Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “test approach”

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

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

At least 109 records · Page 6
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

cTULIP: application of a human-based RNA-seq primary tumor classification tool for cross-species primary tumor classification in canine

The domestic dog, Canis familiaris, is quickly gaining traction as an advantageous model for use in the study of cancer, one of the leading causes of death worldwide. Naturally occurring canine cancers share clinical, histological, and molecular characteristics with the corresponding human diseases. In this study, we take a deep-learning approach to test how similar the gene expression profile of canine glioma and bladder cancer (BLCA) tumors are to the corresponding human tumors. We likewise develop a tool for identifying misclassified or outlier samples in large canine oncological datasets, analogous to that which was developed for human datasets. We test a number of machine learning algorithms and found that a convolutional neural network outperformed logistic regression and random forest approaches. We use a recently developed RNA-seq-based convolutional neural network, TULIP, to test the robustness of a human-data-trained primary tumor classification tool on cross-species primary tumor prediction. Our study ultimately highlights the molecular similarities between canine and human BLCA and glioma tumors, showing that protein-coding one-to-one homologs shared between humans and canines, are sufficient to distinguish between BLCA and gliomas. The results of this study indicate that using protein-coding one-to-one homologs as the features in the input layer of TULIP performs good primary tumor prediction in both humans and canines. Furthermore, our analysis shows that our selected features also contain the majority of features with known clinical relevance in BLCA and gliomas. Our success in using a human-data-trained model for cross-species primary tumor prediction also sheds light on the conservation of oncological pathways in humans and canines, further underscoring the importance of the canine model system in the study of human disease.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2020

Structural Optimization (Final Technical Report V.2.0)

The overall Project (Project) objective is to materially decrease the leveled cost of energy (LCOE) of the Columbia Power (CPwr) StingRAY utility-scale wave energy converter (WEC). This will be achieved by reducing structural material and manufacturing costs and increasing energy output. In this Project, improving the overall Power-to-Weight ratio (PWR) is accomplished through lowering design margins - allowing for weight reduction and more efficient, cost-effective WEC manufacturing and assembly - and by optimizing mass-related WEC performance parameters, such as center of gravity and system inertia. A mixed materials approach to further structural optimization was developed under this Project and validated with extensive laboratory structural testing. This approach substitutes fiber-reinforced plastic (FRP) for steel where appropriate. The benefits of steel are maintained where most useful, for instance at structural joints where the stiffness of steel is required, and the complex geometry is more readily fabricated with steel. However, there are structural spans whose simple shapes are readily fabricated with mandrel-wound FRP and where significant cost and weight savings can be found. An adhesive, double lap shear joint is used to join the FRP and steel subcomponents.

16 TIDAL AND WAVE POWER↗
DOE OSTI2025

Predicting the von Neumann entanglement entropy using a graph neural network

Calculating the von Neumann entanglement entropy from experimental data is challenging due to its dependence on the complete wavefunction, forcing reliance on approximations such as classical mutual information (MI). We propose a machine learning approach using a graph neural network to predict the von Neumann entropy directly from experimentally accessible bitstrings. We test this approach on a Rydberg ladder system and achieve a mean absolute error of $3.6\,\times 10^{-3}$ when evaluating within the training range on a dataset with entropy values ranging from 0 to 1.9. The model achieves a mean absolute percentage error of 1.44% and outperforms MI-based bounds. When tested beyond the training range, the model maintains reasonable accuracy. Furthermore, we demonstrate that fine-tuning the model with small datasets significantly improves performance on data outside the original training range.

graph neural networks↗
DOE OSTI2021

Adiabatic quantum linear regression

Abstract A major challenge in machine learning is the computational expense of training these models. Model training can be viewed as a form of optimization used to fit a machine learning model to a set of data, which can take up significant amount of time on classical computers. Adiabatic quantum computers have been shown to excel at solving optimization problems, and therefore, we believe, present a promising alternative to improve machine learning training times. In this paper, we present an adiabatic quantum computing approach for training a linear regression model. In order to do this, we formulate the regression problem as a quadratic unconstrained binary optimization (QUBO) problem. We analyze our quantum approach theoretically, test it on the D-Wave adiabatic quantum computer and compare its performance to a classical approach that uses the Scikit-learn library in Python. Our analysis shows that the quantum approach attains up to $${2.8 \times }$$ 2.8 × speedup over the classical approach on larger datasets, and performs at par with the classical approach on the regression error metric. The quantum approach used the D-Wave 2000Q adiabatic quantum computer, whereas the classical approach used a desktop workstation with an 8-core Intel i9 processor. As such, the results obtained in this work must be interpreted within the context of the specific hardware and software implementations of these machines.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Detection of Stealthy False Data Injection Attacks in Unobservable Distribution Networks

