Search NASA⌕ Search

SEARCH · Search NASA

Results for “Complex Network Data”

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 1,441 records · Page 80

SBML Level 3: an extensible format for the exchange and reuse of biological models

Systems biology has experienced dramatic growth in the number, size, and complexity of computational models. To reproduce simulation results and reuse models, researchers must exchange unambiguous model descriptions. We review the latest edition of the Systems Biology Markup Language (SBML), a format designed for this purpose. A community of modelers and software authors developed SBML Level 3 over the past decade. Its modular form consists of a core suited to representing reaction-based models and packages that extend the core with features suited to other model types including constraint-based models, reaction-diffusion models, logical network models, and rule-based models. The format leverages two decades of SBML and a rich software ecosystem that transformed how systems biologists build and interact with models. More recently, the rise of multiscale models of whole cells and organs, and new data sources such as single-cell measurements and live imaging, has precipitated new ways of integrating data with models. We provide our perspectives on the challenges presented by these developments and how SBML Level 3 provides the foundation needed to support this evolution.

59 BASIC BIOLOGICAL SCIENCES↗

PMU Data Quality and Sensor Health Monitoring

Phasor Measurement Units (PMUs) play a critical role in the evolution of the electric power industry by providing high-precision, real-time monitoring of essential power system metrics. However, effectively detecting abnormalities and critical events from PMU data is a complex task, complicated by intricate temporal patterns, a scarcity of labeled data for training algo- rithms, and constraints on online computational power. In this study, we apply TranAD, an innovative algorithm that combines transformer architectures with the refinement of adversarial learning, to both synthetic and real-world PMU datasets for developing a data quality and sensor online health monitoring platform for utilities. Our findings reveal that TranAD not only provides efficient detection and localization but also enhances the detail with which abnormalities are detected, marking a a significant step forward in the field of clean data acquisition processes for power system monitoring

deep neural network, machine learning (ML)↗

Efficient estimators for likelihood ratio sensitivity indices of complex stochastic dynamics

We demonstrate that centered likelihood ratio estimators for the sensitivity indices of complex stochastic dynamics are highly efficient with low, constant in time variance and consequently they are suitable for sensitivity analysis in long-time and steady-state regimes. These estimators rely on a new covariance formulation of the likelihood ratio that includes as a submatrix a Fisher information matrix for stochastic dynamics and can also be used for fast screening of insensitive parameters and parameter combinations. The proposed methods are applicable to broad classes of stochastic dynamics such as chemical reaction networks, Langevin-type equations and stochastic models in finance, including systems with a high dimensional parameter space and/or disparate decorrelation times between different observables. Furthermore, they are simple to implement as a standard observable in any existing simulation algorithm without additional modifications.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Dereplication, Aggregation and Scoring Tool (DAS Tool) v1.0

Communities of uncultivated microbes are critical to ecosystem function and microorganism health, and a key objective of metagenomic studies is to analyze organism-specific metabolic pathways and reconstruct community interaction networks. This requires accurate assignment of genes to genomes, yet existing binning methods often fail to predict a reasonable number of genomes and report many bins of low quality and completeness. Furthermore, the performance of existing algorithms varies between samples and biotypes. Here, we present a dereplication, aggregation and scoring strategy, DAS Tool, that combines the strengths of a flexible set of established binning algorithms. DAS Tools applied to a constructed community generated more accurate bins than any automated method. Further, when applied to samples of different complexity, including soil, natural oil seeps, and the human gut, DAS Tool recovered substantially more near-complete genomes than any single binning method alone. Included were three genomes from a novel lineage . The ability to reconstruct many near-complete genomes from metagenomics data will greatly advance genome-centric analyses of ecosystems.

SIEBER, CHRISTIAN↗

Complex Fault Geometry of the 2020 M ww 6.5 Monte Cristo Range, Nevada, Earthquake Sequence

