SEARCH · Engineering Papers

Results for “faults”

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.

SourceResearch connectionFrom yearThrough year

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

At least 289 records · Page 16

All sources NASA NTRS DOE OSTI

DOE OSTI2017

What does fault tolerant Deep Learning need from MPI?

Deep Learning (DL) algorithms have become the {\em de facto} Machine Learning (ML) algorithm for large scale data analysis. DL algorithms are computationally expensive -- even distributed DL implementations which use MPI require days of training (model learning) time on commonly studied datasets. Long running DL applications become susceptible to faults -- requiring development of a fault tolerant system infrastructure, in addition to fault tolerant DL algorithms. This raises an important question: {\em What is needed from MPI for designing fault tolerant DL implementations?} In this paper, we address this problem for permanent faults. We motivate the need for a fault tolerant MPI specification by an in-depth consideration of recent innovations in DL algorithms and their properties, which drive the need for specific fault tolerance features. We present an in-depth discussion on the suitability of different parallelism types (model, data and hybrid); a need (or lack thereof) for check-pointing of any critical data structures; and most importantly, consideration for several fault tolerance proposals (user-level fault mitigation (ULFM), Reinit) in MPI and their applicability to fault tolerant DL implementations. We leverage a distributed memory implementation of Caffe, currently available under the Machine Learning Toolkit for Extreme Scale (MaTEx). We implement our approaches by extending MaTEx-Caffe for using ULFM-based implementation. Our evaluation using the ImageNet dataset and AlexNet neural network topology demonstrates the effectiveness of the proposed fault tolerant DL implementation using OpenMPI based ULFM.

Amatya, Vinay C.↗

DOE OSTI2016

OpenStudio® Fault Models Gem [SWR-15-23]

Overview of the OpenStudio® Fault Models Gem: Creating a comprehensive library of OpenStudio® fault models. Features: - Physics-based fault impact simulation within the capability of EnergyPlus®/OpenStudio® - Fault intensity implementation simulating how severe faults are (e.g., how much portion is fouled in the condenser heat exchanger) and estimating the impact of a fault in different fault intensities. - Fault evolution implementation simulating evolving/increasing fault intensity through certain period of time. Future goals: - what are common/typical faults in different sample spaces (e.g., HVAC system type, building operation type, building location etc.)? - how often (e.g., incidence and prevalence) faults occur in different sample spaces (e.g., HVAC system type, building operation type, building location etc.)?

Frank, Stephen↗

DOE OSTI2024

A strainmeter array as the fulcrum of novel observatory sites along the Alto Tiberina Near Fault Observatory

Fault slip is a complex natural phenomenon involving multiple spatiotemporal scales from seconds to days to weeks. To understand the physical and chemical processes responsible for the full fault slip spectrum, a multidisciplinary approach is highly recommended. The Near Fault Observatories (NFOs) aim at providing high-precision and spatiotemporally dense multidisciplinary near-fault data, enabling the generation of new original observations and innovative scientific products. The Alto Tiberina Near Fault Observatory is a permanent monitoring infrastructure established around the Alto Tiberina fault (ATF), a 60 km long low-angle normal fault (mean dip 20°), located along a sector of the Northern Apennines (central Italy) undergoing an extension at a rate of about 3 mm yr –1 . The presence of repeating earthquakes on the ATF and a steep gradient in crustal velocities measured across the ATF by GNSS stations suggest large and deep (5–12 km) portions of the ATF undergoing aseismic creep. Both laboratory and theoretical studies indicate that any given patch of a fault can creep, nucleate slow earthquakes, and host large earthquakes, as also documented in nature for certain ruptures (e.g., Iquique in 2014, Tōhoku in 2011, and Parkfield in 2004). Nonetheless, how a fault patch switches from one mode of slip to another, as well as the interaction between creep, slow slip, and regular earthquakes, is still poorly documented by near-field observation. With the strainmeter array along the Alto Tiberina fault system (STAR) project, we build a series of six geophysical observatory sites consisting of 80–160 m deep vertical boreholes instrumented with strainmeters and seismometers as well as meteorological and GNSS antennas and additional seismometers at the surface. By covering the portions of the ATF that exhibits repeated earthquakes at shallow depth (above 4 km) with these new observatory sites, we aim to collect unique open-access data to answer fundamental questions about the relationship between creep, slow slip, dynamic earthquake rupture, and tectonic faulting.

