Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “ensemble machine learning”

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 433 records · Page 24
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

MindSynchro

This report presents the developments and results of MindSynchro project as part of DOE OE FOA 1861. DOE and Pacific Northwest National Laboratory (PNNL) have made available to FOA awardees datasets containing years of real historical data recorded from various phasor measurement units (PMUs) which are installed in three large US interconnections: Texas (IC A), Western (IC B), and Eastern (IC C). The main goal of the project, which was successfully achieved, was to develop methods for detection and identification of events which are relevant for power grid operation. Tasks performed for achieving the project goals included data exploration and pre-processing, the development and application of physics-based features, data analysis and labeling based on unsupervised learning approaches, training and testing of DSSL models for classification of events which are relevant for power grid operation, and deployment of solutions to cloud environments. The methods developed in the project can potentially provide relevant benefits to power grid asset owners/operators in general in terms of situational awareness. Two main types of outcomes can be provided by these tools: Identification of specific relevant power grid event types: Semi-supervised ML methods developed in the project can adequately employ not only the relatively scarce labeled data but also the large amount of available unlabeled data to train models for detection of specific event types. Such methods enable the application of trained models for the detection of events in a population of PMUs much larger than that associated to the labeled events. Support in data labeling / label validation: Labels are critical for training of models for identification of specific types of events. However, labeling large amounts of data is a manual and tedious process. This means that such process is error prone and is not scalable. Methods developed in the project, based on ensembles of clustering models, have been successfully employed for turning manual labeling into a scalable process. Accurate identification of specific relevant events can provide the operators with immediate situational awareness that could otherwise require hours or days of analysis from domain experts. We envision that such methods could be initially employed in support of post-mortem analysis of events and, as confidence is gained, they could be employed for online/real-time support, providing, among other benefits, insights for avoiding major events which could happen due to a combination of smaller ones. On the longer term, related methods could potentially be employed to improve protection and control.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2021

Semi-supervised Bayesian Low-shot Learning

Deep neural networks (NNs) typically outperform traditional machine learning (ML) approaches for complicated, non-linear tasks. It is expected that deep learning (DL) should offer superior performance for the important non-proliferation task of predicting explosive device configuration based upon observed optical signature, a task which human experts struggle with. However, supervised machine learning is difficult to apply in this mission space because most recorded signatures are not associated with the corresponding device description, or “truth labels.” This is challenging for NNs, which traditionally require many samples for strong performance. Semi-supervised learning (SSL), low-shot learning (LSL), and uncertainty quantification (UQ) for NNs are emerging approaches that could bridge the mission gaps of few labels and rare samples of importance. NN explainability techniques are important in gaining insight into the inferential feature importance of such a complex model. In this work, SSL, LSL, and UQ are merged into a single framework, a significant technical hurdle not previously demonstrated. Exponential Average Adversarial Training (EAAT) and Pairwise Neural Networks (PNNs) are chosen as the SSL and LSL methods of choice. Permutation feature importance (PFI) for functional data is used to provide explainability via the Variable importance Explainable Elastic Shape Analysis (VEESA) pipeline. A variety of uncertainty quantification approaches are explored: Bayesian Neural Networks (BNNs), ensemble methods, concrete dropout, and evidential deep learning. Two final approaches, one utilizing ensemble methods and one utilizing evidential learning, are constructed and compared using a well-quantified synthetic 2D dataset along with the DIRSIG Megascene.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Assessments of epistemic uncertainty using Gaussian stochastic weight averaging for fluid-flow regression

Here, we use Gaussian stochastic weight averaging (SWAG) to assess the epistemic uncertainty associated with neural-network-based function approximation relevant to fluid flows. SWAG approximates a posterior Gaussian distribution of each weight, given training data, and a constant learning rate. Having access to this distribution, it is able to create multiple models with various combinations of sampled weights, which can be used to obtain ensemble predictions. The average of such an ensemble can be regarded as the 'mean estimation', whereas its standard deviation can be used to construct 'confidence intervals', which enable us to perform uncertainty quantification (UQ) with regard to the training process of neural networks. We utilize representative neural-network-based function approximation tasks for the following cases: (i) a two-dimensional circular-cylinder wake; (ii) the DayMET dataset (maximum daily temperature in North America); (iii) a three-dimensional square-cylinder wake; and (iv) urban flow, to assess the generalizability of the present idea for a wide range of complex datasets. SWAG-based UQ can be applied regardless of the network architecture, and therefore, we demonstrate the applicability of the method for two types of neural networks: (i) global field reconstruction from sparse sensors by combining convolutional neural network (CNN) and multi-layer perceptron (MLP); and (ii) far-field state estimation from sectional data with two-dimensional CNN. We find that SWAG can obtain physically-interpretable confidence-interval estimates from the perspective of epistemic uncertainty. This capability supports its use for a wide range of problems in science and engineering.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Leveraging AI and Spatial Data to Unlock Pipeline Integrity Insights: NETL’s Advanced Infrastructure Integrity Model (AIIM)