In this paper, a composite scheme is proposed for detecting stealthy data manipulation attacks on distribution system which is unobservable with standard least squares based state estimators. This technique has three stages where the process of data imputation, voltage phasor estimation and the bad data detection are carried out in a systematic manner. The proposed approach is then integrated with moving target defense strategies which perturbs the network parameters to reveal stealthy false data injection attacks. The proposed approach is tested is validated on a three-phase, unbalanced 37-node distribution system and its results are presented. It is shown that the proposed approach has the ability to accurately detect the presence of FDI attacks using limited measurements (i.e., the test system is unobservable).

Rajasekaran, James K.↗
DOE OSTI2021

Zn uptake by illite and argillaceous rocks

The uptake of Zn by Illite du Puy (IdP) and two argillaceous rocks, Opalinus Clay (OPA) and Boda Claystone (BODA) was investigated. The uptake of Zn by illite was studied in 0.1 M NaCl at near-neutral pH and Zn loadings varying from 2.1 to 42 mmol/kg. The Zn uptake by the two argillaceous rock systems was carried out in 0.1 M NaCl at pH 7.2 and in their respective porewaters at pH 8.0 to evaluate the influence of porewater composition. The Zn loadings varied from 1.8 to 86 mmol/kg and 1.7 to 60 mmol/kg for OPA and BODA, respectively. The simplified “bottom-up approach” to predict the uptake of Zn to natural rocks with different clay mineral contents and porewater compositions was tested. This approach was found to be applicable in the argillaceous rock systems at low Zn equilibrium ([Zn] eql ) concentrations below 10 -7 M. However, at [Zn]eql above 10 -7 M, the model calculations clearly underpredicted the sorption data. The extended X-ray absorption fine structure (EXAFS) results support findings based on wet chemistry calculations using the 2 site protolysis non-electrostatic surface complexation and cation exchange (2SPNE SC/CE) sorption model. In the IdP system with prolonged reaction times of up to two years, formation of Zn precipitates was not observed. This indicates that the Zn surface complexes are stable over a time period of two years, and dissolution and recrystallization processes might not play a significant role in the Zn-IdP uptake process. In addition, the EXAFS results on illite corroborate previous findings on the existence of weak and strong sites types in dioctahedral clay minerals, as assumed in the 2SPNE SC/CE model. In the OPA and BODA systems, precipitation processes could start to begin at metal loadings of 16 mmol/kg and 24 mmol/kg, respectively. The differences between argillaceous rock samples prepared in NaCl and porewater are modest indicating that in the case of Zn the clay minerals play a predominant role in the uptake process, and that the effects originating from the porewater are minor. Furthermore, despite the differences in mineralogy, both argillaceous rocks show the same uptake behaviour, indicating that the clay minerals are predominantly responsible for the sorption at low and the precipitation processes at higher Zn concentrations. The study demonstrates how results on “pure” systems such as IdP can be transferred to predict the uptake by argillaceous rocks at low sorbate concentrations.

58 GEOSCIENCES↗
DOE OSTI2025

Galaxy cluster profiles: a Gaussian mixture model approach to halo miscentering

Measurements of the galaxy density and weak-lensing profiles of galaxy clusters typically rely on an assumed cluster center, which is taken to be the brightest cluster galaxy or other proxies for the true halo center defined as the minimum in the potential well. Departure of the assumed cluster center from the true halo center bias the resultant profile measurements, an effect known as miscentering bias. Currently, miscentering is typically modeled in stacked profiles of clusters with a two parameter model. We use an alternate approach in which the profiles of individual clusters are used with the corresponding likelihood computed using a Gaussian mixture model. We test the approach using halos and the corresponding subhalo profiles from the IllustrisTNG hydrodynamic simulations. We obtain significantly improved estimates of the miscentering parameters for both 3D and projected 2D profiles relevant for imaging surveys. We discuss applications to upcoming cosmological surveys. Our Python package for the Gaussian mixture model is publicly available at https://github.com/KyleMiller1/Halo-Miscentering-Mixture-Model.