58 GEOSCIENCES↗

NASA NTRS1989

Experiments in fault tolerant software reliability

Twenty functionally equivalent programs were built and tested in a multiversion software experiment. Following unit testing, all programs were subjected to an extensive system test. In the process sixty-one distinct faults were identified among the versions. Less than 12 percent of the faults exhibited varying degrees of positive correlation. The common-cause (or similar) faults spanned as many as 14 components. However, a majority of these faults were trivial, and easily detected by proper unit and/or system testing. Only two of the seven similar faults were difficult faults, and both were caused by specification ambiguities. One of these faults exhibited variable identical-and-wrong response span, i.e. response span which varied with the testing conditions and input data. Techniques that could have been used to avoid the faults are discussed. For example, it was determined that back-to-back testing of 2-tuples could have been used to eliminate about 90 percent of the faults. In addition, four of the seven similar faults could have been detected by using back-to-back testing of 5-tuples. It is believed that most, if not all, similar faults could have been avoided had the specifications been written using more formal notation, the unit testing phase was subject to more stringent standards and controls, and better tools for measuring the quality and adequacy of the test data (e.g. coverage) were used.

Mcallister, David F.↗

NASA NTRS1990

NASA ground terminal communication equipment automated fault isolation expert systems

The prototype expert systems are described that diagnose the Distribution and Switching System I and II (DSS1 and DSS2), Statistical Multiplexers (SM), and Multiplexer and Demultiplexer systems (MDM) at the NASA Ground Terminal (NGT). A system level fault isolation expert system monitors the activities of a selected data stream, verifies that the fault exists in the NGT and identifies the faulty equipment. Equipment level fault isolation expert systems are invoked to isolate the fault to a Line Replaceable Unit (LRU) level. Input and sometimes output data stream activities for the equipment are available. The system level fault isolation expert system compares the equipment input and output status for a data stream and performs loopback tests (if necessary) to isolate the faulty equipment. The equipment level fault isolation system utilizes the process of elimination and/or the maintenance personnel's fault isolation experience stored in its knowledge base. The DSS1, DSS2 and SM fault isolation systems, using the knowledge of the current equipment configuration and the equipment circuitry issues a set of test connections according to the predefined rules. The faulty component or board can be identified by the expert system by analyzing the test results. The MDM fault isolation system correlates the failure symptoms with the faulty component based on maintenance personnel experience. The faulty component can be determined by knowing the failure symptoms. The DSS1, DSS2, SM, and MDM equipment simulators are implemented in PASCAL. The DSS1 fault isolation expert system was converted to C language from VP-Expert and integrated into the NGT automation software for offline switch diagnoses. Potentially, the NGT fault isolation algorithms can be used for the DSS1, SM, amd MDM located at Goddard Space Flight Center (GSFC).

Tang, Y. K.↗

NASA NTRS1993

Fault geometries and extension in the Valles Marineris, Mars

The central troughs of the Valles Marineris system on Mars are generally regarded to be of tectonic origin because they are linear and bounded by faults. Their radial orientation to the Tharsis rise suggests that at least their incipient formation was influenced by the formation of the rise, but the exact role of Tharsis is not yet well understood. Collapse and erosion may have significantly modified the original tectonic troughs. To better understand the magnitude and mechanism of crustal extension across the Valles Marineris, the attitudes of trough-wall faults that parallel the long axes of the troughs were studied. Our results will help to determine if the extension is related to the Tharsis rise, to a local arch along which the troughs lie, or to other mechanisms. Fault-plane attitudes were obtained by performing a photogrammetric study of fault traces in the walls of Candor and Melas Chasmata. If the exposed wall faults are not major trough-bounding faults, it is assumed that these major faults are buried at the base of the walls and have similar geometries to the exposed faults. Three dimensional coordinates of several points along each fault trace were obtained and three-point solutions to their geometries were constructed. As expected, the results show that the faults dip toward the interior of the troughs, suggesting that they are indeed normal faults associated with trough formation.

