Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “statistical analysis”

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 271 records · Page 15

SU-E-J-253: The Radiomics Toolbox in the Computational Environment for Radiological Research (CERR)

Purpose: To present an open source and free platform to facilitate radiomics research — The “Radiomics toolbox” in CERR. Method: There is scarcity of open source tools that support end-to-end modeling of image features to predict patient outcomes. The “Radiomics toolbox” strives to fill the need for such a software platform. The platform supports (1) import of various kinds of image modalities like CT, PET, MR, SPECT, US. (2) Contouring tools to delineate structures of interest. (3) Extraction and storage of image based features like 1st order statistics, gray-scale co-occurrence and zonesize matrix based texture features and shape features and (4) Statistical Analysis. Statistical analysis of the extracted features is supported with basic functionality that includes univariate correlations, Kaplan-Meir curves and advanced functionality that includes feature reduction and multivariate modeling. The graphical user interface and the data management are performed with Matlab for the ease of development and readability of code and features for wide audience. Open-source software developed with other programming languages is integrated to enhance various components of this toolbox. For example: Java-based DCM4CHE for import of DICOM, R for statistical analysis. Results: The Radiomics toolbox will be distributed as an open source, GNU copyrighted software. The toolbox was prototyped for modeling Oropharyngeal PET dataset at MSKCC. The analysis will be presented in a separate paper. Conclusion: The Radiomics Toolbox provides an extensible platform for extracting and modeling image features. To emphasize new uses of CERR for radiomics and image-based research, we have changed the name from the “Computational Environment for Radiotherapy Research” to the “Computational Environment for Radiological Research”.

60 APPLIED LIFE SCIENCES↗

Active Storage with Analytics Capabilities and I/O Runtime System for Petascale Systems

Computational scientists must understand results from experimental, observational and computational simulation generated data to gain insights and perform knowledge discovery. As systems approach the petascale range, problems that were unimaginable a few years ago are within reach. With the increasing volume and complexity of data produced by ultra-scale simulations and high-throughput experiments, understanding the science is largely hampered by the lack of comprehensive I/O, storage, acceleration of data manipulation, analysis, and mining tools. Scientists require techniques, tools and infrastructure to facilitate better understanding of their data, in particular the ability to effectively perform complex data analysis, statistical analysis and knowledge discovery. The goal of this work is to enable more effective analysis of scientific datasets through the integration of enhancements in the I/O stack, from active storage support at the file system layer to MPI-IO and high-level I/O library layers. We propose to provide software components to accelerate data analytics, mining, I/O, and knowledge discovery for large-scale scientific applications, thereby increasing productivity of both scientists and the systems. Our approaches include 1) design the interfaces in high-level I/O libraries, such as parallel netCDF, for applications to activate data mining operations at the lower I/O layers; 2) Enhance MPI-IO runtime systems to incorporate the functionality developed as a part of the runtime system design; 3) Develop parallel data mining programs as part of runtime library for server-side file system in PVFS file system; and 4) Prototype an active storage cluster, which will utilize multicore CPUs, GPUs, and FPGAs to carry out the data mining workload.

97 MATHEMATICS AND COMPUTING↗

Quantitative assessment of particle dispersion in polymeric composites and its effect on mechanical properties

In this work, an automated image analysis tool is developed to establish quantitative correlations between the particle/cluster size distribution and the mechanical properties of particle reinforced polymer composites (PRPC).This automated image analysis tool is developed within python programming software to process and analyze the microstructural images of the polymerbased composite materials. The spent coffee bean powder (SCBP) reinforced poly-propylene carbonate (PPC) polymer composite with differing wt.% of the filler is selected for the analysis. Detailed statistical analysis of the microstructural images reveals that ‘clustering of clusters’ is also presented in addition to the most commonly reported ‘clustering of particles’, and the distribution of particle/clusters is bimodal. Based on these findings, an effective volume fraction for the filler material is proposed to mainly capture the agglomeration effect. With this effective volume fraction, the standard rule-of-mixture model correctly captures the experimentally measured tensile strength and modulus as a function of filler wt.%. Further, the applicability of this effective volume fraction for other theoretical models is also analyzed. The detailed statistical analysis of the microstructure and the proposed effective volume fraction helps to develop a deeper quantitative understanding of the PRPC than the conventional qualitative correlation of microstructural features with the properties and failure processes.