On 15 May 2020 an M ww 6.5 earthquake occurred beneath the Monte Cristo Range in the Mina Deflection region of western Nevada. Rapid deployment of eight temporary seismic stations enabled detailed analysis of its productive and slowly decaying aftershock sequence (p=0.8), which included ~18,000 autodetected events in 3.5 months. Double-difference, waveform-based relative relocation of 16,714 earthquakes reveals a complex network of faults, many of which cross the inferred 35-km-long east–northeast-striking, left-lateral mainshock rupture. Seismicity aligns with left-lateral, right-lateral, and normal mechanism moment tensors of 128 of the largest earthquakes. The mainshock occurred near the middle of the aftershock zone at the intersection of two distinct zones of seismicity. In the western section, numerous subparallel, shallow, north-northeast-striking faults form a broad flower-structure-like fault mesh that coalesces at depth into a near-vertical, left-lateral fault. We infer the near-vertical fault to be a region of significant slip in the mainshock and an eastward extension of the left-lateral Candelaria fault. Near the mainshock hypocenter, seismicity occurs on a northeast-striking, west-dipping structure that extends north from the eastern Columbus Salt Marsh normal fault. Together, these two intersecting structures bound the Columbus Salt Marsh tectonic basin. East of this intersection and the mainshock hypocenter, seismicity occurs in a narrow, near-vertical, east-northeast-striking fault zone through to its eastern terminus. At the eastern end, the aftershock zone broadens and extends northwest toward the southern extension of the northwest-striking, right-lateral Petrified Springs fault system. The eastern section hosts significantly fewer aftershocks than the western section, but has more moment release. We infer that shallow aftershocks throughout the system highlight fault-fracture meshes that connect mapped fault systems at depth. Here, comparing earthquake data with surface ruptures and a simple geodetic fault model sheds light on the complexity of this recent M 6.5 Walker Lane earthquake.

58 GEOSCIENCES↗

Bulk Glassy GeTe 2 : A Missing Member of the Tetrahedral GeX 2 Family and a Precursor for the Next Generation of Phase-Change Materials

Vitreous germanium disulfide GeS 2 and diselenide GeSe 2 belong to canonical chalcogenide glasses extensively studied over the past half century. Their high-temperature orthorhombic polymorphs are congruently melting compounds, and the tetrahedral crystal and glass structure is largely preserved in the melt. In contrast, the ditelluride counterpart is absent in the Ge-Te phase diagram, which shows only a single compound, monotelluride GeTe. Phase-change materials based on GeTe have become a technologically important class of solids, and their structure and properties are also widely studied. Surprisingly, very scarce information is available for alloys having GeTe 2 stoichiometry. Using a fast quenching procedure in silica capillaries, high-energy X-ray diffraction, and Raman spectroscopy supported by first-principles simulations, we show that bulk glassy GeTe 2 differs substantially from the lighter GeX 2 members, revealing 46% of trigonal germanium, 31% of three-fold coordinated tellurium, and only 20% of edge-sharing tetrahedra or pyramids. The fraction of homopolar Ge-Ge bonds is low; however, the population of dominant Te-Te dimers and Te n oligomers, n <= 10, appears to be significant. The complex structural and chemical topology of g-GeTe 2 is directly related to the thermodynamic metastability of germanium ditelluride, schematically represented by the following reaction: GeTe 2 $\rightleftarrows$ GeTe + Te. Disproportionation is complete above liquidus in the temperature range of semiconductor-metal transition, and the dense metallic GeTe 2 liquid, mostly consisting of five-fold coordinated Ge species, exhibits high fluidity, strong fragility (m = 99 ± 5), and presumably a fast structural transformation rate combined with low atomic mobility in the vicinity of the glass transition temperature, favorable for reliable long-term data retention in nonvolatile memories. The observed and predicted characteristic features make GeTe 2 a promising precursor for the next generation of phase-change materials, especially coupled with additional metal doping, depolymerizing the tetrahedral interconnected glass network and accelerating (sub)nanosecond crystallization.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Machine Learning Assisted Safety Modeling and Analysis of Advanced Reactors

