Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Vulnerabilities”

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 1,045 records · Page 58
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Learning-based framework for sensor fault-tolerant building HVAC control with model-assisted learning

As people spend up to 87% of their time indoors, intelligent Heating, Ventilation, and Air Conditioning (HVAC) systems in buildings are essential for maintaining occupant comfort and reducing energy consumption. These HVAC systems in smart buildings rely 'on real-time sensor readings, which in practice often suffer from various faults and could also be vulnerable to malicious attacks. Such faulty sensor inputs may lead to the violation of indoor environment requirements (e.g., temperature, humidity, etc.) and the increase of energy consumption. While many model-based approaches have been proposed in the literature for building HVAC control, it is costly to develop accurate physical models for ensuring their performance and even more challenging to address the impact of sensor faults. In this work, we present a novel learning-based framework for sensor fault-tolerant HVAC control, which includes three deep learning based components for 1) generating temperature proposals with the consideration of possible sensor faults, 2) selecting one of the proposals based on the assessment of their accuracy, and 3) applying reinforcement learning with the selected temperature proposal. Moreover, to address the challenge of training data insufficiency in building-related tasks, we propose a model-assisted learning method leveraging an abstract model of building physical dynamics. Through extensive experiments, we demonstrate that the proposed fault-tolerant HVAC control framework can significantly reduce building temperature violations under a variety of sensor fault patterns while maintaining energy efficiency.

Xu, Shichao↗
DOE OSTI2022

Stealthy Cyber Anomaly Detection On Large Noisy Multi-material 3D Printer Datasets Using Probabilistic Models

As Additive Layer Manufacturing (ALM) becomes pervasive in industry, its applications in safety critical component manufacturing are being explored and adopted. However, ALM's reliance on embedded computing renders it vulnerable to tampering through cyber-attacks. Sensor instrumentation of ALM devices allows for rigorous process and security monitoring, but also results in a massive volume of noisy data for each run. As such, in-situ, near-real-time anomaly detection is very challenging. The ideal algorithm for this context is simple, computationally efficient, minimizes false positives, and is accurate enough to resolve small deviations. In this paper, we present a probabilistic-model-based approach to address this challenge. To test our approach, we analyze current measurements from a polymer composite 3D printer during emulated tampering attacks. Our results show that our approach can consistently and efficiently locate small changes in the presence of substantial operational noise.

Yoginath, Srikanth↗
DOE OSTI2023

XploreNAS : Explore Adversarially Robust and Hardware-efficient Neural Architectures for Non-ideal Xbars

Compute In-Memory platforms such as memristive crossbars are gaining focus as they facilitate acceleration of Deep Neural Networks (DNNs) with high area and compute efficiencies. However, the intrinsic non-idealities associated with the analog nature of computing in crossbars limits the performance of the deployed DNNs. Furthermore, DNNs are shown to be vulnerable to adversarial attacks leading to severe security threats in their large-scale deployment. Thus, finding adversarially robust DNN architectures for non-ideal crossbars is critical to the safe and secure deployment of DNNs on the edge. This work proposes a two-phase algorithm-hardware co-optimization approach called XploreNAS that searches for hardware efficient and adversarially robust neural architectures for non-ideal crossbar platforms. We use the one-shot Neural Architecture Search approach to train a large Supernet with crossbar-awareness and sample adversarially robust Subnets therefrom, maintaining competitive hardware efficiency. Our experiments on crossbars with benchmark datasets (SVHN, CIFAR10, CIFAR100) show up to ~8–16% improvement in the adversarial robustness of the searched Subnets against a baseline ResNet-18 model subjected to crossbar-aware adversarial training. We benchmark our robust Subnets for Energy-Delay-Area-Products (EDAPs) using the Neurosim tool and find that with additional hardware efficiency–driven optimizations, the Subnets attain ~1.5–1.6× lower EDAPs than ResNet-18 baseline.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Automated Adversary-in-the-Loop Cyber-Physical Defense Planning