Maintaining the integrity of natural gas infrastructure plays a critical role in ensuring energy security. Robust, data-driven foundational AI models for pipeline integrity can help address risk management and mitigation issues. Trusted foundational models can help with industry adoption and accelerate innovation by enhancing integrity predictions, reduce costs, and informing infrastructure build-out. The AIIM dashboard was released in 2022 and utilizes multi-ML models for ensemble-type insights. It was expanded to include analytics on reported incidents. It was developed as an ESRI Dashboard to support data visualization & interrogation and contains pipeline data and model results.

Advanced Infrastructure Integrity Model (AIIM)↗
DOE OSTI2026

Inverse Mapping of the Collision Kernel and Wall Flux Scaling in a Tall Convection‐Cloud Chamber Using Local Sensors and Knowledge‐Informed Deep Learning

Droplet collision–coalescence is a crucial process in cloud physics, but accurately representing this process under different dynamical conditions remains challenging. A proposed future convective‐cloud chamber aims to investigate this key process, but the method for observing it remains unclear, even though it is theoretically established that collision‐coalescence will occur. This study serves as a proof‐of‐concept demonstration of how knowledge‐informed deep learning, combined with measurement data from local sensors in the chamber, can be used to estimate the collision kernels, which determine how the droplet size distribution evolves during collision‐coalescence. In addition to estimating the collision kernel, we also address wall fluxes, another uncertain but important process that acts as a source of heat and moisture in the chamber. Ensemble runs of large‐eddy simulations are conducted by scaling the wall fluxes and the collision kernel, while the measured flow and cloud properties are used as inputs for a neural network. Results indicate that this approach successfully maps the scaling of wall fluxes and the collision kernel with biases of approximately 1% or less relative to the range of the target data. This proof‐of‐concept lays the groundwork for future applications; when the real measurements are available, real sensor data combined with the trained model presented in this work will enable estimation of the actual wall fluxes and collision kernel.

cloud chamber↗
DOE OSTI2022

Enhanced analysis of experimental x-ray spectra through deep learning

X-ray spectroscopic data from high-energy-density laser-produced plasmas has long required thorough, time-consuming analysis to extract meaningful source conditions. There are often confounding factors due to rapidly evolving states and finite spatial gradients (e.g., the existence of multi-temperature, multi-density, multi-ionization states, etc.) that make spectral measurements and analysis difficult. Here, in this paper, we demonstrate how deep learning can be applied to enhance x-ray spectral data analysis in both speed and intricacy. Neural networks (NNs) are trained on ensemble atomic physics simulations so that they can subsequently construct a model capable of extracting plasma parameters directly from experimental spectra. Through deep learning, the models can extract temperature distributions as opposed to single or dual temperature/density fits from standard trial-and-error atomic modeling at a significantly reduced computational cost compared to traditional trial-and-error methods. These NNs are envisioned to be deployed with high repetition rate x-ray spectrometers in order to provide detailed real-time analysis of experimental spectra.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2025

Improving Trustworthiness of Data-Driven Power Grid Contingency Analysis With Bayesian Residual Graph Neural Networks

The evolving energy landscape requires novel tools to efficiently perform contingency analysis and reliability assessment of power grids, potentially in real-time. The high computational cost of traditional power flow solvers limits their applicability in practice. Machine learning (ML) surrogates such as deep neural networks (NNs) accelerate power flow solvers computations, enabling high-order contingency analysis and real-time decision-making by learning highly nonlinear functions and integrating grid topology via graph architectures. However, (graph) NNs lack predictive power away from training data and do not provide predictive confidence estimates. Here, we present a Bayesian residual graph NN that integrates knowledge from low-fidelity data via residual training and embeds granular quantification of uncertainties, improving trustworthiness critical for high-consequence decision-making. Applying Bayesian concepts to NNs is challenging due to the high-dimensionality of both the parameter space, complicating derivation of a meaningful prior, and the output space in large grid systems, requiring enhanced techniques to assess the predicted high-dimensional uncertainties. Our contributions include: (1) Deriving a prior for fully connected and graph NNs that leverages low-fidelity data to guide mean predictions and appropriately control prior predictive uncertainty. (2) Integrating this prior within an ensembling with anchoring scheme for efficient approximate posterior inference. (3) Deriving enhanced metrics to assess accuracy of both the mean and uncertainty predictions in high dimensions, appropriately accounting for correlations propagated through graph layers. The resulting Bayesian residual graph NN is tested on a contingency analysis task for 14-bus and 118-bus grids.