36 MATERIALS SCIENCE↗

THE SLOAN DIGITAL SKY SURVEY DATA RELEASE 7 SPECTROSCOPIC M DWARF CATALOG. II. STATISTICAL PARALLAX ANALYSIS

We present a statistical parallax analysis of low-mass dwarfs from the Sloan Digital Sky Survey. We calculate absolute r-band magnitudes (M{sub r} ) as a function of color and spectral type and investigate changes in M{sub r} with location in the Milky Way. We find that magnetically active M dwarfs are intrinsically brighter in M{sub r} than their inactive counterparts at the same color or spectral type. Metallicity, as traced by the proxy {zeta}, also affects M{sub r} , with metal-poor stars having fainter absolute magnitudes than higher metallicity M dwarfs at the same color or spectral type. Additionally, we measure the velocity ellipsoid and solar reflex motion for each subsample of M dwarfs. We find good agreement between our measured solar peculiar motion and previous results for similar populations, as well as some evidence for differing motions of early and late M-type populations in U and W velocities that cannot be attributed to asymmetric drift. The reflex solar motion and the velocity dispersions both show that younger populations, as traced by magnetic activity and location near the Galactic plane, have experienced less dynamical heating. We introduce a new parameter, the independent position altitude (IPA), to investigate populations as a function of vertical height from the Galactic plane. M dwarfs at all types exhibit an increase in velocity dispersion when analyzed in comparable IPA subgroups.

79 ASTRONOMY AND ASTROPHYSICS↗

A local structure model for network analysis

The statistical analysis of networks is a popular research topic with ever widening applications. Exponential random graph models (ERGMs), which specify a model through interpretable, global network features, are common for this purpose. In this study we introduce a new class of models for network analysis, called local structure graph models (LSGMs). In contrast to an ERGM, a LSGM specifies a network model through local features and allows for an interpretable and controllable local dependence structure. In particular, LSGMs are formulated by a set of full conditional distributions for each network edge, e.g., the probability of edge presence/absence, depending on neighborhoods of other edges. Additional model features are introduced to aid in specification and to help alleviate a common issue (occurring also with ERGMs) of model degeneracy. Finally, the proposed models are demonstrated on a network of tornadoes in Arkansas where a LSGM is shown to perform significantly better than a model without local dependence.

97 MATHEMATICS AND COMPUTING↗

Systematic analysis on spectral statistics of odd-A nuclei

Level statistics for given spin and parity below 10 MeV within A≤71, 71≤A≤151, and 151≤A≤221 mass regions are systematically analyzed. We have taken 5316 level spacings from 266 odd-A nuclei in the light mass region, 6897 level spacings from 478 ones in the medium mass region, and 2768 level spacings from 143 ones in the heavy mass region for the statistics. In the analysis, the chaoticity parameter in the Berry–Robnik distribution is determined based on the method of maximum likelihood estimation. The results show a mass number A-dependence in the level statistics. The larger the mass number, the relatively stronger in chaoticity. In addition, with the level statistics in using data sequences classified according to the order of magnitude of half-lives, it is shown that the level energies to be more chaotic for longer lived light and heavy mass odd-A nuclei, and intermediate between regular and chaotic in medium mass odd-A nuclei. The level statistical analysis in concerning E2 transition strengths of E2 or mixed E2+M1 transitions in odd-A nuclei is also made, which shows chaos in level energies likely occurring in those with weaker E2 transition strengths. Therefore, it can be concluded that odd-A nuclei are, more often than not, chaotic. Chaos is likely emergent in level energies of stable light mass nuclei with weaker E2 transition strengths, while it is definitely emergent in level energies of heavy mass nuclei, and even stronger in comparatively stable ones and those with weaker E2 transition strengths.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Accelerated statistical failure analysis of multifidelity TRISO fuel models