Security of cyber-physical systems (CPS) continues to pose new challenges due to the tight integration and operational complexity of the cyber and physical components. To address these challenges, this article presents a domain-aware, optimization-based approach to determine an effective defense strategy for CPS in an automated fashion—by emulating a strategic adversary in the loop that exploits system vulnerabilities, interconnection of the CPS, and the dynamics of the physical components. Our approach builds on an adversarial decision-making model based on a Markov Decision Process (MDP) that determines the optimal cyber (discrete) and physical (continuous) attack actions over a CPS attack graph. The defense planning problem is modeled as a non-zero-sum game between the adversary and defender. We use a model-free reinforcement learning method to solve the adversary’s problem as a function of the defense strategy. We then employ Bayesian optimization (BO) to find an approximate best-response for the defender to harden the network against the resulting adversary policy. This process is iterated multiple times to improve the strategy for both players. We demonstrate the effectiveness of our approach on a ransomware-inspired graph with a smart building system as the physical process. Numerical studies show that our method converges to a Nash equilibrium for various defender-specific costs of network hardening.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Interdisciplinary Approaches to Cybervulnerability Impact Assessment for Energy Critical Infrastructure

As energy infrastructure becomes more interconnected, understanding cybersecurity risks to production systems requires integrating operational and computer security knowledge. We interviewed 18 experts working in the field of energy critical infrastructure to compare what information they find necessary to assess the impact of computer vulnerabilities on energy operational technology. These experts came from two groups: 1) computer security experts and 2) energy sector operations experts. We find that both groups responded similarly for general categories of information and displayed knowledge about both domains, perhaps due to their interdisciplinary work at the same organization. Yet, their discussion of each group’s domain-specific training, motivations, and limitations, as well as their suggestions for collaboration across domains, highlighted how these two groups can work together to help each other secure the energy grid. Our findings inform the development of interdisciplinary security approaches in critical-infrastructure contexts.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Semantic Stealth: Crafting Covert Adversarial Patches for Sentiment Classifiers Using Large Language Models

Deep learning models have been shown to be vulnerable to adversarial attacks, in which perturbations to their inputs cause the model to produce incorrect predictions. As opposed to adversarial attacks in computer vision, where small changes introduced to pixel values can drastically alter a model's output while remaining imperceptible to humans, text-based attacks are difficult to conceal due to the discrete nature of tokens. Consequently, unconstrained gradient-based attacks often produce adversarial examples that lack semantic meaning, rendering them detectable through visual inspection or perplexity filters. In contrast to methods that rely on gradient-based optimization in the embedding space, we propose an approach that leverages a Large Language Model's ability to generate grammatically correct and semantically meaningful text to craft adversarial patches that seamlessly blend in with the original input text. These patches can be used to alter the behavior of a target model, such as a text classifier. Since our approach does not rely on gradient backpropagation, it only requires access to the target model's confidence scores, making it a grey-box attack. We demonstrate the feasibility of our approach using open-source LLMs, including Intel's Neural Chat, Llama2, and Mistral-Instruct, to generate adversarial patches capable of altering the predictions of a distilBERT model fine-tuned on the IMDB reviews dataset for sentiment classification.

Roa Carvajal, Maria↗
DOE OSTI2024

A Compound Data Poisoning Technique with Significant Adversarial Effects on Transformer-based Sentiment Classification Tasks

Transformer-based models have demonstrated much success in various natural language processing tasks. However, they are often vulnerable to adversarial attacks, such as data poisoning, which can intentionally fool the model into generating incorrect results. In this article, we present a novel, compound variant of a data poisoning attack on a transformer-based model that maximizes the poisoning effect while minimizing the scope of poisoning. Here we do so by combining the established data poisoning technique (label flipping) with a novel adversarial artifact selection and insertion technique aimed at minimizing detectability and the scope of the poisoning footprint. We find that by using a combination of these two techniques, we achieve a state-of-the-art attack success rate of approximately 90% while poisoning only 0.5% of the original training set, thus minimizing the scope and detectability of the poisoning action. These findings have the potential to advance the development of better data poisoning detection methods.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