Chadwick, D. J.↗

NASA NTRS2008

Model-Based Fault Tolerant Control

The Model Based Fault Tolerant Control (MBFTC) task was conducted under the NASA Aviation Safety and Security Program. The goal of MBFTC is to develop and demonstrate real-time strategies to diagnose and accommodate anomalous aircraft engine events such as sensor faults, actuator faults, or turbine gas-path component damage that can lead to in-flight shutdowns, aborted take offs, asymmetric thrust/loss of thrust control, or engine surge/stall events. A suite of model-based fault detection algorithms were developed and evaluated. Based on the performance and maturity of the developed algorithms two approaches were selected for further analysis: (i) multiple-hypothesis testing, and (ii) neural networks; both used residuals from an Extended Kalman Filter to detect the occurrence of the selected faults. A simple fusion algorithm was implemented to combine the results from each algorithm to obtain an overall estimate of the identified fault type and magnitude. The identification of the fault type and magnitude enabled the use of an online fault accommodation strategy to correct for the adverse impact of these faults on engine operability thereby enabling continued engine operation in the presence of these faults. The performance of the fault detection and accommodation algorithm was extensively tested in a simulation environment.

Kumar, Aditya↗

NASA NTRS2009

Modeling, Detection, and Disambiguation of Sensor Faults for Aerospace Applications

Sensor faults continue to be a major hurdle for systems health management to reach its full potential. At the same time, few recorded instances of sensor faults exist. It is equally difficult to seed particular sensor faults. Therefore, research is underway to better understand the different fault modes seen in sensors and to model the faults. The fault models can then be used in simulated sensor fault scenarios to ensure that algorithms can distinguish between sensor faults and system faults. The paper illustrates the work with data collected from an electro-mechanical actuator in an aerospace setting, equipped with temperature, vibration, current, and position sensors. The most common sensor faults, such as bias, drift, scaling, and dropout were simulated and injected into the experimental data, with the goal of making these simulations as realistic as feasible. A neural network based classifier was then created and tested on both experimental data and the more challenging randomized data sequences. Additional studies were also conducted to determine sensitivity of detection and disambiguation efficacy to severity of fault conditions.

Balaban, Edward↗

NASA NTRS2006

Real-Time Diagnosis of Faults Using a Bank of Kalman Filters

A new robust method of automated real-time diagnosis of faults in an aircraft engine or a similar complex system involves the use of a bank of Kalman filters. In order to be highly reliable, a diagnostic system must be designed to account for the numerous failure conditions that an aircraft engine may encounter in operation. The method achieves this objective though the utilization of multiple Kalman filters, each of which is uniquely designed based on a specific failure hypothesis. A fault-detection-and-isolation (FDI) system, developed based on this method, is able to isolate faults in sensors and actuators while detecting component faults (abrupt degradation in engine component performance). By affording a capability for real-time identification of minor faults before they grow into major ones, the method promises to enhance safety and reduce operating costs. The robustness of this method is further enhanced by incorporating information regarding the aging condition of an engine. In general, real-time fault diagnostic methods use the nominal performance of a "healthy" new engine as a reference condition in the diagnostic process. Such an approach does not account for gradual changes in performance associated with aging of an otherwise healthy engine. By incorporating information on gradual, aging-related changes, the new method makes it possible to retain at least some of the sensitivity and accuracy needed to detect incipient faults while preventing false alarms that could result from erroneous interpretation of symptoms of aging as symptoms of failures. The figure schematically depicts an FDI system according to the new method. The FDI system is integrated with an engine, from which it accepts two sets of input signals: sensor readings and actuator commands. Two main parts of the FDI system are a bank of Kalman filters and a subsystem that implements FDI decision rules. Each Kalman filter is designed to detect a specific sensor or actuator fault. When a sensor or actuator fault occurs, large estimation errors are generated by all filters except the one using the correct hypothesis. By monitoring the residual output of each filter, the specific fault that has occurred can be detected and isolated on the basis of the decision rules. A set of parameters that indicate the performance of the engine components is estimated by the "correct" Kalman filter for use in detecting component faults. To reduce the loss of diagnostic accuracy and sensitivity in the face of aging, the FDI system accepts information from a steady-state-condition-monitoring system. This information is used to update the Kalman filters and a data bank of trim values representative of the current aging condition.