With the advances in computational power and numerical methods, analysts can now rely on first-principle simulations to predict ultra-fine details in a variety of applications. Advances in machine learning (ML) have produced algorithms that can now learn high-level abstractions via hierarchical models. This project aims to leverage advances in ML techniques and the available high-resolution simulation data to develop a novel modeling and simulation (M\&S) methodology for reactor safety analysis. While application-agnostic ML techniques are available, complex physics constraints need to be incorporated into ML techniques to build ML-based closures for computationally efficient predictive simulations. This project intends to develop a physics-guided data-driven multi-scale methodology for M\&S of advanced reactors. The project focuses on thermal fluid (T/F) phenomena, which play major roles in advanced reactor safety. Specifically, we propose a data-driven coarse-mesh turbulence model based on local flow features for the transient analysis of thermal mixing and stratification in a sodium-cooled fast reactor (SFR). The model has a coarse-mesh setup to ensure computational efficiency, while it is trained by fine-mesh computational fluid dynamics (CFD) data with Reynolds-averaged Navier-Stokes (RANS) turbulence model to ensure accuracy. Three different neural networks are developed and tested for loss-of-flow transients in the hot pool of SFR, i.e. the densely connected convolutional neural network (DCNN), long-short-term-memory network based on proper orthogonal decomposition (POD-LSTM), and the DCNN informed by LSTM (DCNN-LSTM). The performances of these three neural networks are evaluated based on baseline models. The DCNN-LSTM model has been chosen for further hyperparameter optimization. Furthermore, based on a simplified two-dimensional case, uncertainty quantification (UQ) of the developed ML-based closure are investigated with three methods, i.e. Monte Carlo dropout, deep ensemble, and Bayesian neural network. The developed ML-based turbulent viscosity closure relation based on deep ensemble is then integrated into the system analysis module SAM and serves as a term in the conservation equations. Such a SAM-ML based procedure guarantees that the obtained results are consistent with the physical constraints of the thermal-fluid system. The SAM-ML simulation on the same loss-of-flow transient showed comparable accuracy with the CFD simulation but with a much coarser mesh setup. Last but not least, the ML-based closure improvement with the support of higher-fidelity data from large eddy simulation (LES) is discussed. As a first step towards this direction, a baseline LES simulation is performed to obtain comparable data with RANS results. Based on the early results, future investigation on further improving the ML-based closure is discussed. We believe the developed approach that combines scientific machine learning with nuclear system analysis code can benefit the advanced reactor community as more accurate safety analyses will better characterize reactor safety margins and reduce licensing efforts.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Quantification of oil and gas methane emissions in the Delaware and Marcellus basins using a network of continuous tower-based measurements

Abstract. According to the United States Environmental Protection Agency (US EPA), emissions from oil and gas infrastructure contribute 30 % of all anthropogenic methane (CH4) emissions in the US. Studies in the last decade have shown emissions from this sector to be substantially larger than bottom-up assessments, including the EPA inventory, highlighting both the increased importance of methane emissions from the oil and gas sector in terms of their overall climatological impact and the need for independent monitoring of these emissions. In this study we present continuous monitoring of regional methane emissions from two oil and gas basins using tower-based observing networks. Continuous methane measurements were taken at four tower sites in the northeastern Marcellus basin from May 2015 through December 2016 and five tower sites in the Delaware basin in the western Permian from March 2020 through April 2022. These measurements, an atmospheric transport model, and prior emission fields are combined using an atmospheric inversion to estimate monthly methane emissions in the two regions. This study finds the mean overall emission rate from the Delaware basin during the measurement period to be 146–210 Mg CH4 h−1 (energy-normalized loss rate of 1.1 %–1.5 %, gas-normalized rate of 2.5 %–3.5 %). Strong temporal variability in the emissions was present, with the lowest emission rates occurring during the onset of the COVID-19 pandemic. Additionally, a synthetic model–data experiment performed using the Delaware tower network shows that the presence of intermittent sources is not a significant source of uncertainty in monthly quantification of the mean emission rate. In the Marcellus, this study finds the overall mean emission rate to be 19–28 Mg CH4 h−1 (gas-normalized loss rate of 0.30 %–0.45 %), with relative consistency in the emission rate over time. These totals align with aircraft top-down estimates from the same time periods. In both basins, the tower network was able to constrain monthly flux estimates within ±20 % uncertainty in the Delaware and ±24 % uncertainty in the Marcellus. The results from this study demonstrate the ability to monitor emissions continuously and detect changes in the emissions field, even in a basin with relatively low emissions and complex background conditions.

54 ENVIRONMENTAL SCIENCES↗

Learning in Modal Space: Solving Time-Dependent Stochastic PDEs Using Physics-Informed Neural Networks

