Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “THREAT”

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 739 records · Page 41

Analytic Ballistic Performance Model of Whipple Shields

The dual-wall Whipple shield is the shield of choice for lightweight, long-duration flight. The shield uses an initial sacrificial wall to initiate fragmentation and melt an impacting threat that expands over a void before hitting a subsequent shield wall of a critical component. The key parameters to this type of shield are the rear wall and its mass which stops the debris, as well as the minimum pressure generated under threat particle impact of the sacrificial wall and the amount of void that is available for expansion. Ensuring the minimum pressure is sufficiently high to achieve large scale fragmentation/melt of the threat particle enables the expansion of the threat and reduces the momentum flux of the debris on the rear wall. Three key factors in the minimum pressure achieved are the thickness of the sacrificial wall relative to the characteristic dimension of the impacting particle, the density and material cohesion contrast of the sacrificial wall relative to the threat particle and the impact speed. The mass of the rear wall and the sacrificial wall are desirable to minimize for launch costs and dynamic concerns making it important to have an understanding of the effects of density contrast and impact speed. In this paper a fourth key parameter is identified related to fragmentation, which corresponds to the ratio of the size of the projectile relative to the transition from brittle to ductile hole growth in the projectile. Ballistic limit equations have been developed to define the failure limits of a MMOD shield, generally in terms of projectile diameter (or mass), impact velocity, and angle. Within the range of impact velocities relevant for Earth-orbiting spacecraft, three distinct regions of penetration phenomenology have been identified for Whipple shields: center dot Low velocity: the projectile is eroded (and possibly deformed) during its passage through the bumper plate, but is not fragmented. Thus, perforation of the rear wall is by a fragment with a mass and speed equal to or less than the original impactor. center dot Intermediate (shatter) velocity: impact velocities are sufficient to induce projectile fragmentation upon impact with the bumper plate, resulting in a coarse debris cloud with large solid fragments. Increasing velocity within the shatter regime results in increased fragmentation, and eventually melting, of the projectile and bumper fragments, generating a finer and more evenly dispersed debris cloud. Failure of the rear wall is a complicated combination of modes observed at low- and hypervelocity. center dot Hypervelocity: the projectile and holed-out bumper material is completely, or nearly completely, melted and/or vaporized by the initial impact. The resultant debris cloud impacts over a dispersed area of the rear wall, loading it impulsively and inducing failure through rupture or petalling. While each of these regimes are well observed with extensive empirical methods to describe these regions, differences in impactor materials, configurations of shields and questions about the limitations of the attainable impact speeds have left questions that are difficult to answer from completely empirical methods.

Miller, J. E.↗

Analytic Ballistic Performance Model of Whipple Shields

The dual-wall, Whipple shield is the shield of choice for lightweight, long-duration flight. The shield uses an initial sacrificial wall to initiate fragmentation and melt an impacting threat that expands over a void before hitting a subsequent shield wall of a critical component. The key parameters to this type of shield are the rear wall and its mass which stops the debris, as well as the minimum shock wave strength generated by the threat particle impact of the sacrificial wall and the amount of room that is available for expansion. Ensuring the shock wave strength is sufficiently high to achieve large scale fragmentation/melt of the threat particle enables the expansion of the threat and reduces the momentum flux of the debris on the rear wall. Three key factors in the shock wave strength achieved are the thickness of the sacrificial wall relative to the characteristic dimension of the impacting particle, the density and material cohesion contrast of the sacrificial wall relative to the threat particle and the impact speed. The mass of the rear wall and the sacrificial wall are desirable to minimize for launch costs making it important to have an understanding of the effects of density contrast and impact speed. An analytic model is developed here, to describe the influence of these three key factors. In addition this paper develops a description of a fourth key parameter related to fragmentation and its role in establishing the onset of projectile expansion.