Bayesian reasoning↗
DOE OSTI2020

Prediction of condition-specific regulatory genes using machine learning

Recent advances in genomic technologies have generated data on large-scale protein–DNA interactions and open chromatin regions for many eukaryotic species. How to identify condition-specific functions of transcription factors using these data has become a major challenge in genomic research. To solve this problem, we have developed a method called ConSReg, which provides a novel approach to integrate regulatory genomic data into predictive machine learning models of key regulatory genes. Using Arabidopsis as a model system, we tested our approach to identify regulatory genes in data sets from single cell gene expression and from abiotic stress treatments. Our results showed that ConSReg accurately predicted transcription factors that regulate differentially expressed genes with an average auROC of 0.84, which is 23.5–25% better than enrichment-based approaches. To further validate the performance of ConSReg, we analyzed an independent data set related to plant nitrogen responses. ConSReg provided better rankings of the correct transcription factors in 61.7% of cases, which is three times better than other plant tools. We applied ConSReg to Arabidopsis single cell RNA-seq data, successfully identifying candidate regulatory genes that control cell wall formation. Our methods provide a new approach to define candidate regulatory genes using integrated genomic data in plants.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

Draft Rules for Alloy 617 Creep-Fatigue Design Using an EPP+SMT Approach

The report describes an improved new creep-fatigue design method for structural components in high temperature nuclear service. The new method uses an integrated elastic-perfectly plastic (EPP) analysis and Simplified Model Test (SMT) approach that greatly simplifies the design process by avoiding the separate evaluation of creep and fatigue damages and eliminating the requirement of stress classification. The report includes draft design rules, presented in a format compatible with an ASME nuclear Code Case, as well as a commentary on the rules and the validation of the rules using pressurized SMT (p-SMT) test data. Areas of potential improvement have been identified for further evaluation. The report also develops EPP+SMT design charts for Alloy 617 at temperatures between 800°C and 950°C by extrapolating high strain range, short hold test data to low strain range and long hold times, typical for structural components in high temperature nuclear service. The design charts are currently limited to 10,000 cycles to avoid excessive extrapolation outside the test database. More robust alternate extrapolation procedures will be considered in future work. The report includes several verification problems comparing the new EPP+SMT design method with current ASME creep-fatigue design methods. The comparison demonstrates that the EPP+SMT method significantly reduces the over-conservatism in current methods. Finally, the report includes a fully-documented sample problem with multiple load cases detailing the application of proposed EPP+SMT design rules.

36 MATERIALS SCIENCE↗
DOE OSTI2024

SPIDARman: System-Level Physics-Informed Detection of Anomalies in Reactor Collected Data Considering Human Errors

In nuclear power plants (NPPs), anomalies arising from sensors or human errors (HEs) can undermine the performance and reliability of plant operations. Anomaly detection models can be employed to detect sensor errors and HEs. Additionally, physics-informed machine learning models can utilize the known physics of the system, as described by mathematical equations, to ensure that sensor values are consistent with physical laws. Hence, we propose SPIDARman: System-level Physics-Informed Detection of Anomalies in Reactor Collected Data Considering Human Errors, a holistic physics-informed anomaly detection approach based on generative adversarial networks (GANs) to detect anomalies in both automatically collected sensor data and manually collected surveillance data. Here we test our approach on data collected from a flow loop testbed, showcasing its potential to detect anomalies. Results demonstrate that the proposed model performs better than the baseline GAN-based models in detecting sensor and surveillance anomalies, suggesting the potential of physics-informed anomaly detection GAN models in NPPs.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2026

Identification and mitigation of memory block timing issue in ITk ABCStar during ASIC production