ATTNChecker: Highly-Optimized Fault Tolerant Attention for Large Language Model Training

Large Language Models (LLMs) have demonstrated remarkable performance in various natural language processing tasks. However, the training of these models is computationally intensive and susceptible to faults, particularly in the attention mechanism, which is a critical component of transformer-based LLMs. In this paper, we investigate the impact of faults on LLM training, focusing on INF, NaN, and near-INF values in the computation results with systematic fault injection experiments. We observe the propagation patterns of these errors, which can trigger non-trainable states in the model and disrupt training, forcing the procedure to load from checkpoints. To mitigate the impact of these faults, we propose ATTNChecker, the first Algorithm-Based Fault Tolerance (ABFT) technique tailored for the attention mechanism in LLMs. ATTNChecker is designed based on fault propagation patterns of LLM and incorporates performance optimization to adapt to both system reliability and model vulnerability while providing lightweight protection for fast LLM training. Evaluations on four LLMs show that ATTNChecker on average incurs on average 7% overhead on training while detecting and correcting all extreme errors. Compared with the state-of-the-art checkpoint/restore approach, ATTNChecker reduces recovery overhead by up to 49×.

Liang, Yuhang [University of Alabama - Birmingham]↗
DOE OSTI2025

Position-Enhanced Gradient Attack (PEGA) on Medical Language Models

Federated Learning (FL) enables collaborative training of language models on sensitive clinical notes without sharing the data. However, this paradigm is vulnerable to gradient inversion attacks that can reconstruct private data from shared gradients. We find that state-of-the-art attacks are less effective in the medical domain, failing to overcome the unique challenges posed by its specialized vocabulary and unstructured format. To address this, we introduce the Position-Enhanced Gradient Attack (PEGA), a novel attack that makes gradients position-aware by optimizing token and position embeddings simultaneously. PEGA employs two key innovations: a periodic sorting of positional embeddings to resolve token order ambiguity and a late-stage embedding replacement strategy to correct hard-to-recover critical tokens. To evaluate the leakage of sensitive data more directly, we also propose the Unified PHI-Recall (UPHI), a new metric measuring the recovery of Protected Health Information. Experiments on the MIMIC-III dataset show that PEGA significantly outperforms leading attacks like TAG and LAMP, particularly in its ability to reconstruct identifiable patient information, exposing a more severe and nuanced privacy risk in federated medical NLP.

Xu, Nuo [University of Minnesota]↗
DOE OSTI2026

Detect the Unobservable: Abnormality Detection in mixed Autonomy for Lane Change Maneuver with Following Vehicles’ Trajectories Only

Highly Automated Vehicles (HAVs) and Advanced Driver-Assistance Systems (ADAS) are transforming modern transportation with enhanced mobility, safety, and efficiency. Despite their advantages, cybersecurity vulnerabilities in these systems can lead to abnormal behavior, posing significant risks to surrounding human-driven vehicles (HDVs) in mixed traffic environments. Here, this article addresses the challenge of detecting abnormal lateral movements of HAVs/ADAS vehicles using only trajectory profiles of following HDVs. Specifically, we propose a novel modeling approach that captures both normal and abnormal lateral behaviors through vehicle kinematics, integrated decision-making processes, vehicle control using symbolic regression for lane change vehicles. Additionally, we introduce an abnormality detection framework that relies on observable HDV data, even in occlusion scenarios. The framework evaluates the sensitivity of various car-following models to detect abnormal behaviors, providing insights into the interaction between HAVs/ADAS and HDVs in mixed autonomy systems.

