Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “time to failure”

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 55 records · Page 3
All sourcesNASA NTRSDOE OSTI
NASA NTRS1985

Completely monotone regression estimates of software failure rates

A method for estimating the present failure rate of a program is presented. A crude nonparameter estimate of the failure rate function is obtained from past failure times. This estimate is then smoothed by fitting a completely monotonic function, which is the solution of a quadratic programming problem. The value of the smoothed function at present time is used as the estimate of present failure rate. Results of a Monte Carlo study of performance are given.

Miller, D. R.↗
NASA NTRS2002

The Use of Probabilistic Methods to Evaluate the Systems Impact of Component Design Improvements on Large Turbofan Engines

Probabilistic Structural Analysis (PSA) is now commonly used for predicting the distribution of time/cycles to failure of turbine blades and other engine components. These distributions are typically based on fatigue/fracture and creep failure modes of these components. Additionally, reliability analysis is used for taking test data related to particular failure modes and calculating failure rate distributions of electronic and electromechanical components. How can these individual failure time distributions of structural, electronic and electromechanical component failure modes be effectively combined into a top level model for overall system evaluation of component upgrades, changes in maintenance intervals, or line replaceable unit (LRU) redesign? This paper shows an example of how various probabilistic failure predictions for turbine engine components can be evaluated and combined to show their effect on overall engine performance. A generic model of a turbofan engine was modeled using various Probabilistic Risk Assessment (PRA) tools (Quantitative Risk Assessment Software (QRAS) etc.). Hypothetical PSA results for a number of structural components along with mitigation factors that would restrict the failure mode from propagating to a Loss of Mission (LOM) failure were used in the models. The output of this program includes an overall failure distribution for LOM of the system. The rank and contribution to the overall Mission Success (MS) is also given for each failure mode and each subsystem. This application methodology demonstrates the effectiveness of PRA for assessing the performance of large turbine engines. Additionally, the effects of system changes and upgrades, the application of different maintenance intervals, inclusion of new sensor detection of faults and other upgrades were evaluated in determining overall turbine engine reliability.

Packard, Michael H.↗
DOE OSTI2024

Machine Learning and Data Science to Advance Laboratory Earthquake Prediction and Illuminate the Mechanics of Precursors to Failure

Earthquakes represent one of our greatest natural hazards and in recent years human induced seismicity is adding to the threat. Even a modest improvement in the ability to forecast devastating large earthquakes or smaller shallow events associated with fluid injection could save thousands of lives and billions of dollars. Current efforts to forecast earthquakes are limited by knowledge of earthquake physics and hampered by a lack of reliable lab or field observations. However, recent work has provided a critical opportunity for advancement. We have found: 1) clear and consistent precursors prior to earthquake-like failure in the laboratory and 2) that lab earthquakes can be predicted using machine learning (ML). These works show that stick-slip failure events –the lab equivalent of earthquakes– are preceded by a cascade of micro-failure events that radiate elastic energy in a manner that foretells catastrophic failure. Remarkably, ML predicts the fault zone stress state, the failure time and in some cases the magnitude of lab earthquakes. In addition, the observations include clear precursors to failure in the form of changes in fault zone properties prior to lab earthquakes. Precursors have been observed in previous laboratory studies but their origin is poorly understood and their possible connection to ML based earthquake prediction is unknown. The work conducted under our project has dramatically expanded these efforts. We have developed an integrated data science approach to illuminate the physics of earthquake precursors and lab earthquake prediction. Our work has accelerated the development of ML, artificial intelligence (AI), and related data science approaches by providing massive data sets that are tightly connected to critical scientific problems and by bringing together leading subject matter experts and data scientists. Earthquake physics involves phenomena that are far from equilibrium. Our work has leveraged data science methods to illuminate these phenomena and investigate how they relate to earthquake prediction. In addition to a large database with many types of labeled events that is available to everyone, our work has advanced the fundamental understanding of seismic forecasting, earthquake physics, and fault rheology