The ABCStar is a mixed-signal front-end readout ASIC for the strips sensor portion of the ATLAS ITk detector being developed as part of the High-Luminosity LHC upgrade. In pre-production testing, a subtle design flaw was uncovered in the ABCStar that was reducing wafer yields in some manufactured lots from the expected 90% to as low as 2%. The root cause was determined to be a timing issue in the logic synthesized to control previously silicon proven memory blocks re-used for this ASIC. The solutions proposed included manufacturing process changes by the wafer foundry, changes to the operating parameters for the ABCStar in the detector, and the possibility that a redesign might be required. The two mitigation efforts were undertaken in parallel, with the process modification route a less desirable solution since already manufactured wafers would need to be scrapped in favour of the new ones. Based on a knowledge of the existing process, and testing done on the worst performing wafers, it was proposed that raising the core operating voltage of the ABCStar from 1.20V to 1.25V could address the timing issue by sufficiently speeding up its transistors. An extensive testing program that included the effects of temperature and radiation expected over the lifetime of the ITk detector was conducted to validate that approach. Those tests and studies proved that even the worst performing wafers would have yields over 80% with the 1.25V core voltage, and neither the modified process nor redesign would be required for ensuring reliable operation of the ITk. Based on testing, a further timing mitigation was implemented to provide an additional margin of reliability by increasing the duty cycle of the clock to the ABCStar. Testing of all ABCStar wafers has been completed and the production of the detector modules using these ASICs is now well underway as a result of the efforts detailed herein.

FOS: Physical sciences↗
DOE OSTI2022

One-Loop effective action approach to quantum MHV theory

It is well known that the MHV action, i.e. the action containing all the maximally helicity violating vertices, is alone not sufficient for loop computations. In order to develop loop contributions systematically and to ensure that there are no missing terms, we propose to formulate the quantum MHV action via one-loop effective action approach. The quadratic field fluctuations in the light cone Yang-Mills theory are explicitly integrated, followed by the classical canonical field transformation. We test the approach by calculating one loop (++++) and (+++) amplitudes, i.e. amplitudes that cannot be calculated from ordinary MHV action. Such an approach can be further used to unambiguously define loop corrections in other theories related to Yang-Mills theory by field transformations.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2023

Differential early–life survival underlies the adaptive significance of temperature–dependent sex determination in a long–lived reptile

1. Many ectotherms rely on temperature cues experienced during development to determine offspring sex. The first descriptions of temperature-dependent sex determination (TSD) were made over 50 years ago, yet an understanding of its adaptive significance remains elusive, especially in long-lived taxa. 2. One novel hypothesis predicts that TSD should be evolutionarily favoured when two criteria are met—(a) incubation temperature influences annual juvenile survival and (b) sexes mature at different ages. Under these conditions, a sex-dependent effect of incubation temperature on offspring fitness arises through differences in age at sexual maturity, with the sex that matures later benefiting disproportionately from temperatures that promote juvenile survival. 3. The American alligator (Alligator mississippiensis) serves as an insightful model in which to test this hypothesis, as males begin reproducing nearly a decade after females. Here, through a combination of artificial incubation experiments and mark-recapture approaches, we test the specific predictions of the survival-to-maturity hypothesis for the adaptive value of TSD by disentangling the effects of incubation temperature and sex on annual survival of alligator hatchlings across two geographically distinct sites. 4. Hatchlings incubated at male-promoting temperatures (MPTs) consistently exhibited higher survival compared to those incubated at female-promoting temperatures. This pattern appears independent of hatchling sex, as females produced from hormone manipulation at MPT exhibit similar survival to their male counterparts. 5. Additional experiments show that incubation temperature may affect early-life survival primarily by affecting the efficiency with which maternally transferred energy resources are used during development. 6. Results from this study provide the first explicit empirical support for the adaptive value of TSD in a crocodilian and point to developmental energetics as a potential unifying mechanism underlying persistent survival consequences of incubation temperature.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2020

Analysis of Air-Purifying Respirator (APR) Cartridge Performance Testing on Hanford Tanks SX-101 and SX-104 (Vol.1)