Kobayashi, Takahisa↗

NASA NTRS2005

Tool for Viewing Faults Under Terrain

Multi Surface Light Table (MSLT) is an interactive software tool that was developed in support of the QuakeSim project, which has created an earthquake- fault database and a set of earthquake- simulation software tools. MSLT visualizes the three-dimensional geometries of faults embedded below the terrain and animates time-varying simulations of stress and slip. The fault segments, represented as rectangular surfaces at dip angles, are organized into collections, that is, faults. An interface built into MSLT queries and retrieves fault definitions from the QuakeSim fault database. MSLT also reads time-varying output from one of the QuakeSim simulation tools, called "Virtual California." Stress intensity is represented by variations in color. Slips are represented by directional indicators on the fault segments. The magnitudes of the slips are represented by the duration of the directional indicators in time. The interactive controls in MSLT provide a virtual track-ball, pan and zoom, translucency adjustment, simulation playback, and simulation movie capture. In addition, geographical information on the fault segments and faults is displayed on text windows. Because of the extensive viewing controls, faults can be seen in relation to one another, and to the terrain. These relations can be realized in simulations. Correlated slips in parallel faults are visible in the playback of Virtual California simulations.

Siegel, Herbert, L.↗

DOE OSTI2022

What We Learned From Analyzing 18 Million Rows of Commercial Buildings’ HVAC Fault Data

To achieve ambitious decarbonization goals it is critical that buildings operate to their full potential. Commercial HVAC systems, however, experience a wide range of operational faults, adversely affecting energy consumption, occupant comfort, and maintenance costs. Analytical tools such as fault detection & diagnostics (FDD) software identify and help diagnose these types of sensing, mechanical, or control-related faults. While significant energy savings has been documented for FDD, along with limited-scale studies on technical capabilities, there is a lack of empirical data on faults being reported by FDD tools. With FDD deployment accelerating significantly over the past decade there is an opportunity to gather and analyze data on commercial HVAC operational problems at an unprecedented scale. Such data could address many questions such as: [a] What faults are most commonly reported?; and [b] How does fault reporting vary by time of year and other possible drivers? A recent study into FDD fault reporting amassed the largest U.S. dataset of commercial HVAC air-side fault records, drawn from multi-year monitoring across over 60,000 pieces of HVAC equipment. The results of this study provide granular data on fault reporting for over 90 unique fault types. In this paper we provide an overview of the research process and highlight key findings and lessons learned. This study presents an extraordinary level of detail on FDD fault reporting characteristics across many climate zones and building types. Armed with these new insights, commercial building industry stakeholders can make better informed decisions when designing, configuring, and operating commercial HVAC systems.

Crowe, Eliot↗

DOE OSTI2021

A propagation-based fault detection and discrimination method and the optimization of sensor deployment

Industrial processes can be affected by faults having a serious impact on operation when not promptly detected and diagnosed. Here in this paper, a propagation-based fault detection and discrimination(PFDD) method is proposed to develop a strategy for fault diagnosis while in the design phase of a system. The PFDD method constructs the system model using the Integrated System Fault Analysis(ISFA) technique. Based on the system model, the propagation of hardware and software faults are simulated qualitatively. Given the results of the simulation, the process by which a fault propagates can be characterized using the qualitative features of system variables including the deviation of the system variables from their expected values, the variation of the system variables over time, and the order in which each variable is influenced during the propagation of the fault. The strategy by which a fault can be detected and discriminated is defined using those features. The PFDD method supports the detection and discrimination of faults in both steady states and transient states. Based on the PFDD method, the optimization of sensor deployment in a system is discussed. A brute force algorithm is developed to examine the system’s capability at diagnosing faults and the cost of sensor deployment for all possible configurations of sensors. The optimal sensor deployment strategy can be derived accordingly. However, the brute force method is only applicable to small-scale systems due to its high computational cost. A genetic algorithm is used to optimize sensor deployment in large-scale systems. The PFDD and sensor deployment optimization methods are applied to the Experimental Breeder Reactor II (EBR-II) for verification.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2021