58 GEOSCIENCES↗
NASA NTRS2011

Moisture-Induced TBC Spallation on Turbine Blade Samples

Delayed failure of TBCs is a widely observed laboratory phenomenon, although many of the early observations went unreported. "The weekend effect" or "DeskTop Spallation" (DTS) is characterized by initial survival of a TBC after accelerated laboratory thermal cycling, then failure by exposure to ambient humidity or water. Once initiated, failure can occur quite dramatically in less than a second. To this end, the water drop test and digital video recordings have become useful techniques in studies at NASA (Smialek, Zhu, Cuy), DECHMA (Rudolphi, Renusch, Schuetze), and CNRS Toulouse/SNECMA (Deneux, Cadoret, Hervier, Monceau). In the present study the results for a commercial turbine blade, with a standard EB-PVD 7YSZ TBC top coat and Pt-aluminide diffusion bond monitored by weight change and visual appearance. Failures were distributed widely over a 5-100 hr time range, depending on temperature. At some opportune times, failure was captured by video recording, documenting the appearance and speed of the moisture-induced spallation process. Failure interfaces exhibited alumina scale grains, decorated with Ta-rich oxide particles, and alumina inclusions as islands and streamers. The phenomenon is thus rooted in moisture-induced delayed spallation (MIDS) of the alumina scale formed on the bond coat. In that regard, many studies show the susceptibility of alumina scales to moisture, as long as high strain energy and a partially exposed interface exist. The latter conditions result from severe cyclic oxidation conditions, which produce a highly stressed and partially damaged scale. In one model, it has been proposed that moisture reacts with aluminum in the bond coat to release hydrogen atoms that 'embrittle' the interface. A negative synergistic effect with interfacial sulfur is also invoked.

Smialek, James L.↗
NASA NTRS2011

Moisture-Induced TBC Spallation on Turbine Blade Samples

Delayed failure of TBCs is a widely observed laboratory phenomenon, although many of the early observations went unreported. The weekend effect or DeskTop Spallation (DTS) is characterized by initial survival of a TBC after accelerated laboratory thermal cycling, then failure by exposure to ambient humidity or water. Once initiated, failure can occur quite dramatically in less than a second. To this end, the water drop test and digital video recordings have become useful techniques in studies at NASA (Smialek, Zhu, Cuy), DECHMA (Rudolphi, Renusch, Schuetze), and CNRS Toulouse/SNECMA (Deneux, Cadoret, Hervier, Monceau). In the present study the results for a commercial turbine blade, with a standard EB-PVD 7YSZ TBC top coat and Pt-aluminide diffusion bond coat are reported. Cut sections were intermittently oxidized at 1100, 1150, and 1200 C and monitored by weight change and visual appearance. Failures were distributed widely over a 5-100 hr time range, depending on temperature. At some opportune times, failure was captured by video recording, documenting the appearance and speed of the moisture-induced spallation process. Failure interfaces exhibited alumina scale grains, decorated with Ta-rich oxide particles, and alumina inclusions as islands and streamers. The phenomenon is thus rooted in moisture-induced delayed spallation (MIDS) of the alumina scale formed on the bond coat. In that regard, many studies show the susceptibility of alumina scales to moisture, as long as high strain energy and a partially exposed interface exist. The latter conditions result from severe cyclic oxidation conditions, which produce a highly stressed and partially damaged scale. In one model, it has been proposed that moisture reacts with aluminum in the bond coat to release hydrogen atoms that embrittle the interface. A negative synergistic effect with interfacial sulfur is also invoked.

Smialek, James↗
NASA NTRS2001

Control of Flexible Systems in the Presence of Failures

Control of flexible systems under degradation or failure of sensors/actuators is considered. A Linear Matrix Inequality framework is used to synthesize H(sub infinity)-based controllers, which provide good disturbance rejection while capable of tolerating real parameter uncertainties in the system model, as well as potential degradation or failure of the control system hardware. In this approach, a one-at-a-time failure scenario is considered, wherein no more than one sensor or actuator is allowed to fail at any given time. A numerical example involving control synthesis for a two-dimensional flexible system is presented to demonstrate the feasibility of the proposed approach.

