Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “explainable AI”

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 37 records · Page 2
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

Using Explainable Artificial Intelligence to Predict Perovskite Solar Cell Electrical Metastability from Operando Photoluminescence Images in Accelerated Stress Testing

Metal halide perovskite (MHP) solar cells exhibit a metastable response to bias governed by coupled ionic–electronic processes, complicating the conventional reciprocity relation between luminescence intensity and device open-circuit voltage (V oc ). This limits the use of luminescence as a diagnostic for device screening or accelerated stress testing, motivating new approaches that can interpret photoluminescence (PL) signals under nonequilibrium conditions. From the artificial intelligence perspective, we develop an explainable deep learning framework that integrates convolutional neural networks (CNN), long short-term memory (LSTM) layers, and an attention mechanism to learn spatiotemporal features from operando photoluminescence PL image sequences. The model achieves a mean absolute error of ±0.027 V in predicting open-circuit voltage transients and reduces extreme-tail errors by up to 78% compared to physics-based reciprocity calculations. Gradient-weighted Class Activation Mapping (Grad-CAM) provides interpretability by highlighting physically meaningful regions such as electrode edges and emergent defect features. From the engineering application perspective, this framework enables accurate, contactless prediction of device V oc and identification of degradation-relevant features during accelerated aging of perovskite solar cells. This approach demonstrates how explainable AI can enhance operando diagnostics and reliability analysis in photovoltaic devices under nonequilibrium conditions.

14 SOLAR ENERGY↗
DOE OSTI2025

Deep Learning with Reflection High-Energy Electron Diffraction Images to Predict Cation Ratio in Sr 2 x Ti 2(1– x ) O 3 Thin Films

Machine learning (ML) with in-situ diagnostics offers a transformative approach to accelerate, understand, and control thin film synthesis by uncovering relationships between synthesis conditions and material properties. In this study, we demonstrate the application of deep learning to predict the stoichiometry of Sr 2x Ti 2(1–x) O 3 thin films using reflection high-energy electron diffraction images acquired during pulsed laser deposition. A gated convolutional neural network trained for regression of the Sr atomic fraction achieved accurate predictions with a small dataset of 31 samples. Explainable AI techniques revealed a previously unknown correlation between diffraction streak features and cation stoichiometry in Sr 2x Ti 2(1–x) O 3 thin films. Here, our results demonstrate how ML can be used to transform a ubiquitous in-situ diagnostic tool, that is usually limited to qualitative assessments, into a quantitative surrogate measurement of continuously valued thin film properties. Such methods are critically needed to enable real-time control, autonomous workflows, and accelerate traditional synthesis approaches.

77 NANOSCIENCE AND NANOTECHNOLOGY↗
DOE OSTI2024

Projecting Large Fires in the Western US With an Interpretable and Accurate Hybrid Machine Learning Method

More frequent and widespread large fires are occurring in the western United States (US), yet reliable methods for predicting these fires, particularly with extended lead times and a high spatial resolution, remain challenging. In this study, we proposed an interpretable and accurate hybrid machine learning (ML) model, that explicitly represented the controls of fuel flammability, fuel availability, and human suppression effects on fires. The model demonstrated notable accuracy with a F 1 -score of 0.846 ± 0.012, surpassing process-driven fire danger indices and four commonly used ML models by up to 40% and 9%, respectively. More importantly, the ML model showed remarkably higher interpretability relative to other ML models. Specifically, by demystifying the “black box” of each ML model using the explainable AI techniques, we identified substantial structural differences across ML fire models, even among those with similar accuracy. The relationships between fires and their drivers, identified by our model, were aligned closer with established fire physical principles. The ML structural discrepancy led to diverse fire predictions and our model predictions exhibited greater consistency with actual fire occurrence. With the highly interpretable and accurate model, we revealed the strong compound effects from multiple climate variables related to evaporative demand, energy release component, temperature, and wind speed, on the dynamics of large fires and megafires in the western US. Our findings highlight the importance of assessing the structural integrity of models in addition to their accuracy. They also underscore the critical need to address the rise in compound climate extremes linked to large wildfires.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Predicting homopolymer and copolymer solubility through machine learning

In this work, we report the development of multiple new machine learning (ML) models to accurately predict homopolymer/copolymer solubility over a diverse set of polymers & solvents, using explainable AI to provide polymer design recommendations.