Miller, J. E.↗

System Effectiveness

An effective risk assessment system is needed to address the threat posed by an active or passive insider who, acting alone or in collusion, could attempt diversion or theft of nuclear material. It is critical that a nuclear facility conduct a thorough self-assessment of the material protection, control, and accountability (MPC&A) system to evaluate system effectiveness. Self-assessment involves vulnerability analysis and performance testing of the MPC&A system. The process should lead to confirmation that mitigating features of the system effectively minimize the threat, or it could lead to the conclusion that system improvements or upgrades are necessary to achieve acceptable protection against the threat. Analysis of the MPC&A system is necessary to understand the limits and vulnerabilities of the system to internal threats. Self-assessment helps the facility be prepared to respond to internal threats and reduce the risk of theft or diversion of nuclear material. MSET is a self-assessment or inspection tool utilizing probabilistic risk assessment (PRA) methodology to calculate the system effectiveness of a nuclear facility's MPC&A system. MSET analyzes the effectiveness of an MPC&A system based on defined performance metrics for MPC&A functions based on U.S. and international best practices and regulations. A facility's MC&A system can be evaluated at a point in time and reevaluated after upgrades are implemented or after other system changes occur. The total system or specific subareas within the system can be evaluated. Areas of potential performance improvement or system upgrade can be assessed to determine where the most beneficial and cost-effective improvements should be made. Analyses of risk importance factors show that sustainability is essential for optimal performance. The analyses reveal where performance degradation has the greatest detrimental impact on total system risk and where performance improvements have the greatest reduction in system risk. The risk importance factors show the amount of risk reduction achievable with potential upgrades and the amount of risk reduction actually achieved after upgrades are completed. Applying the risk assessment tool gives support to budget prioritization by showing where budget support levels must be sustained for MC&A functions most important to risk. Results of the risk assessment are also useful in supporting funding justifications for system improvements that significantly reduce system risk.

99 GENERAL AND MISCELLANEOUS↗

Explosive Detection in Aviation Applications Using CT

CT scanners are deployed world-wide to detect explosives in checked and carry-on baggage. Though very similar to single- and dual-energy multi-slice CT scanners used today in medical imaging, some recently developed explosives detection scanners employ multiple sources and detector arrays to eliminate mechanical rotation of a gantry, photon counting detectors for spectral imaging, and limited number of views to reduce cost. For each bag scanned, the resulting reconstructed images are first processed by automated threat recognition algorithms to screen for explosives and other threats. Human operators review the images only when these automated algorithms report the presence of possible threats. The US Department of Homeland Security (DHS) has requirements for future scanners that include dealing with a larger number of threats, higher probability of detection, lower false alarm rates and lower operating costs. One tactic that DHS is pursuing to achieve these requirements is to augment the capabilities of the established security vendors with third-party algorithm developers. A third-party in this context refers to academics and companies other than the established vendors. DHS is particularly interested in exploring the model that has been used very successfully by the medical imaging industry, in which university researchers develop algorithms that are eventually deployed in commercial medical imaging equipment. The purpose of this paper is to discuss opportunities for third-parties to develop advanced reconstruction and threat detection algorithms.

42 ENGINEERING↗

Method for Determining the Sensitivity of a Physical Security System.

Modern systems, such as physical security systems, are often designed to involve complex interactions of technological and human elements. Evaluation of the performance of these systems often overlooks the human element. A method is proposed here to expand the concept of sensitivity—as denoted by d’—from signal detection theory (Green & Swets 1966; Macmillan & Creelman 2005), which came out of the field of psychophysics, to cover not only human threat detection but also other human functions plus the performance of technical systems in a physical security system, thereby including humans in the overall evaluation of system performance. New in this method is the idea that probabilities of hits (accurate identification of threats) and false alarms (saying “threat” when there is not one), which are used to calculate d’ of the system, can be applied to technologies and, furthermore, to different functions in the system beyond simple yes-no threat detection. At the most succinct level, the method returns a single number that represents the effectiveness of a physical security system; specifically, the balance between the handling of actual threats and the distraction of false alarms. The method can be automated, and the constituent parts revealed, such that given an interaction graph that indicates the functional associations of system elements and the individual probabilities of hits and false alarms for those elements, it will return the d’ of the entire system as well as d’ values for individual parts. The method can also return a measure of the response bias* of the system. One finding of this work is that the d’ for a physical security system can be relatively poor in spite of having excellent d’s for each of its individual functional elements.