Washington River Protection Solutions (WRPS) conducted tests on two types of chemical cartridges for use in air-purifying respirators to determine the period of time that the cartridges would provide adequate performance1 to protect workers when exposed to a mixture of Chemicals of Potential Concern (COPC) from vapors exiting the headspaces of Hanford tanks SX-101 and SX-104. The Occupational Safety and Health Administration (OSHA) considers cartridge testing a valid approach for establishing cartridge change schedules.2 Testing is commonly applied in situations where mixtures of COPCs exist and where other approaches such as manufacturer recommendations and modeling are less reliable. The tests were designed and conducted to ensure measurement and/or control of the key variables OSHA identified as important to estimate the cartridge service life, including temperature, humidity, COPC concentration, breathing rate, and cartridge adsorption capacity. Testing was conducted from June 16-17, 2017, and June 23-24, 2017, using headspace vapors from Hanford tank SX-104 and Hanford tank SX-101, respectively. Headspace vapors under static conditions were fed to a respirator cartridge test stand developed by WRPS in collaboration with HiLine Engineering (Richland, Washington). Multipurpose respirator cartridges, SCOTT 7422-SD1 and SCOTT 7422-SC1 (SCOTT Safety, Monroe, North Carolina), were assessed on separate days. Sample media (sorbent tubes) were used to collect samples of the vapor stream entering and exiting the respirator cartridges, and subsequently were analyzed for COPC concentrations. Pacific Northwest National Laboratory was tasked with conducting an independent analysis of the analytical results and making recommendations based on the results for respiratory cartridge performance and service life.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2025

Accelerated Fuel Qualification of Fast Modular Reactor Fuel in a Thermal Reactor: Modeling and Simulation Paired with Irradiation Testing

The accelerated fuel qualification (AFQ) methodology is applied by simulating accelerated fuel tests of the General Atomics Electromagnetic Systems’ fuel system for its 44-MW(electric) gas-cooled, fast-spectrum fast modular reactor (FMR). This fuel is comprised of UO 2 pellets in SiGA® cladding, a silicon carbide ceramic matrix composite. Fast reactors, like the FMR, offer many benefits, including high fuel utilization and flexibility, but may require a lengthy material design process if tests are performed using fast neutron irradiation alone. A thermal neutron irradiation can instead be used to rapidly test how well key components of the current material models extend to high burnup. Thermal neutrons produce a different radial power distribution within the pin than fast neutrons. However, the temperature and burnup values for the two neutron types are comparable, and the differences between the simulated fuel responses are relatively small, demonstrating the weak sensitivity of the physics-based fuel model calculations on the neutron type and the irradiation rate. Furthermore, the deformation of the SiGA cladding saturates after about 1 displacement per atom for both neutron spectra. In an accelerated fuel test, the irradiation time required to reach the target fuel burnup can be reduced by a factor of 3 by using a small rodlet with a 45% smaller pellet diameter while maintaining the same linear power. Therefore, the time for data collection up to high burnup can be significantly reduced while maintaining the same temperature profile, which largely determines the material response. Tests of fuel rodlets of standard and compact size will be carried out in the Idaho National Laboratory’s Advanced Test Reactor (ATR), including full size and compact rodlets with varying gap sizes. By applying physics-based mechanistic modeling and simulation in accordance with the AFQ methodology, this type of compact rodlet testing in a thermal test reactor captures the necessary phenomena to test fuel material models up to high burnup and to simulate the expected impact of fast neutron radiation on the fuel in FMR operations. Finally, this approach to testing fast reactor fuels in existing thermal test reactors, paired with advanced physics-based mechanistic modeling and simulation, is expected to be applicable to a range of advanced fuels and will decrease the overall fuel qualification timeframe from decades to years.

Advanced test reactor (ATR)↗
DOE OSTI2022

Computational Imaging for Intelligence in Highly Scattering Aerosols (Final Report)