Stubbs, Christopher D. [Department of Chemistry, C↗
DOE OSTI2026

Continental-Scale Controls on Hyporheic Respiration Revealed by Knowledge-Guided Machine Learning

Hyporheic zone sediments regulate organic matter turnover and in-stream respiration, yet controls on sediment respiration remain poorly constrained across heterogeneous river networks, limiting prediction of stream metabolism and carbon processing at continental scales. Here, we integrate observations from ~90 river corridors across the United States in the WHONDRS consortium with a knowledge-guided machine learning (KGML) framework that couples thermodynamic rate theory with machine learning to identify dominant controls on hyporheic respiration. Diagnostic analyses show that organic matter concentration and thermodynamic favorability define an upper bound on respiration potential, whereas biological catalytic capacity and physical accessibility jointly govern realized respiration rates through interaction effects. To represent unmeasurable accessibility constraints, we use the mechanistic model as a scaffold for KGML, allowing machine learning to target residual structure not explained by process theory. This hybrid framework improves predictive skill relative to both the mechanistic model alone and fully data-driven models while preserving interpretability. These results indicate that variability in hyporheic respiration is largely mechanistically structured and demonstrate how integrating process theory with explainable AI enhances predictive performance while enabling scalable synthesis of river corridor observations.

Zheng, Jianqiu↗
DOE OSTI2026

Data and scripts associated with a manuscript analyzing ELM-FATES parameter sensitivity under pre-fire and postfire scenarios using machine learning

NOTE: The manuscript associated with this data package is currently in review. The data may be revised based on reviewer feedback. Upon manuscript acceptance, this data package will be updated with the final dataset and additional metadata. This data package is associated with the manuscript “Fire Severity-Dependent Shifts in Vegetation Parameter Sensitivity: A Pre- and Post-Fire Analysis Using ELM-FATES and Explainable AI” submitted to Journal of Advances in Modeling Earth Systems (Zahura et al. 2026). The study examines vegetation physiological parameters controlling pre-fire and post-fire vegetation dynamics. To support this analysis, 73 vegetation parameters in Functionally Assembled Terrestrial Ecosystem Simulator (FATES) (Fisher et al., 2018) , which is coupled with E3SM (Energy Exascale Earth System Model) land model (ELM, ELM-FATES), were perturbed using a Sobol sequence to generate 1,024 ensemble members for two plant functional types: needleleaf evergreen extratropical trees (NEET) and C3 grass. Simulations were conducted for the pre-fire period (2016) and post-fire period (2018–2023). Burn severity was represented by modifying the Nesterov index in FATES to 75,000, 150,000, and 300,000 for low, moderate, and high severity, respectively. A no-fire scenario was also included. Simulations were performed for 16 grid cells in the American River Watershed across different burn severities and plant functional types. XGBoost (eXtreme Gradient Boosting) models were trained using the parameter ensembles and ELM-FATES-simulated outputs, including leaf area index (LAI), gross primary productivity (GPP), aboveground biomass, vegetation evaporation, transpiration, and soil evaporation. Models were trained separately for each year and burn severity, followed by SHAP (SHapley Additive exPlanations) analysis to identify changes in dominant parameters after fire disturbance. For details on how to navigate data packages generated by this project, see https://data.ess-dive.lbl.gov/portals/PNNLRiverCorridorSFA/About. The data package contains the ELM-FATES simulation data. The scripts and data related to the analysis will be added later. The inputs and outputs from ELM-FATES are inside the “FATES” folder. “FATES_domain_surface” contains the domain and surface netcdfs that were used to run ELM-FATES in the study area. “FATES_parameters” contains the 1024 ensembles that were generated using Sobol sequence. “FATES_outputs” folder contains ELM-FATES simulated variables. All files are .csv and .nc (NetCDF).