24 - POWER TRANSMISSION AND DISTRIBUTION↗
NASA NTRS2019

Exploring the Utility of Machine Learning-Based Passive Microwave Brightness Temperature Data Assimilation over Terrestrial Snow in High Mountain Asia

This study explores the use of a support vector machine (SVM) as the observation operator within a passive microwave brightness temperature data assimilation framework (herein SVM-DA) to enhance the characterization of snow water equivalent (SWE) over High Mountain Asia (HMA). A series of synthetic twin experiments were conducted with the NASA Land Information System (LIS) at a number of locations across HMA. Overall, the SVM-DA framework is effective at improving SWE estimates (~70% reduction in RMSE relative to the Open Loop) for SWE depths less than 200 mm during dry snowpack conditions. The SVM-DA framework also improves SWE estimates in deep, wet snow (~45% reduction in RMSE) when snow liquid water is well estimated by the land surface model, but can lead to model degradation when snow liquid water estimates diverge from values used during SVM training. In particular, two key challenges of using the SVM-DA framework were observed over deep, wet snowpacks. First, variations in snow liquid water content dominate the brightness temperature spectral difference (TB) signal associated with emission from a wet snowpack, which can lead to abrupt changes in SWE during the analysis update. Second, the ensemble of SVM-based predictions can collapse (i.e., yield a near-zero standard deviation across the ensemble) when prior estimates of snow are outside the range of snow inputs used during the SVM training procedure. Such a scenario can lead to the presence of spurious error correlations between SWE and TB, and as a consequence, can result in degraded SWE estimates from the analysis update. These degraded analysis updates can be largely mitigated by applying rule-based approaches. For example, restricting the SWE update when the standard deviation of the predicted TB is greater than 0.05 K helps prevent the occurrence of filter divergence. Similarly, adding a thin layer (i.e., 5 mm) of SWE when the synthetic TB is larger than 5 K can improve SVM-DA performance in the presence of a precipitation dry bias. The study demonstrates that a carefully constructed SVM-DA framework cognizant of the inherent limitations of passive microwave-based SWE estimation holds promise for snow mass data assimilation.

Kwon, Yonghwan↗
DOE OSTI2022

Entanglement features of random neural network quantum states

Restricted Boltzmann machines (RBMs) are a class of neural networks that have been successfully employed as a variational ansatz for quantum many-body wave functions. Here, we develop an analytic method to study quantum many-body spin states encoded by random RBMs with independent and identically distributed complex Gaussian weights. By mapping the computation of ensemble-averaged quantities to statistical mechanics models, we are able to investigate the parameter space of the RBM ensemble in the thermodynamic limit. We discover qualitatively distinct wave functions by varying RBM parameters, which correspond to distinct phases in the equivalent statistical mechanics model. Notably, there is a regime in which the typical RBM states have near-maximal entanglement entropy in the thermodynamic limit, similar to that of Haar-random states. However, these states generically exhibit nonergodic behavior in the Ising basis, and do not form quantum state designs, making them distinguishable from Haar-random states.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Agilent AgileBioFoundry CRADA (Final Report)