Magahami, Peiman G.↗
DOE OSTI2020

Safety Risk Reliability Model Library

SR2ML is a software package which contains a set of safety and reliability models designed to be interfaced with the INL developed RAVEN code. These models can be employed to perform both static and dynamic system risk analysis and determine risk importance of specific elements of the considered system. Two classes of reliability models have been developed; the first class includes all classical reliability models (Fault-Trees, Event-Trees, Markov models and Reliability Block Diagrams) which have been extended to deal not only with Boolean logic values but also time dependent values. The second class includes several components aging models. Models of these two classes are designed to be included in a RAVEN ensemble model to perform time dependent system reliability analysis (dynamic analysis). Similarly, these models can be interfaced with system analysis codes to determine failure time of systems and evaluate accident progression (static analysis).

Wang, Congjian↗
NASA NTRS2009

Bayes Analysis and Reliability Implications of Stress-Rupture Testing a Kevlar/Epoxy COPV Using Temperature and Pressure Acceleration

Composite Overwrapped Pressure Vessels (COPVs) that have survived a long service time under pressure generally must be recertified before service is extended. Flight certification is dependent on the reliability analysis to quantify the risk of stress rupture failure in existing flight vessels. Full certification of this reliability model would require a statistically significant number of lifetime tests to be performed and is impractical given the cost and limited flight hardware for certification testing purposes. One approach to confirm the reliability model is to perform a stress rupture test on a flight COPV. Currently, testing of such a Kevlar49 (Dupont)/epoxy COPV is nearing completion. The present paper focuses on a Bayesian statistical approach to analyze the possible failure time results of this test and to assess the implications in choosing between possible model parameter values that in the past have had significant uncertainty. The key uncertain parameters in this case are the actual fiber stress ratio at operating pressure, and the Weibull shape parameter for lifetime; the former has been uncertain due to ambiguities in interpreting the original and a duplicate burst test. The latter has been uncertain due to major differences between COPVs in the database and the actual COPVs in service. Any information obtained that clarifies and eliminates uncertainty in these parameters will have a major effect on the predicted reliability of the service COPVs going forward. The key result is that the longer the vessel survives, the more likely the more optimistic stress ratio model is correct. At the time of writing, the resulting effect on predicted future reliability is dramatic, increasing it by about one "nine," that is, reducing the predicted probability of failure by an order of magnitude. However, testing one vessel does not change the uncertainty on the Weibull shape parameter for lifetime since testing several vessels would be necessary.

Phoenix, S. Leigh↗
DOE OSTI2022

In with the old, in with the new: machine learning for time to event biomedical research

The predictive modeling literature for biomedical applications is dominated by biostatistical methods for survival analysis, and more recently some out of the box machine learning approaches. In this article, we show a presentation of a machine learning method appropriate for time-to-event modeling in the area of prostate cancer long-term disease progression. Using XGBoost adapted to long-term disease progression, we developed a predictive model for 118 788 patients with localized prostate cancer at diagnosis from the Department of Veterans Affairs (VA). Our model accounted for patient censoring. Harrell’s c-index for our model using only features available at the time of diagnosis was 0.757 95% confidence interval [0.756, 0.757]. Our results show that machine learning methods like XGBoost can be adapted to use accelerated failure time (AFT) with censoring to model long-term risk of disease progression. Furthermore, the long median survival justifies and requires censoring. Overall, we show that an existing machine learning approach can be used for AFT outcome modeling in prostate cancer, and more generally for other chronic diseases with long observation times.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗
NASA NTRS2003

Failure Criterion For Isotropic Time Dependent Materials Which Accounts for Multi-Axial Loading