Development of a Unified Taxonomy for HVAC System Faults

Detecting and diagnosing HVAC faults is critical for maintaining building operation performance, reducing energy waste, and ensuring indoor comfort. An increasing deployment of commercial fault detection and diagnostics (FDD) software tools in commercial buildings in the past decade has significantly increased buildings’ operational reliability and reduced energy consumption. A massive amount of data has been generated by the FDD software tools. However, efficiently utilizing FDD data for ‘big data’ analytics, algorithm improvement, and other data-driven applications is challenging because the format and naming conventions of those data are very customized, unstructured, and hard to interpret. This paper presents the development of a unified taxonomy for HVAC faults. A taxonomy is an orderly classification of HVAC faults according to their characteristics and causal relations. The taxonomy includes fault categorization, physical hierarchy, fault library, relation model, and naming/tagging scheme. The taxonomy employs both a physical hierarchy of HVAC equipment and a cause-effect relationship model to reveal the root causes of faults in HVAC systems. A structured and standardized vocabulary library is developed to increase data representability and interpretability. The developed fault taxonomy can be used for HVAC system ‘big data’ analytics such as HVAC system fault prevalence analysis or the development of an HVAC FDD software standard. A common type of HVAC equipment-packaged rooftop unit (RTU) is used as an example to demonstrate the application of the developed fault taxonomy. Two RTU FDD software tools are used to show that after mapping FDD data according to the taxonomy, the meta-analysis of the multiple FDD reports is possible and efficient.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗

DOE OSTI2025

A quantitative risk assessment framework for fault reactivation in underground hydrogen storage: Coupled simulation and deep learning approach

Underground hydrogen storage (UHS) is emerging as a critical solution for large-scale energy storage. However, like all subsurface fluid injection activities, UHS poses the risk of injection-induced fault reactivation. Accurate risk assessment is essential to ensuring the safety and efficiency of UHS operations. This study presents the development of deep-learning surrogate models for fault reactivation prediction in UHS, trained on a comprehensive database of fully coupled fluid flow-geomechanics simulations. Our findings reveal that analytical models often yield unreliable estimates, with errors up to 54% in the allowable injection pressure, potentially leading to a 40% reduction in UHS operational capacity. The developed surrogate models were incorporated into a quantitative risk assessment (QRA) framework, enabling probabilistic evaluation of fault reactivation risk while accounting for uncertainties in the input variables. Site-specific features, such as horizontal stress gradients, fault’s dip and strike angles, and operational parameters like bottom-hole injection pressure and well-fault distance, were identified as the primary drivers of fault reactivation across various stress regimes. Whereas other hydraulic, geological, and poroelastic reservoir properties were found to have a secondary impact. Notably, we observed that the risk of fault reactivation for a critically oriented fault with a static friction coefficient greater than 0.55 remains below 10% in a normal faulting stress regime. However, the risk significantly increases as the stress regime transitions from normal to strike-slip and ultimately to reverse faulting conditions. These findings underscore the importance of rigorous site characterization and comprehensive QRA evaluations to optimize UHS performance and minimize geomechanical risks.

25 ENERGY STORAGE↗

DOE OSTI2018

3-D Modeling of Induced Seismicity Along Multiple Faults: Magnitude, Rate, and Location in a Poroelasticity System

Understanding of the potential to injection–induced seismicity along faults requires the response of fault zone system to spatiotemporal perturbations in pore pressure and stress. In this study, three–dimensional (3–D) model system consisting of the caprock, reservoir, and basement is intersected by vertical strike–slip faults. We examine the full poroelastic behavior of the formation and perform the mechanical analysis along each fault zone using the Coulomb stress change. The magnitude, rate, and location of potential earthquakes are predicted using the spatial distribution of stresses and pore pressure over time. Rapid diffusion of pore pressure into conductive faults initiates failure, but the majority of induced seismicity occurs at deep fault zones due to poroelastic stabilization near the injection interval. Less permeable faults can be destabilized by either delayed pore pressure diffusion or poroelastic stressing. A two–dimensional (2–D) horizontal model, representing the interface between the reservoir and the basement, limits diffusion of pore pressure and deformation of the formation in the vertical direction that may overestimate or underestimate the potential of earthquakes along the fault. Lastly, our numerical results suggest that the 3–D modeling of faulting system including poroelastic coupling can reduce the uncertainty in the seismic hazard prediction by considering the hydraulic and mechanical interaction between faults and bounding formations.