Aboveground biomass↗
DOE OSTI2024

Teaching Freight Mode Choice Models New Tricks Using Interpretable Machine Learning Methods

Understanding and forecasting the intricate freight mode choice behavior under various industry, policy, and technology contexts is essential in freight planning and policymaking. Numerous models have been developed in prior studies to provide insights into freight mode selection, the majority of which use discrete choice models such as multinomial logit (MNL) models. However, logit models often rely on linear specifications of independent variables, despite potential nonlinear relationships in the data. Moreover, there often lacks a heuristic and efficient approach to identify such complex relationships to define the logit model specifications. To fill this gap, we developed an MNL model for freight mode choice using the insights from state-of-the- art machine learning (ML) models. ML models can capture the nonlinear nature of the complex decision-making process, and recent advances in 'explainable AI' have greatly improved their interpretability. The interpretable ML methods help enhance the performance of MNL models and advance knowledge of freight mode choice. Specifically, the influential factors and their relationship with individual modes are identified using SHapley Additive exPlanations (SHAP) to improve the MNL's performance. The workflow is demonstrated in a case study of Austin, Texas, and the SHAP results reveal multiple nonlinear relationships predicted by ML models. Incorporating those relationships into MNL model specifications improves the interpretability and accuracy of the MNL model compared to a conventional MNL model. Findings from this study can be used to guide freight planning and inform policymakers and practitioners on how key factors affect freight decision-making.

ADVANCED PROPULSION SYSTEMS,MATHEMATICS AND COMPUT↗
DOE OSTI2025

Edge ML for CAN bus intrusion detection in AVs

Autonomous Vehicles (AVs) are revolutionizing transportation, but their reliance on interconnected cyber-physical systems exposes them to unprecedented cybersecurity risks. This study addresses the critical challenge of detecting real-time cyber intrusions in self-driving vehicles by leveraging a dataset from the Udacity self-driving car project. We simulate four high-impact attack vectors, Denial of Service (DoS), spoofing, replay, and fuzzy attacks, by injecting noise into spatial features (e.g., bounding box coordinates) to replicate adversarial scenarios. We develop and evaluate two lightweight neural network architectures (NN-1 and NN-2) alongside a logistic regression baseline (LG-1) for intrusion detection. The models achieve exceptional performance, with NN-2 attaining an AUC score of 93.15% and 93.15% accuracy, demonstrating their suitability for edge deployment in AV environments. Through explainable AI techniques, we uncover unique forensic fingerprints of each attack type, such as spatial corruption in fuzzy attacks and temporal anomalies in replay attacks, offering actionable insights for feature engineering and proactive defense. Visual analytics, including confusion matrices, ROC curves, and feature importance plots, validate the models' robustness and interpretability. This research sets a new benchmark for AV cybersecurity, delivering a scalable, field-ready toolkit for Original Equipment Manufacturers (OEMs) and policymakers. By aligning intrusion fingerprints with SAE J3061 automotive security standards, we provide a pathway for integrating machine learning into safety-critical AV systems. Our findings underscore the urgent need for security-by-design AI, ensuring that AVs not only drive autonomously but also defend autonomously. This work bridges the gap between theoretical cybersecurity and life-preserving engineering, offering a leap toward safer, more secure autonomous transportation.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Regulation compliant AI for fusion: explainable image-based feedback control of divertor detachment in DIII-D tokamak

While artificial intelligence (AI) has been promising for fusion control, its inherent black-box nature will make compliant implementation in regulatory environments a challenge. This study implements and validates a real-time AI-enabled linear and interpretable control system for successful divertor detachment control with the DIII-D lower divertor camera. Using D 2 gas, we demonstrate successful feedback divertor detachment control with a mean absolute difference of 2% from the target for both detachment and reattachment. This automatic training and linear processing framework can be extended to any image-based diagnostic for future fusion reactors.

computer vision↗
DOE OSTI2024

Human Factors Considerations in Artificial Intelligence Applications for Nuclear Power Plants

In recent years, there has been a wave of artificial intelligence (AI) technologies that offer to solve problems from shopping habits to mortgage approvals to critical systems operations. The rapidity of the development of these systems has led to both excitement and apprehension about the roles these systems should play in our modern societies. Furthermore, this paper focuses on the critical infrastructure industry, in general, and nuclear power generation, in particular, and seeks to scrutinize how we can leverage these novel technologies in human-centered ways to maintain or enhance the established high levels of reliability and resilience in these industries. First, we discuss the broader aspects of cognitive systems and activities that are critical to understanding the human-AI space. Then we explore different approaches to explainability in AI and the notions of trust. We then move on to discuss several human factors concepts and methods and how they can support the design of human-AI teams. We then explore recent research related to nuclear power that has been undertaken and evaluate the current industry and regulatory landscapes. Finally, we discuss identified research gaps and recommendations for solving these for the critical infrastructure space.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2024