The Space Shuttle's Reusable Solid Rocket Motor (RSRM) nozzle program has recently conducted testing to characterize the effects of multi-axial loading, temperature and time on the failure characteristics of TIGA321, EA913NA, EA946 (three filled epoxy adhesives). From the test data a "Multi-Axial, Temperature, and Time Dependent" or MATT failure criterion was developed. It is shown that this criterion simplifies, for constant load and constant load rate conditions, into a form that can be easily used for stress analysis. Failure for TIGA321 and EA913NA are characterized below their glass transition temperature. Failure for EA946 is characterized for conditions that pass through its glass transition. The MATT failure criterion is shown to be accurate for a wide range of conditions for these adhesives.

Richardson, D. E.↗
NASA NTRSDate not supplied

Using Degradation Modeling to Identify Fragile Operational Conditions in Human- and Component-driven Resilience Assessment

Studying failure events shows that many high-impact events result from the complex interactions between precipitating failure events and degraded operational conditions. Often, when a system is put in operations, unforeseen practical realities (e.g., maintenance and/or workforce availability) lead the system to be operated in configurations outside its envisioned nominal range. However, design-time failure models often assume that the failure events are initiated in an idealized, nominal state of system operation, resulting in an incomplete assessment of future risk. To solve this, this paper develops a framework to consider degraded operational performance in scenario-based resilience models which uses a corresponding model of performance degradation to determine the values of deteriorated model parameters in the resilience model. This framework is demonstrated on a remotely-piloted rover to determine the (individual and combined) effect of drive-train wear and operator fatigue on the resilience of the rover to drive-train faults. This demonstration showed the substantial impact that degradation has on resilience, highlighting the need to account for degradation in resilience models–specifically, unconsidered degradation can lead to overestimates of resilience (and thus underestimates of safety margin) and because resilience can degrade prior to visible unreliability, which can lead to an operational environment with a high propensity for high-impact unforeseen failure events.

resilience↗
NASA NTRSDate not supplied

Using Degradation Modeling to Identify Fragile Operational Conditions in Human- and Component-driven Resilience Assessment

Studying failure events shows that many high-impact events result from the complex interactions between precipitating failure events and degraded operational conditions. Often, when a system is put in operations, unforeseen practical realities (e.g., maintenance and/or workforce availability) lead the system to be operated in configurations outside its envisioned nominal range. However, design-time failure models often assume that the failure events are initiated in an idealized, nominal state of system operation, resulting in an incomplete assessment of future risk. To solve this, this paper develops a framework to consider degraded operational performance in scenario-based resilience models which uses a corresponding model of performance degradation to determine the values of deteriorated model parameters in the resilience model. This framework is demonstrated on a remotely-piloted rover to determine the (individual and combined) effect of drive-train wear and operator fatigue on the resilience of the rover to drive-train faults. This demonstration showed the substantial impact that degradation has on resilience, highlighting the need to account for degradation in resilience models--specifically, unconsidered degradation can lead to overestimates of resilience (and thus underestimates of safety margin) and because resilience can degrade prior to visible unreliability, which can lead to an operational environment with a high propensity for high-impact unforeseen failure events.

Daniel Hulse↗
NASA NTRS2009

Bayes Analysis and Reliability Implications of Stress-Rupture Testing a Kevlar/Epoxy COPV using Temperature and Pressure Acceleration