One of the open problems in scientific computing is the long-time integration of nonlinear stochastic partial differential equations (SPDEs), especially with arbitrary initial data. We address this problem by taking advantage of recent advances in scientific machine learning and the spectral dynamically orthogonal (DO) and borthogonal (BO) methods for representing stochastic processes. The recently introduced DO/BO methods reduce the SPDE to solving a system of deterministic PDEs and a system of stochastic ordinary differential equations. Specifically, we propose two new physics-informed neural networks (PINNs) for solving time-dependent SPDEs, namely the neural network (NN)-DO/BO methods. The proposed methods incorporate the DO/BO constraints into the loss function (along with the modal decomposition of the SPDE) with an implicit form instead of generating explicit expressions for the temporal derivatives of the DO/BO modes. Hence, the NN-DO/BO methods can overcome some of the drawbacks of the original DO/BO methods. For example, we do not need the assumption that the covariance matrix of the random coefficients is invertible as in the original DO method, and we can remove the assumption of no eigenvalue crossing as in the original BO method. Moreover, the NN-DO/BO methods can be used to solve time-dependent stochastic inverse problems with the same formulation and same computational complexity as for forward problems. Furthermore, we demonstrate the capability of the proposed methods via several numerical examples, namely: (1) A linear stochastic advection equation with deterministic initial condition: we obtain good results with the proposed methods, while the original DO/BO methods cannot be applied directly in this case. (2) Long-time integration of the stochastic Burgers' equation: we show the good performance of NN-DO/BO methods, especially the effectiveness of the NN-BO approach for such problems with many eigenvalue crossings during the whole time evolution, while the original BO method fails. (3) Nonlinear reaction diffusion equation: we consider both the forward problem and the inverse problems, including very noisy initial point values, to investigate the flexibility of the NN-DO/BO methods in handling inverse and mixed type problems. Taken together, these simulation results demonstrate that the NN-DO/BO methods can be employed to effectively quantify uncertainty propagation in a wide range of physical problems, but future work should address the efficiency issue of PINNs for forward problems.

97 MATHEMATICS AND COMPUTING↗

Molecular basis of differential HLA class I-restricted T cell recognition of a highly networked HIV peptide

Cytotoxic-T-lymphocyte (CTL) mediated control of HIV-1 is enhanced by targeting highly networked epitopes in complex with human-leukocyte-antigen-class-I (HLA-I). However, the extent to which the presenting HLA allele contributes to this process is unknown. Here we examine the CTL response to QW9, a highly networked epitope presented by the disease-protective HLA-B57 and disease-neutral HLA-B53. Despite robust targeting of QW9 in persons expressing either allele, T cell receptor (TCR) cross-recognition of the naturally occurring variant QW9_S3T is consistently reduced when presented by HLA-B53 but not by HLA-B57. Crystal structures show substantial conformational changes from QW9-HLA to QW9_S3T-HLA by both alleles. The TCR-QW9-B53 ternary complex structure manifests how the QW9-B53 can elicit effective CTLs and suggests sterically hindered cross-recognition by QW9_S3T-B53. We observe populations of cross-reactive TCRs for B57, but not B53 and also find greater peptide-HLA stability for B57 in comparison to B53. These data demonstrate differential impacts of HLAs on TCR cross-recognition and antigen presentation of a naturally arising variant, with important implications for vaccine design.

60 APPLIED LIFE SCIENCES↗

TroR is the primary regulator of the iron homeostasis transcription network in the halophilic archaeon Haloferax volcanii

Abstract Maintaining the intracellular iron concentration within the homeostatic range is vital to meet cellular metabolic needs and reduce oxidative stress. Previous research revealed that the haloarchaeon Halobacterium salinarum encodes four diphtheria toxin repressor (DtxR) family transcription factors (TFs) that together regulate the iron response through an interconnected transcriptional regulatory network (TRN). However, the conservation of the TRN and the metal specificity of DtxR TFs remained poorly understood. Here we identified and characterized the TRN of Haloferax volcanii for comparison. Genetic analysis demonstrated that Hfx. volcanii relies on three DtxR transcriptional regulators (Idr, SirR, and TroR), with TroR as the primary regulator of iron homeostasis. Bioinformatics and molecular approaches revealed that TroR binds a conserved cis-regulatory motif located ∼100 nt upstream of the start codon of iron-related target genes. Transcriptomics analysis demonstrated that, under conditions of iron sufficiency, TroR repressed iron uptake and induced iron storage mechanisms. TroR repressed the expression of one other DtxR TF, Idr. This reduced DtxR TRN complexity relative to that of Hbt. salinarum appeared correlated with natural variations in iron availability. Based on these data, we hypothesize that variable environmental conditions such as iron availability appear to select for increasing TRN complexity.