Generative AI for Power Grid Operations

Generative artificial intelligence (AI) has captured into the mainstream, demonstrating capabilities that once belonged solely to the realm of human cognition. From defeating world champions in complex games to generating human-quality text and images, Generative AI has proven its potential to revolutionize countless industries. The electric power grid is no exception. Generative AI's ability to process vast amounts of data rapidly, assist decision support and identify patterns could significantly enhance power grid operations. For example, Generative AI could improve state estimation where measurements are not available or integrate renewable energy sources more efficiently with probabilistic forecasting. The key contributions of this whitepaper are outlined below: (1) Comprehensive overview of Generative AI's applications in power grid operations: It highlights the opportunities in areas such as forecasting, state estimation, and demonstrating the potential for enhancing efficiency, reliability, and resilience. (2) Expanding Generative AI's impact through synergies with emerging technologies: The paper introduce NREL developed eGridGPT and explores how AI orchestration, multi-agent systems, and Digital Twins can collaborate to optimize grid operations, addressing the complexities of a decarbonized and electrified future. (3) In-depth analysis of challenges in implementing Generative AI: This includes considerations like data availability and quality, model validation, certification, and ethical concerns, ensuring responsible AI deployment. (4) Emphasizing human-AI collaboration: The whitepaper underscores the importance of trustworthy, transparency, and explainability in AI systems to promote seamless interaction between human operators and AI, ultimately improving decision-making. (5) Exploring future research and development: It identifies critical areas for further advancement to fully realize Generative AI's potential in power grid operations. This whitepaper serves as a valuable resource for researchers, practitioners, and policymakers looking to harness Generative AI for a more reliable, stable, and cost-effective power grid.

24 POWER TRANSMISSION AND DISTRIBUTION↗
NASA NTRS2022

TruePAL – An AI Assistant for First Responder Safety

This paper presents the development of an AI assistant, Trusted and Explainable Artificial Intelligence for Saving Lives (TruePAL), to provide real-time warning of risks of potential crashes to the first responders. The TruePAL system employs an AI and deep learning technology for saving first responders and roadside crews lives in and around active traffic. A deep neural network (DNN) and a Non-Axiomatic Reasoning System (NARS) are implemented as an AI system. A mobile app with AI interface is developed to perform verbal communication with the first responders. The TruePAL team has developed an explainable AI approach by opening up the DNN blackbox to extract the activation filters of various features and parts of the targeted objects. The combination of DNN and NARS makes the TruePAL system explainable to the users. TruePAL ingests on-board cameras, radar, and other sensor signals, analyzes the environment and traffic patterns to generate timely warning to drivers and roadside crews to avoid crashes. The TruePAL team, in collaboration with the Miami/Dade Police Dept., has designed five use cases and multiple sub-scenarios in a CARLA driving simulator to test the capability of TruePAL in timely warning to the first responder drivers in potential crash scenarios. We have successfully demonstrated its capability of timely warning in over a dozen scenarios based on the use cases. The preliminary test simulation results show that TruePAL could provide the drivers and crew members advanced warning before a crash occurs.

Chow, Edward↗
DOE OSTI2026

What Is the Agent Doing? Visualizing Agentic AI Querying Workflows

We explore how visualizations can help users understand what an AI agent is doing as it builds and runs queries over data. As part of the LinkQ system, a natural language interface for querying knowledge graphs with a large language model (LLM), we designed two complementary views: A State Diagram that shows where the agent is within a larger workflow, and a Live Action Display that gives real-time updates about the agent's current task. In a study with 14 practitioners, we found that these visuals helped participants build stronger mental models of the agent's behavior while also increasing their confidence in the system. However, we also observed that users sometimes trusted incorrect outputs simply because the agent appeared to be doing the "right" thing. Our findings point to both the value and risk of visualizing agent behavior in interactive AI systems.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Explainable machine learning reveals that local structural motifs encode the thermodynamic state across the CuZr metallic glass-forming range