Statistical nuclear fuel failure analysis is critical for the design and development of advanced reactor technologies. Although Monte Carlo Sampling (MCS) is a standard method of statistical failure analysis for fuels, the low failure probabilities of some advanced fuel forms and the correspondingly large number of required model evaluations limit its application to low-fidelity (e.g., 1-D) fuel models. In this paper, we present four other statistical methods for fuel failure analysis in Bison, considering tri-structural isotropic (TRISO)-coated particle fuel as a case study. The statistical methods considered are Latin hypercube sampling (LHS), adaptive importance sampling (AIS), subset simulation (SS), and the Weibull theory. Using these methods, we analyzed both 1-D and 2-D representations of TRISO models to compute failure probabilities and the distributions of fuel properties that result in failures. The results of these methods compare well across all TRISO models considered. Overall, SS and the Weibull theory were deemed the most efficient, and can be applied to both 1-D and 2-D TRISO models to compute failure probabilities. Moreover, since SS also characterizes the distribution of parameters that cause TRISO failures, and can consider failure modes not described by the Weibull criterion, it may be preferred over the other methods. Finally, a discussion on the efficacy of different statistical methods of assessing nuclear fuel safety is provided.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

On the importance of phase saturation heterogeneity in the analysis of laboratory studies of hydrate dissociation

Methane hydrates (MHs) have been considered as the future source of energy because of its vast resource volume and high energy density. Energy recovery from MH-bearing sediments has attracted intensifying research activities. Fundamentally, heat transfer, fluid flow through porous media, and the kinetics of hydrate reaction are the three key processes controlling the behavior of MH dissociation and the associated fluid production. Earlier studies have suggested that heterogeneous spatial distribution of SH is inevitable in MH-bearing samples synthesized in laboratory. In this paper, we extend our study to analyze numerically the simulation results from the two realizations of the samples (homogeneous and heterogeneous) to identify differences in the fluid production and to determine if they are sufficiently different. Additionally, we conduct a sensitivity analysis and a statistical analysis on the key transport and kinetic rate parameters that could affect hydrate dissociation and fluid production in the context of a heterogeneous hydrate-bearing sample, in an effort to provide insights that could lead to improved designs for laboratory experiments and (possibly) field applications. Our results suggest that the approximation of an artificial hydrate-bearing core with heterogeneous phase saturations by an assumption of uniform phase saturation distributions results in practically similar fluid production profile except for the very early stage with maximum 20.0% deviation in the water production. From the sensitivity and statistical analysis, we determine that gas production depends strongly on the kinetic rate constant, K d 0 and the composite thermal conductivity of the hydrate-bearing sediments, λ θ ; while, water production is very sensitive to K d 0 and the absolute permeability of the sandy medium, k. Understanding the effect of phase heterogeneity and the relative importance of key parameters on the production behavior of hydrate-bearing sediments could provide basis for novel production technologies that lead to enhanced gas production and energy efficiency in the energy recovery process.

42 ENGINEERING↗

Efficient high-fidelity TRISO statistical failure analysis using Bison: Applications to AGR-2 irradiation testing

The ability of tri-structural isotropic (TRISO) fuel to contain fission products is largely dictated by the quality of the manufacturing process, since most of the fission product release is expected to occur due to coating layer failure in a small number of particles containing defects. The Bison fuel performance code has capabilities to predict failure in individual particles, accounting for the presence of defects, and to apply statistical analysis methods to compute the probability of failure in a set of fuel particles. Bison has recently undergone significant development both to improve its physical representations of fuel particle behavior and to improve the efficiency of its statistical failure calculations. Physical model improvements include new capabilities to account for the pressure generated by fission gases on inner pyrolytic carbon (IPyC) crack surfaces and to use local material coordinate orientation to accurately incorporate the anisotropy in the material properties in aspherical particles. To improve statistical modeling efficiency, a direct integration approach which involves directly integrating the failure probability function associated with statistically varying parameters has been developed. The direct integration approach is much more efficient than the Monte Carlo (MC) schemes commonly employed, and allows Bison to directly run high-dimensional fuel performance models, which improves the accuracy of failure probability calculations. Finally, a set of benchmark problems is considered here to compare the MC and direct integration approaches, and a statistical failure analysis of compacts in the Advanced Gas Reactor (AGR)-2 experiments is performed using the direct integration approach.