98 NUCLEAR DISARMAMENT, SAFEGUARDS, AND PHYSICAL P↗

Improving Cyber Situational Understanding

Effective cybersecurity operations require the ability to analyze large amounts of information to assess security risks and formulate defensive strategies against adversaries. This has become more complex in recent years as the sprawl and interconnectivity of devices grows through implementation of virtualization, cloud computing, and Internet of Things (IoT). The amount of data and analysis required for effective cybersecurity command and control decisions far exceeds humans’ capacity to perform manually. We characterize the analysis problem as cyber situational understanding. The research presented to improve cyber situational understanding focuses on vulnerability analysis and threat intelligence. Regarding vulnerabilities, entities must analyze and plan work for between thousands and tens of thousands of software vulnerabilities annually. Entities heavily use network firewalls to limit vulnerability exposure. As a result, some of these vulnerabilities permit exposure to adversarial exploitation, whereas others are inaccessible and therefore present negligible risk of exploitation. Distinguishing between high and low risk software vulnerabilities requires a deep understanding of the vulnerability, network firewall protection, and characteristics of the targeted device. This problem is solved by extracting network service features from vulnerability data features using both machine-learning and natural language processing. Then, the network firewall topology is parsed to determine which vulnerabilities are reachable by adversaries. Ultimately, a state-based safety analysis ascertains which vulnerabilities are unsafe. A related vulnerability analysis problem occurs in cybersecurity operations when associating an entity’s hardware and software assets to public vulnerability databases. Assets often reveal hardware and software through installation artifacts and network service identification, and entities store these artifacts in inventory databases. However, software and hardware vendors apply a standard Common Platform Enumeration (CPE) naming convention when publicly reporting vulnerabilities. Associating these two datasets often requires many hours to days of manual inspection. The proposed solution automates the mapping approach of human analysts using fuzzy matching techniques, natural language processing, and, ultimately, machine learning to present a small set of recommendations for mapping the two datasets. The result significantly reduces human analysis time and reduces the occurrence of false positives in vulnerability notifications. Finally, cyber threat intelligence (CTI) requires associating cyber observable artifacts, such as IP addresses, URIs, and file hashes, with cyber threat tactics, techniques, and procedures. Unfortunately, most CTI data is compartmentalized across multiple organizations and cannot be shared due to the legal and reputational risk with cyber threat being associated with the entity. The approach to solving this problem inovlves using a distributed ledger with anonymous token spending and authentication. This allows a consortium of semi-trusted entities to share the workload of curating CTI for a threat sharing community’s cooperative benefit.

Huff, Philip↗

Radiation Anomaly Detection Algorithms for Field-Acquired Gamma Energy Spectra

The Remote Sensing Laboratory (RSL) is developing a tactical, networked radiation detection system that will be agile, reconfigurable, and capable of rapid threat assessment with a high degree of fidelity and certainty. Our design is driven by the needs of users such as law enforcement personnel who must make decisions by evaluating threat signatures in urban settings. The most efficient tool available to identify the nature of the threat object is real-time gamma spectroscopic analysis, as it is fast and has a very low probability of producing false positive alarm conditions. Urban radiological searches are inherently challenged by the rapid and large spatial variation of background gamma radiation, the presence of benign radioactive materials in terms of the normally occurring radioactive materials (NORM), and shielded and/or masked threat sources. Multiple spectral anomaly detection algorithms have been developed by national laboratories and commercial vendors. For example, the Gamma Detector Response and Analysis Software (GADRAS) a one-dimensional deterministic radiation transport software capable of calculating gamma ray spectra using physics-based detector response functions, was developed at Sandia National Laboratories. The nuisance-rejection spectral comparison ratio anomaly detection algorithm (or N-SCRAD), developed at Pacific Northwest National Laboratory, uses spectral comparison ratios to detect deviation from benign medical and NORM radiation source and can work in spite of strong presence of NORM and or medical sources. RSL has developed its own wavelet-based gamma energy spectral anomaly detection algorithm called WAVRAD. Test results and relative merits of these different algorithms will be discussed and demonstrated.