Metallic glasses derive their properties from the statistics of local atomic motifs rather than from long-range order, yet a quantitative, chemistry-specific link between motif populations and the underlying glassy state has remained elusive. In this work we combine large-scale molecular dynamics, Voronoi tessellation, deep neural networks, and SHapley Additive exPlanations (SHAP) to identify which local structural motifs define the glassy state of Cu—Zr metallic glasses. A dataset of 17,180 atomistic configurations spanning ten compositions (Cu 20 Zr 80 –Cu 80 Zr 20 ) and four quench rates (10 9 –10 12 K/s) is used to train a feed-forward neural network that regresses temperature across the 50–2000 K liquid–supercooled–glass range, achieving a mean absolute error of 19.89 K and R 2 = 0.9974, confirming that the local structural state is faithfully encoded in motif-level structure. SHAP analysis then reveals that a tightly coupled near-icosahedral family of motifs (coordination numbers (CN) 11–13, including the full icosahedron 001200 and its single-atom-perturbation sibling 10930) collectively encodes the thermodynamic state of the system across the full glass-forming range. The CN = 11–13 ordered members carry negative SHAP values at high populations, tracking the most deeply-quenched configurations, while 10930 shows the reversed signature consistent with its role as a soft-spot host whose population shrinks as the icosahedral network deepens. The analysis demonstrates that explainable machine learning can isolate the minimal motif vocabulary defining the glassy state and recovers the near-icosahedral building blocks previously identified by data-driven analyses of Cu—Zr. The approach provides a general, chemistry-specific route for characterizing the structural state of disordered materials.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Concept Lens: Visual Comparison and Evaluation of Generative Model Manipulations

Generative models are becoming a transformative technology for the creation and editing of images. However, it remains challenging to harness these models for precise image manipulation. These challenges often manifest as inconsistency in the editing process, where both the type and amount of semantic change, depend on the image being manipulated. Moreover, there exist many methods for computing image manipulations, whose development is hindered by the matter of inconsistency. This paper aims to address these challenges by improving how we evaluate, compare, and explore the space of manipulations offered by a generative model. We present Concept Lens, a visual interface that is designed to aid users in understanding semantic concepts carried in image manipulations, and how these manipulations vary over generated images. Given the large space of possible images produced by a generative model, Concept Lens is designed to support the exploration of both generated images, and their manipulations, at multiple levels of detail. To this end, the layout of Concept Lens is informed by two hierarchies: a hierarchical organization of (1) original images, grouped by their similarities, and (2) image manipulations, where manipulations that induce similar changes are grouped together. This layout allows one to discover the types of images that consistently respond to a group of manipulations, and vice versa, manipulations that consistently respond to a group of codes. We show the benefits of this design across multiple use cases, specifically, studying the quality of manipulations for a single method, and offering a means of comparing different methods.

clustering↗
DOE OSTI2025

Physical Interpretation of Early Battery Life Prediction Models

Early battery life prediction models are most useful for R&D if they help us understand the early changes in battery electrochemical response that correspond with long-term degradation and failure. Linear regression models such as Fused lasso and Partial Least Squares can fit coefficients directly to high-dimensional electrochemical data like capacity-voltage and ΔV–state-of-charge, i.e., Q(V) and ΔV(SOC) curves, learning coefficients that can be physically interpreted. We leverage the ISU-ILCC battery aging data set to learn high-dimensional coefficients for early battery life prediction from traditional slow-rate capacity check data, demonstrating learning on Q(V), d Q· d V −1 , and ΔV(SOC) curves. A thorough study on the dependence of coefficient values on train/test size and data preprocessing methods is made, demonstrating the reliability of high-dimensional regression approaches unless very small amounts of data are used for model training. For this data set, coefficients from Q(V) and d Q· d V −1 models highlight changes in electrode stoichiometry due to lithium loss, while ΔV(SOC) coefficients highlight changes in positive electrode diffusivity due to particle cracking as well as electrode stoichiometry shifts. By directly interpreting the coefficients of a regression model, we make physical insights into battery degradation mechanisms without requiring the assumptions of traditional battery data analysis methods.