36 MATERIALS SCIENCE↗

Global Sensitivity Analysis for Statistical Model Parameters

Global sensitivity analysis (GSA) is typically used to analyze the influence of uncertain parameters in mathematical models and simulations. In principle, tools from GSA may be extended to analyze the influence of parameters in statistical models. Such analyses may enable reduced or parsimonious modeling and greater predictive capability. Yet, difficulties such as parameter correlation, model stochasticity, multivariate model output, and unknown parameter distributions prohibit a direct application of GSA tools to statistical models. By leveraging a loss function associated with the statistical model, we introduce a novel framework to address these difficulties and enable efficient GSA for statistical model parameters. Theoretical and computational properties are considered and hihglighted on a synthetic example. The framework is applied to a Gaussian process model from the literature, which depends on 95 parameters. Noninfluential parameters are discovered through GSA, and a reduced model with equal or stronger predictive capability is constructed by using only 79 parameters.

97 MATHEMATICS AND COMPUTING↗

Statistical data analysis of x-ray spectroscopy data enabled by neural network accelerated Bayesian inference

Bayesian inference applied to x-ray spectroscopy data analysis enables uncertainty quantification necessary to rigorously test theoretical models. However, when comparing to data, detailed atomic physics and radiation transfer calculations of x-ray emission from non-uniform plasma conditions are typically too slow to be performed in line with statistical sampling methods, such as Markov Chain Monte Carlo sampling. Furthermore, differences in transition energies and x-ray opacities often make direct comparisons between simulated and measured spectra unreliable. Here, we present a spectral decomposition method that allows for corrections to line positions and bound–bound opacities to best fit experimental data, with the goal of providing quantitative feedback to improve the underlying theoretical models and guide future experiments. In this work, we use a neural network (NN) surrogate model to replace spectral calculations of isobaric hot-spots created in Kr-doped implosions at the National Ignition Facility. The NN was trained on calculations of x-ray spectra using an isobaric hot-spot model post-processed with Cretin, a multi-species atomic kinetics and radiation code. The speedup provided by the NN model to generate x-ray emission spectra enables statistical analysis of parameterized models with sufficient detail to accurately represent the physical system and extract the plasma parameters of interest.

47 OTHER INSTRUMENTATION↗

MAPPING THE LOCAL HALO: STATISTICAL PARALLAX ANALYSIS OF SDSS LOW-MASS SUBDWARFS

We present a statistical parallax study of nearly 2000 M subdwarfs with photometry and spectroscopy from the Sloan Digital Sky Survey (SDSS). Statistical parallax analysis yields the mean absolute magnitudes, mean velocities, and velocity ellipsoids for homogenous samples of stars. We selected homogeneous groups of subdwarfs based on their photometric colors and spectral appearance. We examined the color-magnitude relations of low-mass subdwarfs and quantified their dependence on the newly refined metallicity parameter, {zeta}. We also developed a photometric metallicity parameter, {delta}{sub (g-r)}, based on the g - r and r - z colors of low-mass stars and used it to select stars with similar metallicities. The kinematics of low-mass subdwarfs as a function of color and metallicity were also examined and compared to main-sequence M dwarfs. We find that the SDSS subdwarfs share similar kinematics to the inner halo and thick disk. The color-magnitude relations derived in this analysis will be a powerful tool for identifying and characterizing low-mass metal-poor subdwarfs in future surveys such as Gaia and LSST, making them important and plentiful tracers of the stellar halo.

