A number of methods have used transcriptomics data to predict operons, based on the premise that contiguous genes in an operon will be expressed at similar levels. While promising results have been observed using these methods, most of them do not address uncertainty caused by technical variability between experiments, which is especially relevant when the amount of data available is small. In addition, many existing methods do not provide the flexibility to determine the stringency with which genes should be evaluated for being in an operon pair. We present OperonSEQer, a set of machine learning algorithms that uses the statistic and p-value from a non-parametric analysis of variance test (Kruskal-Wallis) to determine the likelihood that two adjacent genes are expressed from the same RNA molecule. We implement a voting system to allow users to choose the stringency of operon calls depending on whether your priority is high recall or high specificity. In addition, we provide the code so that users can retrain the algorithm and re-establish hyperparameters based on any data they choose, allowing for this method to be expanded as additional data is generated. We show that our approach detects operon pairs that are missed by current methods by comparing our predictions to publicly available long-read sequencing data. OperonSEQer therefore improves on existing methods in terms of accuracy, flexibility, and adaptability.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

Impact of genotype‐calling methodologies on genome‐wide association and genomic prediction in polyploids

Abstract Discovery and analysis of genetic variants underlying agriculturally important traits are key to molecular breeding of crops. Reduced representation approaches have provided cost‐efficient genotyping using next‐generation sequencing. However, accurate genotype calling from next‐generation sequencing data is challenging, particularly in polyploid species due to their genome complexity. Recently developed Bayesian statistical methods implemented in available software packages, polyRAD, EBG, and updog, incorporate error rates and population parameters to accurately estimate allelic dosage across any ploidy. We used empirical and simulated data to evaluate the three Bayesian algorithms and demonstrated their impact on the power of genome‐wide association study (GWAS) analysis and the accuracy of genomic prediction. We further incorporated uncertainty in allelic dosage estimation by testing continuous genotype calls and comparing their performance to discrete genotypes in GWAS and genomic prediction. We tested the genotype‐calling methods using data from two autotetraploid species, Miscanthus sacchariflorus and Vaccinium corymbosum , and performed GWAS and genomic prediction. In the empirical study, the tested Bayesian genotype‐calling algorithms differed in their downstream effects on GWAS and genomic prediction, with some showing advantages over others. Through subsequent simulation studies, we observed that at low read depth, polyRAD was advantageous in its effect on GWAS power and limit of false positives. Additionally, we found that continuous genotypes increased the accuracy of genomic prediction, by reducing genotyping error, particularly at low sequencing depth. Our results indicate that by using the Bayesian algorithm implemented in polyRAD and continuous genotypes, we can accurately and cost‐efficiently implement GWAS and genomic prediction in polyploid crops.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2015

Rendering the Topological Spines

Many tools to analyze and represent high dimensional data already exits yet most of them are not flexible, informative and intuitive enough to help the scientists make the corresponding analysis and predictions, understand the structure and complexity of scientific data, get a complete picture of it and explore a greater number of hypotheses. With this in mind, N-Dimensional Data Analysis and Visualization (ND²AV) is being developed to serve as an interactive visual analysis platform with the purpose of coupling together a number of these existing tools that range from statistics, machine learning, and data mining, with new techniques, in particular with new visualization approaches. My task is to create the rendering and implementation of a new concept called topological spines in order to extend ND²AV's scope. Other existing visualization tools create a representation preserving either the topological properties or the structural (geometric) ones because it is challenging to preserve them both simultaneously. Overcoming such challenge by creating a balance in between them, the topological spines are introduced as a new approach that aims to preserve them both. Its render using OpenGL and C++ and is currently being tested to further on be implemented on ND²AV. In this paper I will present what are the Topological Spines and how they are rendered.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Federated Learning for Efficient Condition Monitoring and Anomaly Detection in Industrial Cyber-Physical Systems