networked sensor, gamma spectroscopy, false alarm ↗

Hungary 908 Event - Risk Based Graded Approach to ITM

This presentation, Risk-Based, Graded Approach to Insider Threat Mitigation: Human Measures, introduces a structured framework for managing insider threat risk using internationally recognized guidance from the International Atomic Energy Agency (IAEA) Nuclear Security Series No. 8-G (Rev. 1) and the Joint Statement on Mitigating Insider Threats (INFCIRC/908). The presentation emphasizes that effective insider threat mitigation (ITM) depends on both positional controls, which manage inherent risk based on access, authority, and knowledge, and human measures, which address residual risk reflected in behavior, motivation, and reliability. Using a risk-informed and graded approach, the presentation outlines methods for identifying and prioritizing high-risk positions, applying layered organizational controls, and integrating human reliability mechanisms such as the Behavior Observation Program (BOP), Fitness-for-Duty (FFD) evaluations, Employee Assistance Programs (EAP), and Nuclear Security Culture (NSC). The human-focused portion examines behavioral and organizational indicators of opportunity, vulnerability, motivation, and crisis, demonstrating how early detection, deterrence, and response can prevent insider events. The session concludes with a case review of the Millstone Nuclear Power Station incident involving engineer George Galatis. The case illustrates how weak leadership and a poor safety culture can create conditions for failure and how a comprehensive ITM framework could have altered the outcome. The objective of this presentation is to help practitioners apply a risk-based, graded philosophy to human factors and promote a culture of accountability, communication, and resilience within nuclear organizations.

99 - GENERAL AND MISCELLANEOUS↗

Text-based Analytics for Biosurveillance

The ability to prevent, mitigate, or control a biological threat depends on how quickly the threat is identified and characterized. Ensuring the timely delivery of data and analytics is an essential aspect of providing adequate situational awareness in the face of a disease outbreak. This chapter outlines an analytic pipeline for supporting an advanced early warning system that can integrate multiple data sources and provide situational awareness of potential and occurring disease situations. The pipeline, includes real-time automated data analysis founded on natural language processing (NLP), semantic concept matching, and machine learning techniques, to enrich content with metadata related to biosurveillance. Online news articles are presented as an example use case for the pipeline, but the processes can be generalized to any textual data. In this chapter, the mechanics of a streaming pipeline are briefly discussed as well as the major steps required to provide targeted situational awareness. The text-based analytic pipeline includes various processing steps as well as identifying article relevance to biosurveillance (e.g., relevance algorithm) and article feature extraction (who, what, where, why, how, and when). The ability to prevent, mitigate, or control a biological threat depends on how quickly the threat is identified and characterized. Ensuring the timely delivery of data and analytics is an essential aspect of providing adequate situational awareness in the face of a disease outbreak. This chapter outlines an analytic pipeline for supporting an advanced early warning system that can integrate multiple data sources and provide situational awareness of potential and occurring disease situations. The pipeline, includes real-time automated data analysis founded on natural language processing (NLP), semantic concept matching, and machine learning techniques, to enrich content with metadata related to biosurveillance. Online news articles are presented as an example use case for the pipeline, but the processes can be generalized to any textual data. In this chapter, the mechanics of a streaming pipeline are briefly discussed as well as the major steps required to provide targeted situational awareness. The text-based analytic pipeline includes various processing steps as well as identifying article relevance to biosurveillance (e.g., relevance algorithm) and article feature extraction (who, what, where, why, how, and when).

Biomedical NLP↗

Modeling and Evaluation of Cyber-Attacks on Grid-Interactive Efficient Buildings

Grid-interactive efficient buildings (GEBs) are not only exposed to passive threats (e.g., physical faults) but also active threats such as cyber-attacks launched on the network-based control systems. The impact of cyber-attacks on GEB operation are not yet fully understood, especially as regards the performance of grid services. To quantify the consequences of cyber-attacks on GEBs, this paper proposes a modeling and simulation framework that includes different cyber-attack models and key performance indexes to quantify the performance of GEB operation under cyber-attacks. The framework is numerically demonstrated to model and evaluate cyber-attacks such as data intrusion attacks and Denial-of-Service attacks on a typical medium-sized office building that uses the BACnet/IP protocol for communication networks. Simulation results show that, while different types of attacks could compromise the building systems to different extents, attacks via the remote control of a chiller yield the most significant consequences on a building system’s operation, including both the building service and the grid service. It is also noted that a cyber-attack impacts the building systems during the attack period as well as the post-attack period, which suggests that both periods should be considered to fully evaluate the consequences of a cyber-attack.