79 ASTRONOMY AND ASTROPHYSICS↗

Site-specific Design Case Study for Wet Waste Hydrothermal Liquefaction and Biocrude Upgrading to Hydrocarbon Fuels

Hydrothermal liquefaction (HTL) is a thermal process that converts wet biomass to renewable hydrocarbon fuel blendstocks (i.e., renewable naphtha, renewable diesel, and sustainable aviation fuel (SAF)). It can utilize a wide range of pure and blended wet feedstocks, including sewage sludge from water resource recovery facilities (WRRF), food and agriculture wastes, algae, fats, oils and greases (FOG) and blends of dry and wet wastes/feedstocks. Historically, techno-economic analysis (TEA) and annual state of technology (SOT) assessments with standard economic assumptions used by the Bioenergy Technologies Office (BETO) were conducted for the wet waste HTL pathway leveraging experimental data collected from Pacific Northwest National Laboratory’s (PNNL) continuous flow reactor systems. The objective of the SOT assessment has been to guide and track progress of BETO’s HTL research and development (R&D) toward reduced cost and greenhouse gas (GHG) emissions for the pathway. However, gaps exist between BETO’s traditional SOT updates and the needs of key external stakeholders that – if addressed – will accelerate technology adoption. This Business Case Study aims to bridge this gap by providing an updated design, TEA, and LCA based on PNNL’s FY23 R&D with added analyses and information that provide enhanced relevance for stakeholders of the HTL technology. This includes specific siting, regional wet waste resource inventory and transportation cost analyses, fuel market information, sustainable fuel policy impacts, economic metrics of net present value (NPV) and internal rate of return (IRR), greenhouse gas (GHG) emissions analysis, and statistical analysis of cost and technical uncertainties of the HTL plant design. The study focuses on the “Detroit combined statistical area (CSA)” region for siting of a wet waste HTL plant adjacent to the Great Lakes Water Authority (GLWA) facility with guidance from industry participants. Regional resource and siting analyses were conducted to identify feedstock availability, scale, and cost, as well as a beneficial site location. TEA with detailed rigorous capital cost estimation for the specific site application was conducted to evaluate the key economic metrics of most value to industrial partners. These include total capital investment, operating costs, minimum fuel selling price (MFSP) of the biocrude and fuel blendstock, and NPV and internal rate of return IRR with sustainable fuel credits. Life cycle analysis was conducted to evaluate the supply chain greenhouse gas (GHG) emissions for the wet waste HTL process as compared with petroleum derived diesel. This study is also informed by years of R&D and process de-risking learnings and was conducted with a basic engineering HTL plant design and costing that akin to a “first-of-a-kind” plant economics. This differs from our conventional “nth plant ” SOT assessments. Specifically, the HTL process model has been updated with more operationally reliable methods for feed heating and phase separations. Further, we have implemented additional spare equipment for redundancy, a more rigorous installed equipment cost estimation approach, and additional costs associated with feed formatting and delivery, building, piping and site development. An Excel-based cost sheet based on the basic engineering design is also released alongside the report that allows users to conduct customized TEA with their own feed composition and financial assumptions.

09 BIOMASS FUELS↗

Supporting data for Site-specific Design Case Study for Wet Waste Hydrothermal Liquefaction and Biocrude Upgrading to Hydrocarbon Fuels