The mission of this CRADA with Agilent was to couple powerful MS platforms (QQQ, IM-QTOF-MS) with Agilent’s novel Ultra-High-Performance Liquid Chromatography (UHPLC) fast metabolomic workflows and perform ABF Machine Learning (ML) to generated datasets. Agilent transferred UHPLC methods to PNNL and LBNL and methods were implemented and demonstrated in both labs, achieving total acquisition times of < 10 min. Metabolites analyzed using Agilent’s shared methods included metabolites from central carbon metabolism, common across hosts, and metabolites unique to engineered strains. Standards were acquired in an UHPLC-Drift Tube Ion Mobility Mass Spectrometer (DTIMS) system for the first time within the context of ABF and methods were optimized based on Agilent’s protocols. Samples from ABF hosts Pseudomonas putida, Aspergillus pseudoterreus, Aspergillus niger and Rhodosporidium toruloides were analyzed using the UHPLC-DTIMS platform for a total of 276 runs. A data analysis workflow compatible with the Experimental Data Depot (EDD) and completely shareable was developed for the acquired UHPLC-DTIMS data. Samples were analyzed using a Data Independent Acquisition Approach (DIA), which for most of the standards provided more transitions therefore increasing detection confidence. Using the data acquired by PNNL, LBNL, and Agilent’s specifications from previous ML projects, SNL applied an ensemble ML strategy to pick the best performing model for automated LC-method selection. Finally, with the contribution of the participant labs and Agilent, SNL developed an Automated Method Selection (AMS) software tool to predict the best liquid chromatography method for analysis of any new molecules of interest. Samples with novel pathways and new metabolite targets of interest are generated at a high pace in the ABF. Therefore, our accomplishment in this CRADA improved the efficiency and accuracy of strain testing by developing and implementing fast analytical methods, robust processing tools, and software for predicting the best methods for UHPLC analysis.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2023

The Statistical Emulators of GGCMI Phase 2: Responses of Year-to-Year Variation of Crop Yield to CO 2 , Temperature, Water and Nitrogen Perturbations

Understanding the impact of climate change on year-to-year variation of crop yield is critical to global food stability and security. While crop model emulators are believed to be lightweight tools to replace the models, few emulators have been developed to capture such interannual variation of crop yield in response to climate variability. In this study, we developed a statistical emulator with a machine learning algorithm to reproduce the response of year-to-year variation of four crop yields to CO 2 (C), temperature (T), water (W), and nitrogen (N) perturbations defined in the Global Gridded Crop Model Intercomparison Project (GGCMI) phase 2. The emulators were able to explain more than 52 % of the variance of simulated yield and performed well in capturing the year-to-year variation of global average and gridded crop yield over current croplands in the baseline. With the changes in CO 2 –temperature–water–nitrogen (CTWN) perturbations, the emulators could reproduce the year-to-year variation of crop yield well over most current cropland. The variation of R and the mean absolute error was small under the single CTWN perturbations and dual-factor perturbations. These emulators thus provide statistical response surfaces of yield, including both its mean and interannual variability, to climate factors. They could facilitate spatiotemporal downscaling of crop model simulation, projecting the changes in crop yield variability in the future and serving as a lightweight tool for multi-model ensemble simulation. The emulators enhanced the flexibility of crop yield estimates and expanded the application of large-ensemble simulations of crop yield under climate change.

Climate change↗
DOE OSTI2024

PreMevE‐MEO: Predicting Ultra‐Relativistic Electrons Using Observations From GPS Satellites

Abstract Ultra‐relativistic electrons with energies greater than or equal to two megaelectron‐volt (MeV) pose a major radiation threat to spaceborne electronics, and thus specifying those highly energetic electrons has a significant meaning to space weather communities. Here we report the latest progress in developing our predictive model for MeV electrons in the outer radiation belt. The new version, primarily driven by electron measurements made along medium‐Earth‐orbits (MEO), is called PREdictive MEV Electron (PreMevE)‐MEO model that nowcasts ultra‐relativistic electron flux distributions across the whole outer belt. Model inputs include >2 MeV electron fluxes observed in MEOs by a fleet of GPS satellites as well as electrons measured by one Los Alamos satellite in the geosynchronous orbit. We developed an innovative Sparse Multi‐Inputs Latent Ensemble NETwork (SmileNet) which combines convolutional neural networks with transformers, and we used long‐term in situ electron data from NASA's Van Allen Probes mission to train, validate, optimize, and test the model. It is shown that PreMevE‐MEO can provide hourly nowcasts with high model performance efficiency and high correlation with observations. This prototype PreMevE‐MEO model demonstrates the feasibility of making high‐fidelity predictions driven by observations from longstanding space infrastructure in MEO, thus has great potential of growing into an invaluable space weather operational warning tool.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2024