Natural and man-made degraded visual environments pose major threats to national security. The random scattering and absorption of light by tiny particles suspended in the air reduces situational awareness and causes unacceptable down-time for critical systems and operations. To improve the situation, we have developed several approaches to interpret the information contained within scattered light to enhance sensing and imaging in scattering media. These approaches were tested at the Sandia National Laboratory Fog Chamber facility and with tabletop fog chambers. Computationally efficient light transport models were developed and leveraged for computational sensing. The models are based on a weak angular dependence approximation to the Boltzmann or radiative transfer equation that appears to be applicable in both the moderate and highly scattering regimes. After the new model was experimentally validated, statistical approaches for detection, localization, and imaging of objects hidden in fog were developed and demonstrated. A binary hypothesis test and the Neyman-Pearson lemma provided the highest theoretically possible probability of detection for a specified false alarm rate and signal-to-noise ratio. Maximum likelihood estimation allowed estimation of the fog optical properties as well as the position, size, and reflection coefficient of an object in fog. A computational dehazing approach was implemented to reduce the effects of scatter on images, making object features more readily discernible. We have developed, characterized, and deployed a new Tabletop Fog Chamber capable of repeatably generating multiple unique fog-analogues for optical testing in degraded visual environments. We characterized this chamber using both optical and microphysical techniques. In doing so we have explored the ability of droplet nucleation theory to describe the aerosols generated within the chamber, as well as Mie scattering theory to describe the attenuation of light by said aerosols, and correlated the aerosol microphysics to optical properties such as transmission and meteorological optical range (MOR). This chamber has proved highly valuable and has supported multiple efforts inclusive to and exclusive of this LDRD project to test optics in degraded visual environments. Circularly polarized light has been found to maintain its polarization state better than linearly polarized light when propagating through fog. This was demonstrated experimentally in both the visible and short-wave infrared (SWIR) by imaging targets made of different commercially available retroreflective films. It was found that active circularly polarized imaging can increase contrast and range compared to linearly polarized imaging. We have completed an initial investigation of the capability for machine learning methods to reduce the effects of light scattering when imaging through fog. Previously acquired experimental long-wave images were used to train an autoencoder denoising architecture. Overfitting was found to be a problem because of lack of variability in the object type in this data set. The lessons learned were used to collect a well labeled dataset with much more variability using the Tabletop Fog Chamber that will be available for future studies. We have developed several new sensing methods using speckle intensity correlations. First, the ability to image moving objects in fog was shown, establishing that our unique speckle imaging method can be implemented in dynamic scattering media. Second, the speckle decorrelation over time was found to be sensitive to fog composition, implying extensions to fog characterization. Third, the ability to distinguish macroscopically identical objects on a far-subwavelength scale was demonstrated, suggesting numerous applications ranging from nanoscale defect detection to security. Fourth, we have shown the capability to simultaneously image and localize hidden objects, allowing the speckle imaging method to be effective without prior object positional information. Finally, an interferometric effect was presented that illustrates a new approach for analyzing speckle intensity correlations that may lead to more effective ways to localize and image moving objects. All of these results represent significant developments that challenge the limits of the application of speckle imaging and open important application spaces. A theory was developed and simulations were performed to assess the potential transverse resolution benefit of relative motion in structured illumination for radar systems. Results for a simplified radar system model indicate that significant resolution benefits are possible using data from scanning a structured beam over the target, with the use of appropriate signal processing.

58 GEOSCIENCES↗
DOE OSTI2023

RX-ADS: Interpretable Anomaly Detection Using Adversarial ML for Electric Vehicle CAN Data

Recent year has brought considerable advancements in Electric Vehicles (EVs) and associated infrastructures/ communications. Intrusion Detection Systems (IDS) are widely deployed for anomaly detection in such critical infrastructures. This paper presents an Interpretable Anomaly Detection System (RX-ADS) for intrusion detection in CAN protocol communication in EVs. Contributions include: 1) window based feature extraction method; 2) deep Autoencoder based anomaly detection method; and 3) adversarial machine learning based explanation generation methodology. The presented approach was tested on two benchmark CAN datasets: OTIDS and Car Hacking. The anomaly detection performance of RX-ADS was compared against the state-of-the-art approaches on these datasets: HIDS and GIDS. The RX-ADS approach presented performance comparable to the HIDS approach (OTIDS dataset) and has outperformed HIDS and GIDS approaches (Car Hacking dataset). Further, the proposed approach was able to generate explanations for detected abnormal behaviors arising from various intrusions. Furthermore, these explanations were later validated by information used by domain experts to detect anomalies. Other advantages of RX-ADS include: 1) the method can be trained on unlabeled data; 2) explanations help experts in understanding anomalies and root course analysis, and also help with AI model debugging and diagnostics, ultimately improving user trust in AI systems.

42 ENGINEERING↗