Fu, Yanyang↗

Emulation and detection of physical faults and cyber-attacks on building energy systems through real-time hardware-in-the-loop experiments

The increasing use of remote or mobile access, integrated wearable technologies, data exchange, and cloud-based data analytics in modern smart buildings is steering the building industry towards open communication technologies. The increased connectivity and accessibility could lead to more cyber-attacks in smart buildings. On the other hand, physical faults (e.g., HVAC -heating, ventilation, and air-conditioning faults) may have similar adverse impacts as those from the cyber-attacks on building energy systems, such as occupant discomfort, energy wastage, and equipment downtime. However, current physical behavior-based anomaly detection methods fail to differentiate between cyber-attacks and physical faults in building energy systems. Moreover, the challenge in collecting real-world threat data with ground truth has led researchers to rely on numerical models with user-defined assumptions, which may not accurately reflect real-world conditions due to the lack of in-situ experimental datasets. To address these challenges and gaps, this paper presents a flexible hardware-in-the-loop (HIL) testbed for generating cyber-attack and physical fault datasets and demonstrating threat detection algorithms in a real building automation system (BAS) environment. This testbed combines hardware (i.e., real BAS with local HVAC controllers and a physical network) with software (i.e., high-fidelity models to represent behaviors of building envelope and HVAC energy systems), enabling emulations of realistic threats. Five HIL experiments, including one baseline without any threats, two with physical faults, and two with cyber-attacks, were conducted to generate datasets containing detailed network traffic and system states. A joint classification framework, incorporating a network analyzer and a physical HVAC fault detector, was proposed to automatically detect cyber-physical abnormalities on BAS at both the network and the physical HVAC levels. The network analyzer comprises a conditional random fields (CRF) based command validator and a statistics-based detection strategy. The fault detector employs a weather and schedule-based pattern matching and feature-based principal component analysis (WPM-FPCA) method. Evaluation of the classification using four metrics from the multi-class confusion matrix revealed an average accuracy of 90.2%, recall of 89.7%, precision of 88.5% and F1-score of 89.2%. Finally, these results demonstrate that the proposed joint classification framework can effectively differentiate between specific types of cyber-attacks (e.g., device reinitialization attack, network Denial-of-Service attack) and physical faults (e.g., air handling unit operational fault, cooling coil valve stuck) in real time for improved building energy management.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Stable-carbon isotope ratios for sourcing the nerve-agent precursor methylphosphonic dichloride and its products

The ability to connect a chemical threat agent to a specific batch of a synthetic precursor can provide a fingerprint to contribute to effective forensic investigations. Stable isotope analysis can leverage intrinsic, natural isotopic variability within the molecules of a threat agent to unlock embedded chemical fingerprints in the material. Methylphosphonic dichloride (DC) is a chemical precursor to the nerve agent sarin. It is converted to methylphosphonic difluoride (DF) as part of the sarin synthesis process. We used a suite of commercially available DC stocks to both evaluate the potential for δ13C analysis to be used as a fingerprinting tool in sarin-related investigations and to develop sample preparation techniques (using chemical hydrolysis) that can simplify isotopic analysis of DC and its synthetic products. We demonstrate that natural isotopic variability in DC results in at least three distinct, isotope-resolved clusters within the thirteen stocks we analyzed. Isotopic variability in the carbon feedstock (i.e., methanol) used for DC synthesis is likely inherited by the DC samples we measured. We demonstrate that the hydrolysis of DC and DF to methylphosphonic acid (MPA) can be used as a preparative step for isotopic analysis because the reaction does not impart a measureable isotopic fractionation. MPA is more chemically stable, less toxic, and easier to handle than DC or DF. Further, the hydrolysis method we demonstrated can be applied to a suite of other precursors or to sarin itself, thereby providing a potentially valuable forensic tool.

Moran, James J.↗

Random forest models accurately classify synthetic opioids using high-dimensionality mass spectrometry datasets

Detection of novel threat agents presents several challenges, a principle one being the development of untargeted methods to screen an increasing number of threat chemicals whose exact structures are unknown. With the use of Machine Learning (ML) tools, we can guide the development of analytical methods for broad-spectrum detection of unbounded threat chemical families in complex mixtures. Toward this goal, we used nominal mass and high-resolution mass spectrometry data for hundreds of synthetic opioids and non-opioid compounds. We tested two ML techniques, logistic regression and random forest, to develop models towards a practical, implementable method for opioid detection. We found that of these tested ML methods, random forest models resulted in the highest validation accuracy (95+%) for both nominal mass and high-resolution classification of opioids versus non-opioids, with low false positive and false negative rates. The RF models were then used to successfully predict the classification of 10 compounds—five opioids and five non-opioids not part of the training and validation analysis. This application of ML is a critical step towards the development of field-deployable nominal mass spectrometers with ML-driven analyses for classification of emergent threats.