59 BASIC BIOLOGICAL SCIENCES↗

eCounter: Inline Per-IP Network Monitoring at Millisecond Resolution via eBPF

Scientific data acquisition (SciDAQ) systems are shifting from archive-based workflows to streaming paradigms, where real-time, fine-grained network monitoring becomes essential. While P4-enabled devices offer per-packet in-band observability, they require specialized switches and routers. Host-side tools like Prometheus exporters lack sufficient temporal granularity. To bridge this gap, we present eCounter, a lightweight, hardware-agnostic, inline telemetry agent built on extended Berkeley Packet Filter (eBPF). eCounter captures per-interface ingress and egress traffic, categorized by IP address and protocol, at millisecond to sub-millisecond resolution. In a 100 Gbps environment, it continuously exports up to 3,257 time-series bins per second with only 4% CPU utilization at a 35¿KiB/s data rate. We evaluate eCounter across diverse NIC MTU settings, hook types, CPU architectures and operating systems, and observed negligible impact on concurrent high-throughput streaming applications. Complexity analysis confirms that it can be readily scaled to distributed SciDAQ deployments.

Mei, Xinxin [Computational Sciences and Technology↗

Towards Generalizable and Efficient Circuit Topology Design: A Graph-Transformer-based Surrogate Model with Curriculum Learning

Unlike circuit parameter and sizing optimizations, the automated design of analog circuit topologies poses significant challenges for learning-based approaches. One challenge arises from the combinatorial growth of the topology space with circuit size, which limits the topology optimization efficiency. Moreover, traditional circuit evaluation methods are time-consuming, while the presence of data discontinuity in the topology space makes the accurate prediction of circuit performance exceptionally difficult for unseen topologies. To tackle these challenges, we design a novel Graph-Transformer-based Network (GTN) as the surrogate model for circuit evaluation, offering a substantial acceleration in the speed of circuit topology optimization without sacrificing performance. Our GTN model architecture is designed to embed voltage changes in circuit loops and current flows in connected devices, enabling accurate performance predictions for circuits with unseen topologies. To address the cold start problem when scaling GTN to large-scale circuits, we further introduce a curriculum learning strategy that progressively trains GTN from small-scale to large-scale circuits. This approach enables the model to first learn fundamental physical principles from simpler topologies and gradually adapt to complex configurations, effectively bridging the circuit complexity gap and improving prediction accuracy. Taking the power converter circuit design as an experimental task, our GTN model significantly outperforms an analytical approach and baseline methods directly utilizing graph neural networks. Furthermore, GTN achieves less than 5% relative error and 196× speed-up compared with high-fidelity simulation. Notably, our GTN surrogate model empowers an automatic circuit design framework to discover circuits of comparable quality to those identified through high-fidelity simulation while reducing the time required by up to 98.2%. With curriculum learning, the enhanced GTN achieves a 51% improvement for performance prediction of large-scale circuits compared to the GTN model without this strategy. These advancements establish GTN as a scalable framework for automated analog circuit design across varying circuit complexity levels.

Lu, Haoshu [New Jersey Institute of Technology (NJ↗

Anticipating gelation and vitrification with medium amplitude parallel superposition (MAPS) rheology and artificial neural networks

Abstract Anticipating qualitative changes in the rheological response of complex fluids (e.g., a gelation or vitrification transition) is an important capability for processing operations that utilize such materials in real-world environments. One class of complex fluids that exhibits distinct rheological states are soft glassy materials such as colloidal gels and clay dispersions, which can be well characterized by the soft glassy rheology (SGR) model. We first solve the model equations for the time-dependent, weakly nonlinear response of the SGR model. With this analytical solution, we show that the weak nonlinearities measured via medium amplitude parallel superposition (MAPS) rheology can be used to anticipate the rheological aging transitions in the linear response of soft glassy materials. This is a rheological version of a technique called structural health monitoring used widely in civil and aerospace engineering. We design and train artificial neural networks (ANNs) that are capable of quickly inferring the parameters of the SGR model from the results of sequential MAPS experiments. The combination of these data-rich experiments and machine learning tools to provide a surrogate for computationally expensive viscoelastic constitutive equations allows for rapid experimental characterization of the rheological state of soft glassy materials. We apply this technique to an aging dispersion of Laponite ® clay particles approaching the gel point and demonstrate that a trained ANN can provide real-time detection of transitions in the nonlinear response well in advance of incipient changes in the linear viscoelastic response of the system.