Detecting and localizing anomalies in cyber-physical systems (CPS) has become increasingly challenging as systems grow in complexity, particularly due to varying sensor reliability and node failures in distributed environments. While federated learning (FL) offers a foundation for distributed model training, existing approaches lack mechanisms to handle these CPS-specific challenges. This paper presents an enhanced FL framework that introduces three key innovations: adaptive model aggregation based on sensor reliability, dynamic node selection for resource optimization, and Weibull-based checkpointing for fault tolerance. Our framework enables reliable condition monitoring while addressing the computational and reliability challenges of industrial CPS deployments. Experiments on NASA Bearing and Hydraulic System Datasets demonstrate superior performance over state-of-the-art FL methods, achieving 99.5% AUC-ROC in anomaly detection and maintaining accuracy under node failures. Statistical validation using Mann-Whitney (U) test confirms significant improvements (p < 0.05) in both detection accuracy and computational efficiency across diverse operational scenarios.1

Marfo, William [University of Texas at El Paso,Dep↗
DOE OSTI2023

Compilation of a Solar Mirror Materials Database and an Analysis of Natural and Accelerated Mirror Exposure and Degradation

The National Renewable Energy Laboratory (NREL) has been conducting exposure experiments on solar reflectors for over four decades. Thousands of mirror samples from over one hundred suppliers have been exposed to and monitored in a range of relevant environments. These test conditions include outdoor test settings and several controlled laboratory environments. These samples have been rigorously individually characterized using a series of reflectance measurements, visual inspections, and in some cases, in-depth composition analysis to identify degradation modes, reflectance losses, and other mirror properties integral to understanding the solar reflector's life cycle. Here, this paper compiles the decades of measurement data into a concise statistical analysis. It includes exposure and degradation data for numerous reflector types, including secondary-surface reflector permutations of polymer and glass superstrates with silver and aluminum reflectors as well as front-surface reflectors. The results herein are intended to analyze environmental stressors and degradation trends among various historical and state-of-the-art solar reflectors. It may be used to support solar reflector design, effective testing methodology, and inform manufacturing decisions moving forward. Presented are the results of the compiled database and an initial analysis for degradation rate modeling using full-spectrum and wavelength-dependent approaches. The database is a growing resource hosted on a live, publicly accessible website. In conjunction with the analysis presented here, it provides a valuable resource to the solar reflector manufacturing and testing industry.

14 SOLAR ENERGY↗
DOE OSTI2016

WE-FG-BRA-08: Potential Role of the Glycolytic Oscillator in Acute Hypoxia in Tumors

Purpose: Oscillatory dynamics in acute hypoxia have been observed, but poorly understood. They have mostly been attributed to vascular perturbations, but no link has yet been made to metabolic causes. We set out to determine the fundamental frequencies and test for coherence in tumor oxygen dynamics and spatial properties. Methods: Severe combined immunodeficient (SCID) mice were inoculated onto bilateral flanks with human derived head and neck carcinoma (UW-SCC22) cell line xenografts. Oxygen dynamics were monitored in the tumor every minute for an hour using three modalities: blood oxygen level dependent - magnetic resonance imaging (BOLD-MRI), hemoglobin oxygen saturation photoacoustic, and locally manufactured optical probes for spectral fitting. A statistical test was used to separate fluctuating from non-fluctuating voxels and pixels in BOLD-MRI and photoacoustic data respectively. The power spectrum density (PSD) and the autocorrelation functions were calculated for the time series of each voxel, pixel and region, of the BOLD-MRI, photoacoustic or fiber optic data respectively. Results: Using all three techniques, intermittent oxygen dynamics with both coherent and incoherent signatures was observed in the tumors. Upon averaging the PSDs of fluctuating voxels and pixels, it was found that these oscillations occurred with periods of minutes to tens of minutes from all three approaches. Observations from the BOLD-MRI and photoacoustic data showed that clusters of voxels oscillated in a synchronized manner. Conclusion: We were able to use three different modalities to show that fluctuation in tumor oxygen is both coherent and incoherent, with periods of minutes to tens of minutes. These periods are very similar to those from the well-established metabolic, non-linear biomechanical phenomenon called the glycolytic oscillator. This may provide an additional explanation to the cause of cyclic hypoxia. Such dynamics could have profound implications in hypofractionated radiotherapy regiments and could help guide treatment and make it more patient specific. The authors would like to thank the University of Wisconsin Carbone Cancer Center (UWCCC) for the funds to complete this project. This work is also supported in part by NIH/NCI P30 CA014520- UW Comprehensive Cancer Center Support”.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2021