Chemistry↗

Enhancing Network Anomaly Detection Using Graph Neural Networks

In the world of Internet of Things (IoT) networks, where devices are constantly communicating, keeping them secure from cyber threats is critical. This paper introduces a novel approach to detecting unusual and potentially harmful activities in these networks using graph neural networks (GNNs). We combine two specific types of GNNs-GraphSAGE and graph attention networks (GAT)-to create a model that understands and represents the behaviors and interactions in a network. GraphSAGE creates an embedding of network activities by examining local data interactions, while GAT directs the model's focus to the most critical interactions. By integrating these two methods in a single model that considers different types of interactions (both host and flow nodes), we aim to create a system that accurately represents the current state of a network and can also spot anomalies effectively while reducing false positives and negatives. Our innovative approach has demonstrated promising results, achieving an accuracy of 98% on the UNSW-NB15 dataset, significantly outperforming standalone GraphSAGE and GAT models. This underscores its potential as a robust framework for securing IoT networks against cyber threats and anomalies.

Marfo, William↗

Structural, Functional, and Immunogenic Insights on Cu,Zn Superoxide Dismutase Pathogenic Virulence Factors from Neisseria meningitidis and Brucella abortus

ABSTRACT Bacterial pathogens Neisseria meningitidis and Brucella abortus pose threats to human and animal health worldwide, causing meningococcal disease and brucellosis, respectively. Mortality from acute N. meningitidis infections remains high despite antibiotics, and brucellosis presents alimentary and health consequences. Superoxide dismutases are master regulators of reactive oxygen and general pathogenicity factors and are therefore therapeutic targets. Cu,Zn superoxide dismutases (SODs) localized to the periplasm promote survival by detoxifying superoxide radicals generated by major host antimicrobial immune responses. We discovered that passive immunization with an antibody directed at N. meningitidis SOD (NmSOD) was protective in a mouse infection model. To define the relevant atomic details and solution assembly states of this important virulence factor, we report high-resolution and X-ray scattering analyses of NmSOD and of SOD from B. abortus (BaSOD). The NmSOD structures revealed an auxiliary tetrahedral Cu-binding site bridging the dimer interface; mutational analyses suggested that this metal site contributes to protein stability, with implications for bacterial defense mechanisms. Biochemical and structural analyses informed us about electrostatic substrate guidance, dimer assembly, and an exposed C-terminal epitope in the NmSOD dimer. In contrast, the monomeric BaSOD structure provided insights for extending immunogenic peptide epitopes derived from the protein. These collective results reveal unique contributions of SOD to pathogenic virulence, refine predictive motifs for distinguishing SOD classes, and suggest general targets for antibacterial immune responses. The identified functional contributions, motifs, and targets distinguishing bacterial and eukaryotic SOD assemblies presented here provide a foundation for efforts to develop SOD-specific inhibitors of or vaccines against these harmful pathogens. IMPORTANCE By protecting microbes against reactive oxygen insults, SODs aid survival of many bacteria within their hosts. Despite the ubiquity and conservation of these key enzymes, notable species-specific differences relevant to pathogenesis remain undefined. To probe mechanisms that govern the functioning of Neisseria meningitidis and Brucella abortus SODs, we used X-ray structures, enzymology, modeling, and murine infection experiments. We identified virulence determinants common to the two homologs, assembly differences, and a unique metal reservoir within meningococcal SOD that stabilizes the enzyme and may provide a safeguard against copper toxicity. The insights reported here provide a rationale and a basis for SOD-specific drug design and an extension of immunogen design to target two important pathogens that continue to pose global health threats.

38 RADIATION CHEMISTRY, RADIOCHEMISTRY, AND NUCLEA↗

Random forest models accurately classify synthetic opioids using high-dimensionality mass spectrometry datasets

Detection of novel threat agents presents several challenges, a principle one being the development of untargeted methods to screen an increasing number of threat chemicals whose exact structures are unknown. With the use of Machine Learning (ML) tools, we can guide the development of analytical methods for broad-spectrum detection of unbounded threat chemical families in complex mixtures. Toward this goal, we used nominal mass and high-resolution mass spectrometry data for hundreds of synthetic opioids and non-opioid compounds. We tested two ML techniques, logistic regression and random forest, to develop models towards a practical, implementable method for opioid detection. We found that of these tested ML methods, random forest models resulted in the highest validation accuracy (95+%) for both nominal mass and high-resolution classification of opioids versus non-opioids, with low false positive and false negative rates. The RF models were then used to successfully predict the classification of 10 compounds—five opioids and five non-opioids not part of the training and validation analysis. This application of ML is a critical step towards the development of field-deployable nominal mass spectrometers with ML-driven analyses for classification of emergent threats.