Lennon, Kyle R. (ORCID:0000000212515461)↗

Adsorption and Capillary Condensation-Induced Imbibition in Nanoporous Media

Multiphase flow phenomena in nanoporous media are encountered in many science and engineering applications. Shales, for example, possessing complex nanopore networks, have considerable importance as source rocks for unconventional oil and gas production and as low-permeability seals for geologic carbon sequestration or nuclear waste disposal. This work presents a theoretical investigation of the processes controlling adsorption, capillary condensation, and imbibition in such nanoporous media, with a particular focus on understanding the effects of fluid-fluid and fluid-pore wall interaction forces in the interconnected nanopore space. Building on a new theoretical framework, we developed a numerical model for the multiphase nanoporous flow and tested it against water vapor uptake measurements conducted on a shale core sample. The model, which is based on the density functional approach, explicitly includes the relevant interaction forces among fluids and solids while allowing for a continuum representation of the porous medium. The experimental data include gravimetrically measured mass changes in an initially dry core sample exposed to varying levels of relative humidity, starting with a low relative humidity (rh = 0.31) followed by a period of a higher relative humidity (rh = 0.81). During this process, water vapor uptake in the dry core is recorded as a function of time. Our model suggests that, under low rh conditions, the flow within the shale sample is controlled by adsorption- and diffusion-type processes. After increasing the rh to 0.81, the uptake of water vapor becomes more significant, and according to our model, this can be explained by capillary condensation followed by immiscible displacement in the core sample. It appears that strong fluid-pore wall attractive forces cause condensation near the inlet, which then induces water imbibition further into sample.

58 GEOSCIENCES↗

Insights on the dissolution of water in an albite melt at high pressures and temperatures from a direct structural analysis

The water dissolution mechanism in silicate melts under high pressures is not well understood. Here we present the first direct structure investigation of a water-saturated albite melt to monitor the interactions between water and the network structure of silicate melt at the molecular level. In situ high-energy X-ray diffraction was carried out on the NaAlSi 3 O 8 -H 2 O system at 800 °C and 300 MPa, at the Advanced Photon Source synchrotron facility. The analysis of the X-ray diffraction data was augmented with classical Molecular Dynamics simulations of a hydrous albite melt, incorporating accurate water-based interactions. The results show that metal–oxygen bond breaking at the bridging sites occurs overwhelmingly at the Si site upon reaction with H 2 O, with subsequent Si–OH bond formation and negligible Al–OH formation. Furthermore, we see no evidence for the dissociation of the Al 3+ ion from the network structure upon breaking of the Si–O bond in the hydrous albite melt. The results also indicate that the Na + ion is an active participant in the modifications of the silicate network structure of the albite melt upon water dissolution at high P–T conditions. We do not find evidence for the Na + ion dissociating from the network structure upon depolymerization and subsequent formation of NaOH complexes. Instead, our results show that the Na + ion persists as a structure modifier with a shift away from Na–BO bonding to an increase in the extent of Na-NBO bonding, in parallel with pronounced depolymerization of the network. Our MD simulations show that the Si–O and Al–O bond lengths are expanded by about 6% in the hydrous albite melt compared to those of the dry melt at high P–T conditions. The changes in the network silicate structure of a hydrous albite melt at high pressure and temperature, as revealed in this study, must be considered in the advancement of water dissolution models of hydrous granitic (or alkali aluminosilicate) melts.

36 MATERIALS SCIENCE↗

Uncertainty-Aware, Structure-Preserving Machine Learning Approach for Domain Shift Detection From Nonlinear Dynamic Responses of Structural Systems

Complex structural systems deployed for aerospace, civil, or mechanical applications must operate reliably under varying operational conditions. Structural health monitoring (SHM) systems help ensure the reliability of these systems by providing continuous monitoring of the state of the structure. SHM relies on synthesizing measured data with a predictive model to make informed decisions about structural states. However, these models—which may be thought of as a form of a digital twin—need to be updated continuously as structural changes (e.g., due to damage) arise. We propose an uncertainty-aware machine learning model that enforces distance preservation of the original input state space and then encodes a distance-aware mechanism via a Gaussian process (GP) kernel. The proposed approach leverages the spectral-normalized neural GP algorithm to combine the flexibility of neural networks with the advantages of GP, subjected to structure-preserving constraints, to produce an uncertainty-aware model. This model is used to detect domain shift due to structural changes that cannot be observed directly because they may be spatially isolated (e.g., inside a joint or localized damage). This work leverages detection theory to detect domain shift systematically given statistical features of the prediction variance produced by the model. The proposed approach is demonstrated on a nonlinear structure being subjected to damage conditions. In conclusion, it is shown that the proposed approach is able to rely on distances of the transformed input state space to predict increased variance in shifted domains while being robust to normative changes.

Algorithms↗