Deep energy-pressure regression for a thermodynamically consistent EOS model

Abstract In this paper, we aim to explore novel machine learning (ML) techniques to facilitate and accelerate the construction of universal equation-Of-State (EOS) models with a high accuracy while ensuring important thermodynamic consistency. When applying ML to fit a universal EOS model, there are two key requirements: (1) a high prediction accuracy to ensure precise estimation of relevant physics properties and (2) physical interpretability to support important physics-related downstream applications. We first identify a set of fundamental challenges from the accuracy perspective, including an extremely wide range of input/output space and highly sparse training data. We demonstrate that while a neural network (NN) model may fit the EOS data well, the black-box nature makes it difficult to provide physically interpretable results, leading to weak accountability of prediction results outside the training range and lack of guarantee to meet important thermodynamic consistency constraints. To this end, we propose a principled deep regression model that can be trained following a meta-learning style to predict the desired quantities with a high accuracy using scarce training data. We further introduce a uniquely designed kernel-based regularizer for accurate uncertainty quantification. An ensemble technique is leveraged to battle model overfitting with improved prediction stability. Auto-differentiation is conducted to verify that necessary thermodynamic consistency conditions are maintained. Our evaluation results show an excellent fit of the EOS table and the predicted values are ready to use for important physics-related tasks.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

COVID-19 dynamics across the US: A deep learning study of human mobility and social behavior

This paper presents a deep learning framework for epidemiology system identification from noisy and sparse observations with quantified uncertainty. The proposed approach employs an ensemble of deep neural networks to infer the time-dependent reproduction number of an infectious disease by formulating a tensor-based multi-step loss function that allows us to efficiently calibrate the model on multiple observed trajectories. The method is applied to a mobility and social behavior-based SEIR model of COVID-19 spread. The model is trained on Google and Unacast mobility data spanning a period of 66 days, and is able to yield accurate future forecasts of COVID-19 spread in 203 US counties within a time-window of 15 days. Interestingly, a sensitivity analysis that assesses the importance of different mobility and social behavior parameters reveals that attendance of close places, including workplaces, residential, and retail and recreational locations, has the largest impact on the effective reproduction number. Furthermore, the model enables us to rapidly probe and quantify the effects of government interventions, such as lock-down and re-opening strategies. Taken together, the proposed framework provides a robust workflow for data-driven epidemiology model discovery under uncertainty and produces probabilistic forecasts for the evolution of a pandemic that can judiciously provide information for policy and decision making. All codes and data accompanying this manuscript are available at https://github.com/PredictiveIntelligenceLab/DeepCOVID19.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2020

CoSim: A Simulator for Co-Scheduling of Batch and On-Demand Jobs in HPC Datacenters

The increasing scale and complexity of scientific applications are rapidly transforming the ecosystem of tools, methods, and workflows adopted by the high-performance computing (HPC) community. Big data analytics and deep learning are gaining traction as essential components in this ecosystem in a variety of scenarios, such as, steering of experimental instruments , acceleration of high-fidelity simulations through surrogate computations, and guided ensemble searches. In this context, the batch job model traditionally adopted by the supercomputing in-frastructures needs to be complemented with support to schedule opportunistic on-demand analytics jobs, leading to the problem of efficient preemption of batch jobs with minimum loss of progress. In this paper, we design and implement a simulator, CoSim, that enables on-the-fly analysis of the trade-offs arising between delaying the start of opportunistic on-demand jobs, which leads to longer analytics latency, and loss of progress due to preemption of batch jobs, which is necessary to make room for on-demand jobs. To this end, we propose an algorithm based on dynamic programming with predictable performance and scalability that enables supercomputing infrastructure schedulers to analyze the aforementioned trade-off and take decisions in near real-time. Compared with other state-of-art approaches using traces of the Theta pre-Exascale machine, our approach is capable of finding the optimal solution, while achieving high performance and scalability.

batch job preemption↗
DOE OSTI2025

Enhanced Boundary Layer Height Detection Using Ceilometer, Surface Meteorology, and Radiation Products With a Random Forest Ensemble Method