Hydrothermal liquefaction (HTL) is a thermal process that converts wet biomass to renewable hydrocarbon fuel blendstocks (i.e., renewable naphtha, renewable diesel, and sustainable aviation fuel (SAF)). It can utilize a wide range of pure and blended wet feedstocks, including sewage sludge from water resource recovery facilities (WRRF), food and agriculture wastes, algae, fats, oils and greases (FOG) and blends of dry and wet wastes/feedstocks. Historically, techno-economic analysis (TEA) and annual state of technology (SOT) assessments with standard economic assumptions used by the Bioenergy Technologies Office (BETO) were conducted for the wet waste HTL pathway leveraging experimental data collected from Pacific Northwest National Laboratory’s (PNNL) continuous flow reactor systems. The objective of the SOT assessment has been to guide and track progress of BETO’s HTL research and development (R&D) toward reduced cost and greenhouse gas (GHG) emissions for the pathway. However, gaps exist between BETO’s traditional SOT updates and the needs of key external stakeholders that – if addressed – will accelerate technology adoption. This Business Case Study aims to bridge this gap by providing an updated design, TEA, and LCA based on PNNL’s FY23 R&D with added analyses and information that provide enhanced relevance for stakeholders of the HTL technology. This includes specific siting, regional wet waste resource inventory and transportation cost analyses, fuel market information, sustainable fuel policy impacts, economic metrics of net present value (NPV) and internal rate of return (IRR), greenhouse gas (GHG) emissions analysis, and statistical analysis of cost and technical uncertainties of the HTL plant design. The study focuses on the “Detroit combined statistical area (CSA)” region for siting of a wet waste HTL plant adjacent to the Great Lakes Water Authority (GLWA) facility with guidance from industry participants. Regional resource and siting analyses were conducted to identify feedstock availability, scale, and cost, as well as a beneficial site location. TEA with detailed rigorous capital cost estimation for the specific site application was conducted to evaluate the key economic metrics of most value to industrial partners. These include total capital investment, operating costs, minimum fuel selling price (MFSP) of the biocrude and fuel blendstock, and NPV and internal rate of return IRR with sustainable fuel credits. Life cycle analysis was conducted to evaluate the supply chain greenhouse gas (GHG) emissions for the wet waste HTL process as compared with petroleum derived diesel. This study is also informed by years of R&D and process de-risking learnings and was conducted with a basic engineering HTL plant design and costing that akin to a “first-of-a-kind” plant economics. This differs from our conventional “nth plant ” SOT assessments. Specifically, the HTL process model has been updated with more operationally reliable methods for feed heating and phase separations. Further, we have implemented additional spare equipment for redundancy, a more rigorous installed equipment cost estimation approach, and additional costs associated with feed formatting and delivery, building, piping and site development. An Excel-based cost sheet based on the basic engineering design is also released alongside the report that allows users to conduct customized TEA with their own feed composition and financial assumptions.

Li, Shuyun↗

Searches for New Long-Lived Particles and Upgrade to the ATLAS Inner Detector (Final Technical Report)

The search for new fundamental particles is one of the defining goals of the Large Hadron Collider (LHC). The discovery of the Higgs Boson by the ATLAS and CMS collaborations provided the capstone of the Standard Model of particle physics, but outstanding questions remain. Why does the Higgs boson have a mass of 125 GeV when its natural mass would be many orders of magnitude larger? Is there a universal symmetry which unites all three forces described by the Standard Model? Can that symmetry be extended to include gravity? Is dark matter, evidenced by astronomical observations, made of a particle that interacts via Standard Model forces with the rest of matter? Together, these motivations provide compelling arguments that new physical processes await discovery. This project addressed some outstanding questions about the fundamental particles and their interactions with the ATLAS experiment at the Large Hadron Collider. In particular, the project improved the discovery potential for new, long- lived particles produced via electroweak processes in proton-proton collisions and set world-leading limits on their existence for certain values of their potential mass and lifetime. To achieve this, the project developed new data analysis methods, developed new triggers to select events with new long-lived particles during data-taking of the ATLAS experiment, and analyzed the largest proton–proton collision dataset ever produced. The project also supported significant development of the data acquisition software for the upgrade to the ATLAS inner detector, the Inner TracKer (ITk). The upgrade of the ATLAS inner detector is essential to the success of the entire Phase II physics program on ATLAS. Personnel supported by the project provided support for integration, assembly, and testing of the inner two layers of the ITk pixel system during its prototype and pre-production phase. Four PhD students and two post-doctoral scholars were supported by the grant and received invaluable scientific training as part of the research endeavor. The students and postdocs gained essential professional skills in the areas of advanced data analysis techniques, statistical analysis of data and simulation, programming in C++ and Python, hardware and instrumentation development, and presentation and collaboration skills. Additionally, approximately ten undergraduate students supported through other funding sources participated in research activities synergistic with the goals of this project, receiving essential mentorship from the personnel supported by this project.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Multivariate Analysis as a Tool for Validating Tester Matching