Refinement of Salt Dissolution Inhibitor Requirements (Final Report)

At Savannah River Site (SRS), High-Level Waste is stored in below-grade carbon steel tanks. This waste in part consists of sludge, salt cake, and/or supernate. Preparation of this waste for future processing involves dissolution of the salt cake layer. The salt dissolution process can create conditions that leave the carbon steel tanks susceptible to localized corrosion. The salt to be dissolved contains high concentrations of nitrate, that once released, create an environment that may be conducive to pitting corrosion and/or stress corrosion cracking (SCC) of carbon steel. The salt dissolution process also liberates interstitial liquid trapped between the salt crystals. This liquid is initially high in nitrite and hydroxide concentration. High pH and greater ratios of nitrite to nitrate act as inhibitors to minimize corrosion of carbon steel in high nitrate environments. However, as dissolution proceeds, the concentration of nitrate will increase, while the hydroxide and nitrite concentration of the interstitial liquid will deplete and become insufficient to prevent the onset of corrosion attack. Tank blending and addition of inhibitors are used to ensure adequate concentrations of hydroxide and nitrite. However, this is not desirable during salt dissolution as it can reduce process efficiency and increase the amount of waste that needs processing. It has been proposed that these corrosion control limits be revisited to evaluate the corrosion susceptibility of carbon steel in environments that more closely resemble current operating conditions at SRS. An experimental matrix was designed to evaluate the use of the pitting factor for supernate chemistries characteristic to SRS, particularly during the salt dissolution process. Two electrochemical methods were identified to determine the susceptibility of A537 and A285 low-carbon steels to pitting corrosion with this chemistry envelope at temperatures up to 75 °C. The predominant electrochemical test method was Cyclic Potentiodynamic Polarization (CPP) studies. Through CPP, the pitting factor was used, based on Hanford Site corrosion studies, to accurately identify pitting susceptibility within the compositional range studied with some conservatism. Additionally, sulfate was determined to have no statistically significant influence, at concentrations up to 0.6 M, on pitting behavior in more concentrated solutions where other aggressive species govern pitting susceptibility. Where CPP was inconclusive, Modified ASTM G192 was successfully used to evaluate pitting susceptibility conditions and allowed for a pass/fail result to be determined. In all cases, the pitting factor was determined to be applicable to the simulants tested, with this metric accurately predicting incidences in which pitting occurred. Based upon the findings in this work, a pitting factor of 1.2 is being proposed to build in a safety factor and remain consistent with the Hanford Site approach. Additionally, a minimum pH limit of 12 is proposed to ensure carbon steel passivity and localized corrosion the primary degradation mechanism. Susceptibility to SCC was evaluated using a reduced matrix of tests at 75 °C. No failures due SCC were observed at open circuit. In addition, tests polarized anodically by 200 mV only resulted in failures for trials with pitting factors less than 0.86. However, a test with a passing condition based upon the pitting factor metric (pitting factor = 1.40) did exhibit a failure with an applied potential of +300 mV vs. OCP. This result is contrary to the prediction based upon the pitting factor, however, a polarization of 300 mV, or even 200 mV, from open circuit is substantial. The relationship between these testing parameters and service environment/conditions and the desired level of conservatism in the metric should be further evaluated in the determination of the significance of this result. While the pitting factor accurately predicted susceptibility to SCC at temperatures up to 75 °C and with positive overpotentials up to 200 mV, the relatively small sample matrix and failure of a passing pitting factor with a 300 mV polarization resulted in an inconclusive determination of whether the pitting factor may be used for predicting susceptibility to SCC at temperatures between 50 °C and 75 °C. As such, additional testing is recommended to evaluate the validity of the pitting factor for SCC susceptibility prediction at temperatures between 50 °C and 75 °C.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2021