This study develops and evaluates a Random Forest (RF) model for estimating planetary boundary layer height (PBLH) using 9 years of data from the Atmospheric Radiation Measurement Southern Great Plains (ARM SGP) user facility, with potential application in the NOAA Surface Radiation (SURFRAD) Network. The model integrates ceilometer, surface meteorology, and radiation measurements, and is trained using thermodynamic PBLH estimates derived from radiosondes. This approach aims to bridge gaps between aerosol-based and thermodynamic-based PBLH estimates. The RF model outperformed traditional methods during daytime and better captured transition periods, demonstrating improved accuracy and robustness. At ARM SGP, it showed a substantial reduction in both bias and RMSE, with a bias near zero (−4.9 m) compared with traditional Haar Wavelet (HW) (70.9 m) and Vaisala BL-View software (124.1 m), and an RMSE of 303.2 m, lower than both BL-View (566.9 m) and HW (404.6 m). During daytime hours, RF consistently outperformed both alternatives, maintaining lower bias and RMSE across all periods. At a second evaluation site, RF achieved the lowest overall RMSE (323.7 m), similar to HW (326.4 m) and significantly better than BL-View (738.3 m). However, all models showed reduced accuracy under stable nighttime conditions, limiting the reliability of PBLH estimates. Key predictors for the model included the lifting condensation level height (LCLH), aerosol gradients, and month for seasonal variability. The study underscores the potential of integrating machine learning with multiple data sets such as surface energy and thermodynamic data to advance PBLH estimation.

boundary layer height↗
NASA NTRSDate not supplied

A Novel Machine Learning Method for Surface PM2.5 Estimations from Geostationary Satellites

Particulate matter (PM) with a diameter of less or equal to 2.5 μm, known as PM , affects human health as it penetrates the respiratory system. The Environmental Protection Agency (EPA) measures the atmospheric concentration of PM using air quality monitors stationed throughout the Continental United States (CONUS). Such measurements are points on a spatial domain and therefore, might not be representative of the air quality at nearby areas considering that the composition of the atmosphere is highly variable from place to place. Satellite based AOD permits a spatially uniform means of estimating PM and new geostationary satellites provide high temporal and spatial resolution estimation of AOD. However, the concentration of PM is non-linearly dependent on other atmospheric parameters that include relative humidity, temperature, and height of the planetary boundary layer. This information may be estimated at similar spatial and temporal resolutions as AOD from numerical modeling such as from the National Oceanic and Atmospheric Administration’s (NOAA) High Resolution Rapid Refresh (HRRR) model which resolves near real-time atmospheric conditions over the CONUS. The estimation of PM concentration is a multi-parametric problem that considers the effect of temporal dependencies among the different parameters. Deep learning approaches are appropriate for such complex estimation problems as they intrinsically capture relations among multiple non-linear parameters. This study compares deep-learning methods to traditional regression analysis to demonstrate the capabilities of these methods in predicting PM2.5 concentrations. Additionally, a novel ensemble learning approach is employed to identify scientific processes that could further improve the estimation of PM concentration. Utilizing Long Short-Term Memory (LSTM) neural networks, which are suitable for multivariate time series estimation problems as they are capable of learning long-term dependencies, individual models are created for each EPA station and trained on the aforementioned dataset collocated over each station. Individual station models are merged if the model's performance is improved by reducing the root mean squared error (RMSE) metric. This ensemble training method ultimately reduces the RMSE value. Evaluation of these results provide insights into physical processes and related observable parameters that may contribute to PM concentrations. Identified parameters evaluated to be statistically different between the merged and unmerged models are expected to improve overall performance. These new parameters are then utilized for reevaluation of the deep learning methods with an extreme gradient boosting model with an RMSE of 5.5 providing the best results.

George Priftis↗
DOE OSTI2025

Uncertainty Quantification for Data-Driven Machine Learning Models in Nuclear Engineering Applications: Where We Are and What Do We Need?

Machine learning (ML) has been leveraged to tackle a diverse range of tasks in almost all branches of nuclear engineering. Many of the successes in ML applications can be attributed to the recent performance breakthroughs in deep learning, the growing availability of computational power, data, and easy-to-use ML libraries. However, these empirical successes have often outpaced our formal understanding of the ML algorithms. An important but under-rated area is uncertainty quantification (UQ) of ML. ML-based models are subject to approximation uncertainty when they are used to make predictions, due to sources including but not limited to, data noise, data coverage, extrapolation, imperfect model architecture and the stochastic training process. The goal of this paper is to clearly explain and illustrate the importance of UQ of ML. We will elucidate the differences in the basic concepts of UQ of physics-based models and data-driven ML models. Various sources of uncertainties in physical modeling and data-driven modeling will be discussed, demonstrated, and compared. We will also present and demonstrate a few techniques to quantify the ML prediction uncertainties, including Monte Carlo dropout, deep ensemble, Bayesian neural networks, Gaussian Processes and conformal prediction. Lastly, we will discuss the need for building a verification, validation and UQ framework to establish ML credibility.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