A method of applying Principal Component Analysis, Soft Independent Modeling of Class Analysis, and statistical analysis is described that can be applied to many types of testers to ascertain how well matched the performance of the testers in the analysis are to one another or how well matched a tester is to itself at a later time. This method is most useful for situations for which the same units have not been run across the testers being analyzed for matched performance.

Multari, Rosalie A [Sandia National Laboratories (↗

Data Mining and Machine Learning for Power System Monitoring, Understanding, and Impact Evaluation

This chapter presents results from the Big Data analysis framework to improve power system situational awareness and system reliability. For this purpose, a dataset with real-world phasor measurement unit data and historical transmission system outage data has been created and used to carry out the analysis. Several statistical analysis and machine learning methods have been developed and implemented for event and anomaly detection and modeling. Detection and analysis results for actual examples of power system events are presented. Finally, data-driven characterization and risk assessment methods for weather-related extremes in power systems are developed and demonstrated on the Bonneville Power Administration system. These applications demonstrate the capability of Machine Learning (ML) methods to monitor system abnormalities, to predict system events, and to characterize the impact of extreme events on power grid

data mining, power grid, machine learning, anomaly↗

The OVIS analysis architecture.

This report summarizes the current statistical analysis capability of OVIS and how it works in conjunction with the OVIS data readers and interpolators. It also documents how to extend these capabilities. OVIS is a tool for parallel statistical analysis of sensor data to improve system reliability. Parallelism is achieved using a distributed data model: many sensors on similar components (metaphorically sheep) insert measurements into a series of databases on computers reserved for analyzing the measurements (metaphorically shepherds). Each shepherd node then processes the sheep data stored locally and the results are aggregated across all shepherds. OVIS uses the Visualization Tool Kit (VTK) statistics algorithm class hierarchy to perform analysis of each process's data but avoids VTK's model aggregation stage which uses the Message Passing Interface (MPI); this is because if a single process in an MPI job fails, the entire job will fail. Instead, OVIS uses asynchronous database replication to aggregate statistical models. OVIS has several additional features beyond those present in VTK that, first, accommodate its particular data format and, second, improve the memory and speed of the statistical analyses. First, because many statistical algorithms are multivariate in nature and sensor data is typically univariate, interpolation of data is required to provide simultaneous observations of metrics. Note that in this report, we will refer to a single value obtained from a sensor as a measurement while a collection of multiple sensor values simultaneously present in the system is an observation. A base class for interpolation is provided that abstracts the operation of converting multiple sensor measurements into simultaneous observations. A concrete implementation is provided that performs piecewise constant temporal interpolation of multiple metrics across a single component. Secondly, because calculations may summarize data too large to fit in memory OVIS analyses batches of observations at a time and aggregates these intermediate intra-process models as it goes before storing the final model for inter-process aggregation via database replication. This reduces the memory footprint of the analysis, interpolation, and the database client and server query processing. This also interleaves processing with the disk I/O required to fetch data from the database - also improving speed. This report documents how OVIS performs analyses and how to create additional analysis components that fetch measurements from the database, perform interpolation, or perform operations on streamed observations (such as model updates or assessments). The rest of this section outlines the OVIS analysis algorithm and is followed by sections specific to each subtask. Note that we are limiting our discussion for now to the creation of a model from a set of measurements, and not including the assessment of observations using a model. The same framework can be used for assessment but that use case is not detailed in this report.

97 MATHEMATICS AND COMPUTING↗