Arasteh, Kourosh [Lawrence Livermore National Labo↗

Severe Convective Storms in Limited Instability Organized by Pattern and Distribution

Abstract Severe convection occurring in high-shear, low-CAPE (HSLC) environments is a common cool-season threat in the southeastern United States. Previous studies of HSLC convection document the increased operational challenges that these environments present compared to their high-CAPE counterparts, corresponding to higher false-alarm ratios and lower probability of detection for severe watches and warnings. These environments can exhibit rapid destabilization in the hours prior to convection, sometimes associated with the release of potential instability. Here, we use self-organizing maps (SOMs) to objectively identify environmental patterns accompanying HSLC cool-season severe events and associate them with variations in severe weather frequency and distribution. Large-scale patterns exhibit modest variation within the HSLC subclass, featuring strong surface cyclones accompanied by vigorous upper-tropospheric troughs and northward-extending regions of instability, consistent with prior studies. In most patterns, severe weather occurs immediately ahead of a cold front. Other convective ingredients, such as lower-tropospheric vertical wind shear, near-surface equivalent potential temperature (θ e ) advection, and the release of potential instability, varied more significantly across patterns. No single variable used to train SOMs consistently demonstrated differences in the distribution of severe weather occurrence across patterns. Comparison of SOMs based on upper and lower quartiles of severe occurrence demonstrated that the release of potential instability was most consistently associated with higher-impact events in comparison to other convective ingredients. Overall, we find that previously developed HSLC composite parameters reasonably identify high-impact HSLC events. Significance Statement Even when atmospheric instability is not optimal for severe convective storms, in some situations they can still occur, presenting increased challenges to forecasters. These marginal environments may occur at night or during the cool season, when people are less attuned to severe weather threats. Here, we use a sorting algorithm to classify different weather patterns accompanying such storms, and we distinguish which specific patterns and weather system features are most strongly associated with severe storms. Our goals are to increase situational awareness for forecasters and to improve understanding of the processes leading to severe convection in marginal environments.

Meteorology & Atmospheric Sciences↗

Performing Numerical Analysis of Cybersecurity Options Using Dynamic Risk Analysis Tool EMRALD

Cyberattacks can have many different attack paths, durations, and goals. There are also many different mitigation options involving hardware, software, and/or humans. Considering a cyber threat should involve defense-in-depth methods and a quantitative or numerical evaluation of overall effectiveness against dynamic, time-dependent attacks to make cost and risk-informed decisions. Typical cyberattack modeling methods only provide a qualitative evaluation. The main areas of cybersecurity are confidentiality, integrity, and availability. For companies with cyber-physical systems such as advanced nuclear reactors, cyber-related safety is a requirement set by North American Electric Reliability and the U.S. Nuclear Regulatory Commission. They are also concerned about availability or reliability as a business case. As cyber threats are evolving to a business-for-hire structure, more attacks may focus on disrupting business success and reliability, causing financial and economic stability risk. Companies want to know business reliability and recovery from those threats, and that requires modeling physical behavior of the targets. Dynamic-state-based and Markov-based modeling provides a method for better cyber scenario modeling with different tools having issues such as state-base explosion. Dynamic modeling enables time and conditional features not found in other numerical evaluation methods. EMRALD (Event Modeling Risk Assessment using Lined Diagrams) is a dynamic risk analysis modeling and simulation tool and has features that reduce modeling issues. It has been used to model different time-dependent events including plant behavior and operator procedures. As a general modeling tool, EMRALD can also be used to model cyberattack scenarios with varying mitigation options and quantify effectiveness, producing numerical data for risk-informed decisions. This paper uses EMRALD to demonstrate that dynamic numerical risk analysis can be used for cyber threat modeling to provide insights for design decision-making and optimize defense strategies. Keywords: cyber modeling; cyber-physical systems; numerical cyber modeling

97 - MATHEMATICS AND COMPUTING↗