Connected and Automated vehicles↗
DOE OSTI2023

Metrics for Decision-Making in Energy Justice

Energy equity and justice have become priority considerations for policymakers, practitioners, and scholars alike. To ensure that energy equity is incorporated into actual decisions and analysis, it is necessary to design, use, and continually improve energy equity metrics. In this article, we review the literature and practices surrounding such metrics. We present a working definition for energy justice and equity, and connect them to both criteria for and frameworks of metrics. We then present a large sampling of energy equity metrics, including those focused on vulnerability, wealth creation, energy poverty, life cycle, and comparative country-level dynamics. We conclude with a discussion of the limitations, gaps, and trade-offs associated with these various metrics and their interactions thereof.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗
DOE OSTI2023

Implications of Large- N c QCD for the NN Interaction

We present a method for ordering two-nucleon interactions based upon their scaling with the number of QCD colors, N c , in the limit that N c becomes large. Available data in the two-nucleon sector show general agreement with this ordering, indicating that the method may be useful in other contexts where data are less readily available. However, several caveats and potential pitfalls can make the large-N c ordering fragile and/or vulnerable to misinterpretation. We discuss the application of the large-N c analysis to two- and three-nucleon interactions, including those originating from weak and BSM (beyond the Standard Model) interactions, as well as two-nucleon external currents. Finally, we discuss some open questions in the field.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2026

The Impact of Operational Strategies and Load Fluctuations on Anion Exchange Membrane Water Electrolyzer Degradation

This study reports on the durability of anion exchange membrane water electrolyzers (AEMWEs) at the cell level in a supporting electrolyte (1 M) and the impact of operational profiles on cell degradation. While constant load shows significant voltage loss, including a current hold at 1 A cm −2 and a potential hold at 2 V, it generally occurs in the first 70 h of testing. These losses are likely due to passivation altering site reactivity and cell kinetics, and are largely recoverable upon load cycling and diagnostics at the end of test. Furthermore, square and triangular-wave cycling do not show cell degradation and indicate that AEMWEs are not vulnerable to load cycling. When operating outside of this potential window (1.45–2 V), cell voltage decay rates markedly increase and the losses are not recoverable. This includes a current density hold at 4.0 A cm −2 with high cell voltages (>2.4 V for test duration) and cycling between 0 and 2 V to simulate start–stop, where more than a 50% loss in current density occurs over 50 h. The findings presented here demonstrate the high durability of AEMWEs, particularly when operating with load fluctuations, and may allow for durability advantages compared to iridium and proton exchange membrane systems.

08 HYDROGEN↗
DOE OSTI2022

Nomogram for Predicting Postoperative Portal Venous Systemic Thrombosis in Patients with Cirrhosis Undergoing Splenectomy and Esophagogastric Devascularization

Objectives. The aim of the study is to develop a nomogram for predicting postoperative portal venous systemic thrombosis (PVST) in patients with cirrhosis undergoing splenectomy and esophagogastric devascularization. Methods. In total, 195 eligible patients were included. Demographic characteristics were collected, and the results of perioperative routine laboratory investigations and ultrasound examinations were also recorded. Blood cell morphological traits, including the red cell volume distribution width (RDW), mean platelet volume, and platelet distribution width, were identified. Univariate and multivariate logistic regressions were implemented for risk factor filtration, and an integrated nomogram was generated and then validated using the bootstrap method. Results. A color Doppler abdominal ultrasound examination on a postoperative day (POD) 7 (38.97%) revealed that 76 patients had PVST. The results of the multivariate logistic regression suggested that a higher RDW on POD3 (RDW3) (odds ratio (OR): 1.188, 95% confidence interval (CI): 1.073–1.326), wider portal vein diameter (OR: 1.387, 95% CI: 1.203–1.642), history of variceal hemorrhage (OR: 3.407, 95% CI: 1.670–7.220), and longer spleen length (OR: 1.015, 95% CI: 1.001–1.029) were independent risk parameters for postoperative PVST. Moreover, the nomogram integrating these four parameters exhibited considerable capability in PVST forecasting. The nomogram’s receiver operating characteristic curve reached 0.83 and achieved a sensitivity and specificity of 0.711 and 0.848, respectively, at its cutoff. The nomogram’s calibration curve demonstrated that it was well calibrated. Conclusion. The nomogram exhibited excellent performance in PVST prediction and might assist surgeons in identifying vulnerable patients and administering timely prophylaxis.