25 ENERGY STORAGE↗
NASA NTRSDate not supplied

Beyond Fair: Engagement, Data Usability, and Open Community Productivity through the NASA Open Science Data Repository

The FAIR principle (findable, accessible, interoperable, and reusable) governs the storage and sharing of NASA space biology and health data[1]. These guiding principles maximize reuse of data and the reproducibility of scientific findings. The NASA Open Science Data Repository (OSDR; an expansion of NASA GeneLab) was built on the FAIR principles and houses over 500 studies and close to 1000 datasets from decades of space life sciences experiments. OSDR embodies the FAIR principles through data governance that includes mediated, embargoed, and fully open access data. The FAIR data governance principles were recently proposed to be expanded to encompass a FAIREST framework for assessing research data repositories (FAIR + Engagement, Social connections, and Trust)[2]. FAIREST emphasizes the importance of data repositories engaging with the scientific community and gaining the trust of researchers regarding data quality. Trust also refers to the TRUST principles developed for assessment of digital repositories: Transparency, Responsibility, User Focus, Sustainability, Technology[3]. We present the “Open Science for Life in Space” Analysis Working Groups (AWGs) as evidence regarding the power of engagement, social connections, and trust which has enhanced OSDR’s capabilities and productivity. AWG members engage in two main activities. One, members provide feedback on OSDR scientific standards for data ingestion, curation, and reuse (study, subject and assay metadata; processing pipelines; dataset formats and uniformed structures for machine-readability). Two, AWG members collaborate to mine-reuse OSDR data to conduct scientific analysis. With nearly 800 active members, the AWGs have resulted in 32 publications re-using OSDR data and contributed many papers in two major special issues in Cell (2020) and Nature (2024). AWGs also serve as networking groups, facilitate social connections between researchers at all levels of experience, and also have a social online ‘Forum’ used to keep members informed on projects and opportunities. This community-centric, productive, and trustworthy data culture has resulted in a broader effect with international space agencies, academics, and the commercial space sector wanting to submit their data to OSDR. Ten studies of Inspiration 4 data were recently publicly released by OSDR, as were some JAXA human data. Coming up soon in OSDR are data submissions from the European Space Agency, Virgin Galactic PIs, and SpaceX Polaris Dawn. A major benefit of OSDR is the array of standardized and uniformly formatted data (which was developed through AWG member consensus), from which visualization tools, analysis tools, and machine learning models can be built or trained. This talk will cover the Multi-Study Visualization Tool, the Environmental Data Application, RadLab, and a UCSF-NSF funded knowledge graph biomedical health discovery tool ‘SPOKE’ currently being integrated with OSDR. OSDR also provides training programs in bioinformatics and machine learning to improve the scientific community’s awareness of data availability and to boost their ability to perform data analysis. The increasing engagement of the scientific community and the public with technologies powered by artificial intelligence (AI) heightens the need for data analysis to be transparent. The AI for Life in Space initiative leverages the data products provided in OSDR to train AI models, with an emphasis on explainable and trustworthy AI, which would not be possible without FAIR data and metadata. Overall, here we will demonstrate the importance for NASA life sciences data repositories to adhere to the FAIREST framework, by providing examples and success stories from different aspects of OSDR.

data↗
DOE OSTI2026

Explainable tokamak-agnostic forecasting of fusion plasma instability via megahertz turbulent fluctuations

Scientific applications of artificial intelligence (AI) often remain limited by device-specific training and unexplained “black-box” approaches, creating fundamental barriers to cross-system generalization. This challenge is critical for nuclear fusion, where future reactors will have limited operational data for AI training. Here, we demonstrate that our neural network, trained solely on megahertz-scale turbulence measurements from one machine (DIII-D), forecasts Type-I edge localized mode (ELM) onsets in a different tokamak (KSTAR) through zero-shot weight transfer following physics-consistent preprocessing without device-specific retraining. Through an explainable AI framework combining gradient-weighted class activation mapping with physics validation, we reveal that our network can internalize physics relationships governing the ELM instabilities rather than memorizing device-specific patterns. The network perceives spatiotemporal features that correlate consistently with independently calculated instability growth rates, magnetohydrodynamic stability limits, and pedestal structure dynamics. Statistical analyses of dimensionally-reduced saliency features reveal the identical triangular features between the saliency representations, instability growth rates, and prediction probability across tokamaks, providing evidence that our forecasting system can show tokamak-agnostic generalization. This work contributes to a foundation for explainable scientific AI systems, where cross-system developments are essential for transcending traditional domain-specific constraints.

AI↗