3‐D modeling↗

DOE OSTI2022

Precursory Off-Fault Deformation in Restraining and Releasing Step Overs: Insights From Discrete Element Method Models

Accelerating geophysical activity is detected preceding some, but not all, large earthquakes. This observation may indicate that no precursors occur before some earthquakes, or that the instrumentation lacks the required sensitivity. Here, to aid crustal monitoring efforts, we use discrete element method models to identify the locations and styles of deformation that may provide useful information about approaching fault reactivation. We model the reactivation of two healed rough faults in a variety of step over configurations, embedded in a host rock with varying amounts of damage subject to shear velocity loading parallel to the faults. Both the fault geometry and ratio of fault to host rock strength control the amount of off-fault deformation. Consistent with field observations, models with larger steps and more preexisting host rock damage produce higher amounts of off-fault deformation. We assess the size of the continuous regions of high velocities and strains to compare the value of the precursory information of each velocity and strain component. Comparing the three components of the velocity vector suggests that the fault-parallel velocity produces the largest and most temporally continuous regions of elevated velocity. The size of these regions increases toward failure, indicating the usefulness of tracking this component. Comparing the volumetric and shear components of the three-dimensional strain tensor suggests that during most of the interseismic period, the shear strain provides more information about approaching fault slip than the volumetric strain. However, in the days and months preceding fault reactivation, both the shear and volumetric strains provide similarly valuable information.

58 GEOSCIENCES↗

DOE OSTI2024

Accelerating cavity fault prediction using deep learning at Jefferson Laboratory

Abstract Accelerating cavities are an integral part of the continuous electron beam accelerator facility (CEBAF) at Jefferson Laboratory. When any of the over 400 cavities in CEBAF experiences a fault, it disrupts beam delivery to experimental user halls. In this study, we propose the use of a deep learning model to predict slowly developing cavity faults. By utilizing pre-fault signals, we train a long short-term memory-convolutional neural network binary classifier to distinguish between radio-frequency (RF) signals during normal operation and RF signals indicative of impending faults. We optimize the model by adjusting the fault confidence threshold and implementing a multiple consecutive window criterion to identify fault events, ensuring a low false positive rate. Results obtained from analysis of a real dataset collected from the accelerating cavities simulating a deployed scenario demonstrate the model’s ability to identify normal signals with 99.99% accuracy and correctly predict 80% of slowly developing faults. Notably, these achievements were achieved in the context of a highly imbalanced dataset, and fault predictions were made several hundred milliseconds before the onset of the fault. Anticipating faults enables preemptive measures to improve operational efficiency by preventing or mitigating their occurrence.

43 PARTICLE ACCELERATORS↗

DOE OSTI2025

SVM-Based Synchronized Fault Detection for 100% Renewable Microgrids

Traditional protection schemes face significant challenges when applied to microgrids with high penetrations of renewables with inverter-based resources (IBRs). The proliferation of advanced sensing and communication technologies has generated copious data, offering an opportunity to overcome these limitations using data-driven machine learning approaches. This work proposes a novel approach based on a support vector machine (SVM) for detecting faults within a 100% renewable microgrid. The approach encompasses a systematic offline training stage for the development of a linear SVM-based fault detection algorithm. This process covers offline data collection from the microgrid under study, the extraction of features such as positive- and negative-sequence components and the total harmonic distortion of the voltage and current measurements of the relays, and the design of the linear SVM-based classifier. During the online implementation, however, different classifiers can exhibit asynchronicity in detecting the fault inception at different subcycle-to-cycle period-level delays. To circumvent this asynchronicity issue, a separate algorithm is developed for each relay to estimate the fault inception time as close to the real fault time. The performance of the proposed SVM-based synchronized fault detection method is evaluated using online time-domain simulation studies on a microgrid test system. The results corroborate the reliability of the fault detection scheme when tested under various fault cases (fault types, locations, and impedances) and non-fault cases during both grid-tied and islanded operation modes.

100% microgrid↗