Chen, Miao↗
DOE OSTI2025

Hantavirus is Associated With Open Developed Areas and Arid Climates, Highlighting Increased Risk in the Western United States

In the United States, hantaviruses can cause hantavirus pulmonary syndrome (HPS) in humans, an acute respiratory illness with a high mortality rate. Most people contract HPS from exposure to infected rodent excrement. The interannual dynamics of hantavirus transmission are tied to both environmental and human-related factors, including changes in annual climate conditions, rodent populations, and the built environment in which humans are more likely to be exposed. Similar environmental conditions and socioeconomic factors also likely determine the long-term risk of hantavirus exposure. Here, we use ecological niche models and human cases of HPS in the U.S. from 1993 to 2022 to assess hantavirus risk using four socioeconomic variables, 17 land use variables, one variable of rodent richness, and seven climate variables to determine both the geographical locations of highest exposure risk and leading environmental predictors. We found that areas with higher relative risk tend to be where it is drier, higher social vulnerability, increased rodent richness, and more open to low levels of development—this largely mapped to the western U.S. We found evidence that fringe ecosystems may be important areas of hantavirus transmission, similar to other emerging diseases. Increased rodent richness was associated with increased hantavirus risk, warranting further investigation into how the abundance and community composition of rodents could impact long-term risk. These risk maps can help public health officials develop plans for mitigating hantavirus, especially for the most susceptible populations. They can also be used to further investigate regions estimated to be at high risk for hantavirus where disease cases have not been as common but may be underreported.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

AMIGO

The code provides methods for perturbing binaries without breaking functionality. It is useful for evaluating the security/robustness of existing detection systems. The files generated by AMIGO can also be used to retrain a vulnerable detector.

Koch, Lucas↗
DOE OSTI2021

Infrastructure Expression: Discovery Tools

The Infrastructure Expression: Discovery Tools is a toolset that allows for the automated collection of various network, hardware, and software vulnerabilities and details to be collected from systems on a given network. These details are then converted into a STIX 2.1 format, allowing for easy viewing in existing applications as well as sharing and interfacing with tools that already use the STIX 2.1 format.

Cutshaw, Michael↗
DOE OSTI2022

Time-Based CAN IDS Paper Results Code

Modern vehicles are complex cyber-physical systems made of hundreds of electronic control units (ECUs) that communicate over controller area networks (CANs). This inherited complexity has expanded the CAN attack surface which is vulnerable to message injection attacks. These injections change the overall timing characteristics of messages on the bus, and thus, to detect these malicious messages, time-based intrusion detection systems (IDSs) have been proposed. However, time-based IDSs are usually trained and tested on low-fidelity datasets with unrealistic, labeled attacks. This makes difficult the task of evaluating, comparing, and validating IDSs. Here we detail and benchmark four time-based IDSs against the newly published ROAD dataset, the first open CAN IDS dataset with real (non-simulated) stealthy attacks with physically verified effects. We found that methods that perform hypothesis testing by explicitly estimating message timing distributions have lower performance than methods that seek anomalies in a distribution related statistic. In particular, these “distribution-agnostic” based methods outperform “distribution-based” methods by at least 55% in area under the precision-recall curve (AUC-PR). Our results expand the body of knowledge of CAN time-based IDSs by providing details of these methods and reporting their results when tested on datasets with real advanced attacks. Finally, we develop an after-market plug-in detector using lightweight hardware, which can be used to deploy the best performing IDS method on nearly any vehicle.

Moriano, Pablo [Oak Ridge National Lab. (ORNL), Oa↗