A Nonstationary and Non-Gaussian Moving Average Model for Solar Irradiance

Historically, power has flowed from large power plants to customers. Increasing penetration of distributed energy resources such as solar power from rooftop photovoltaic has made the distribution network a two-way-street with power being generated at the customer level. The incorporation of renewables introduces additional uncertainty and variability into the power grid. Distribution network operation studies are being adapted to include renewables; however, such studies require high quality solar irradiance data that adequately reflect realistic meteorological variability. Data from satellite-based products are spatially complete, but temporally coarse, whereas solar irradiances exhibit high frequency variation at very fine timescales. We propose a new stochastic method for temporally downscaling global horizontal irradiance (GHI) to 1 min resolution, but we do not consider the spatial aspect due to limited availability of the in situ irradiance measurements. Solar irradiance's first and second-order structures vary diurnally and seasonally, and our model adapts to such nonstationarity. Empirical irradiance data exhibits highly non-Gaussian behavior; we develop a nonstationary and non-Gaussian moving average model that is shown to capture realistic solar variability at multiple timescales. We also propose a new estimation scheme based on Cholesky factors of empirical autocovariance matrices, bypassing difficult and inaccessible likelihood-based approaches. The model is demonstrated for a case study of three locations that are located in diverse climates through the United States. The model is compared against competitors from the literature and is shown to provide better uncertainty and variability quantification on testing data.

Cholesky factor↗
DOE OSTI2018

Pilot Demonstration Test Bed for ICC Desiccant Based Cooling System Final Report CRADA No. TSB-737-93

In this CRADA project, LLNL provided a field engineering, operation, and maintenance evaluation of ICC’s Desiccant Based Cooling (DBC) System by installing and activating an instrumented test bed in a real application at LLNL. LLNL was approached by two manufacturers of newly emerging heating, ventilation and air-conditioning (HVAC) equipment for technical assistance in obtaining engineering thermodynamic evaluation of their units. Despite promising improvements in technical HVAC performance and the elimination of environmentally sensitive materials (CFC’S), they experienced market resistance in the federal sector because of the need to develop a record of technical and reliability data. LLNL selected three sites as areas where the units could potentially produce significant improvements in energy consumption and healthy air conditions. This project covered the installation activation of the pilot test bed at LLNL. The collaborative effort provided statistical and requirements bases to make a valid evaluation of the units.

42 ENGINEERING↗
DOE OSTI2023

National forest timber bids and export price interlinkages in the USA: The bounds testing approach

We examine the interrelationships of national forest timber bid and log export prices for Douglas-fir (Region 6) in 2003–2021 and loblolly pine (Region 8) using time-series stationarity and bounds-testing approaches. The analysis shows the difference in time series properties across different geographies and time scales. Bid prices for these dominant species of timber in each of two Forest Service regions are found to be stationary, while log export prices for those same species appear to be non-stationary. Bounds testing conducted with autoregressive distributed lag modeling methods provides strong evidence for a cointegrating relationship between timber bid prices and log export prices in Forest Service Region 6. In Region 8, evidence for a cointegrating relationship was weak. Here, the share of salvage wood in national forest timber sales is a statistically significant control variable in all models estimated.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2026

Asymptotic inconsistency of the cumulative algorithm for laser-induced damage probability analysis

The “cumulative algorithm” is a data analysis method that has been proposed to provide an objective, nonparametric determination of laser-induced damage probability as a function of fluence from experimental data that contain both damaged sites and undamaged sites (i.e., 1-on-1 or S-on-1 testing protocols). In this work, the limitations of this approach are explored by considering the asymptotic limit of a large number of test sites. It is shown that the cumulative algorithm does not converge to the true probability distribution and significantly underestimates the damage probability near the damage onset. Here, based on the results of this work, the cumulative algorithm is not recommended for accurate estimation of damage probability.

Computational methods↗
DOE OSTI2014

An exact and efficient first passage time algorithm for reaction–diffusion processes on a 2D-lattice

We present an exact and efficient algorithm for reaction–diffusion–nucleation processes on a 2D-lattice. The algorithm makes use of first passage time (FPT) to replace the computationally intensive simulation of diffusion hops in KMC by larger jumps when particles are far away from step-edges or other particles. Our approach computes exact probability distributions of jump times and target locations in a closed-form formula, based on the eigenvectors and eigenvalues of the corresponding 1D transition matrix, maintaining atomic-scale resolution of resulting shapes of deposit islands. We have applied our method to three different test cases of electrodeposition: pure diffusional aggregation for large ranges of diffusivity rates and for simulation domain sizes of up to 4096×4096 sites, the effect of diffusivity on island shapes and sizes in combination with a KMC edge diffusion, and the calculation of an exclusion zone in front of a step-edge, confirming statistical equivalence to standard KMC simulations. The algorithm achieves significant speedup compared to standard KMC for cases where particles diffuse over long distances before nucleating with other particles or being captured by larger islands.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2015

A first-principle calculation of the XANES spectrum of Cu{sup 2+} in water

The progress in high performance computing we are witnessing today offers the possibility of accurate electron density calculations of systems in realistic physico-chemical conditions. In this paper, we present a strategy aimed at performing a first-principle computation of the low energy part of the X-ray Absorption Spectroscopy (XAS) spectrum based on the density functional theory calculation of the electronic potential. To test its effectiveness, we apply the method to the computation of the X-ray absorption near edge structure part of the XAS spectrum in the paradigmatic, but simple case of Cu{sup 2+} in water. In order to keep into account the effect of the metal site structure fluctuations in determining the experimental signal, the theoretical spectrum is evaluated as the average over the computed spectra of a statistically significant number of simulated metal site configurations. The comparison of experimental data with theoretical calculations suggests that Cu{sup 2+} lives preferentially in a square-pyramidal geometry. The remarkable success of this approach in the interpretation of XAS data makes us optimistic about the possibility of extending the computational strategy we have outlined to the more interesting case of molecules of biological relevance bound to transition metal ions.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

Adaptive Client Selection in Federated Learning: A Network Anomaly Detection Use Case

Federated Learning (FL) has become a ubiquitous approach for training machine learning models on decentralized data, addressing the myriad privacy concerns inherent in traditional centralized methods. However, the efficiency of FL depends on effective client selection and robust privacy preservation mechanisms. Inadequate client selection may lead to suboptimal model performance, while insufficient privacy measures risk exposing sensitive data. This paper proposes a client selection framework for FL that integrates differential privacy and fault tolerance. Our adaptive approach dynamically adjusts the number of selected clients based on model performance and system constraints, ensuring privacy through calibrated noise addition. We evaluate our method on a network anomaly detection use case using the UNSW-NB15 and ROAD datasets. Results show up to a 7% increase in accuracy and a 25% reduction in training time compared to FedL2P. Moreover, we highlight the trade-offs between privacy budgets and model performance, with higher privacy budgets reducing noise and improving accuracy. Our fault tolerance mechanism, while causing a slight performance drop, enhances robustness to client failures. Statistical validation using Mann-Whitney U tests confirms the significance of these improvements (p < 0.05).

Marfo, William [University of Texas at El Paso,Dep↗