Composite Overwrapped Pressure Vessel (COPVs) that have survived a long service time under pressure generally must be recertified before service is extended. Sometimes lifetime testing is performed on an actual COPV in service in an effort to validate the reliability model that is the basis for certifying the continued flight worthiness of its sisters. Currently, testing of such a Kevlar49(registered TradeMark)/epoxy COPV is nearing completion. The present paper focuses on a Bayesian statistical approach to analyze the possible failure time results of this test and to assess the implications in choosing between possible model parameter values that in the past have had significant uncertainty. The key uncertain parameters in this case are the actual fiber stress ratio at operating pressure, and the Weibull shape parameter for lifetime; the former has been uncertain due to ambiguities in interpreting the original and a duplicate burst test. The latter has been uncertain due to major differences between COPVs in the data base and the actual COPVs in service. Any information obtained that clarifies and eliminates uncertainty in these parameters will have a major effect on the predicted reliability of the service COPVs going forward. The key result is that the longer the vessel survives, the more likely the more optimistic stress ratio is correct. At the time of writing, the resulting effect on predicted future reliability is dramatic, increasing it by about one nine , that is, reducing the probability of failure by an order of magnitude. However, testing one vessel does not change the uncertainty on the Weibull shape parameter for lifetime since testing several would be necessary.

Phoenix, S. Leigh↗
DOE OSTI2024

Chloride-Induced Stress Corrosion Cracking of Friction Stir-Welded 304L Stainless Steel: Effect of Microstructure and Temperature

Dry storage canisters of used nuclear fuels are fabricated using SUS 304L stainless steel. Chloride-induced stress corrosion cracking (CISCC) is one of the major failure modes of dry storage canisters. The cracked canisters can be repaired by friction stir welding (FSW), a low-heat input ‘solid-phase’ welding process. It is important to evaluate the ClSCC resistance of the friction stir welded material. Stress corrosion cracking (SCC) studies were carried out on mill-annealed base materials and friction stir welded 304L stainless U-bend specimens in 3.5% NaCl + 5 N H2SO4 solution at room temperature and boiling MgCl2 solution at 155 °C. The engineering stress on the outer fiber of the FSW U-bend specimen was ~60% higher than that of the base metal (BM). In spite of the higher stress level of the FSW, both materials (FSW and BM) showed almost similar SCC failure times in the two different test solutions. The SCC occurred in the thermo-mechanically affected zone (TMAZ) of the FSW specimens in the 3.5% NaCl + 5 N H2SO4 solution at room temperature, while the stirred zone (SZ) was relatively crack-free. The failure occurred at the stirred zone when tested in the boiling MgCl2 solution. Hydrogen reduction was the cathodic reaction in the boiling MgCl2 solution, which promoted hydrogen-assisted cracking of the heavily deformed stirred zone. The emergence of the slip step followed by passive film rupture and dissolution of the slip step could be the SCC events in the 3.5% NaCl + 5 N H2SO4 solution at room temperature. However, the slip step height was not sufficient to cause passivity breakdown in the fine-grained SZ. Therefore, the SCC occurred in the partially recrystallized softer TMAZ. Overall, the friction-stirred 304L showed higher tolerance to ClSCC than the 304L base metal.

304L austenitic stainless steel↗
NASA NTRS1984

Time dependency of strainrange partitioning life relationships

The effect of exposure time (or creep rate) on the CP life relationship is established by conducting isothermal CP tests at varying exposure times on 316 Ss at 1300 and 1500 F. A reduction in the CP cycle life is observed with an increase in the exposure time of the CP test at a given inelastic strain-range. This phenomenon is characterized by modifying the Manson-Coffin type of CP relationship. Two new life relationships: (1) the Steady State Creep Rate (SSRC) Modified CP life relationship, and (2) the Failure Time (FT) Modified CP life relationship, are developed in this report. They account for the effect of creep rate and exposure time within the CP type of waveform. The reduction in CP cyclic life in the long exposure time tests is attributed to oxidation and the precipitation of carbides along the grain boundaries.

Kalluri, S.↗
NASA NTRS1995

Adaptive Finite Element Methods for Continuum Damage Modeling

The paper presents an application of adaptive finite element methods to the modeling of low-cycle continuum damage and life prediction of high-temperature components. The major objective is to provide automated and accurate modeling of damaged zones through adaptive mesh refinement and adaptive time-stepping methods. The damage modeling methodology is implemented in an usual way by embedding damage evolution in the transient nonlinear solution of elasto-viscoplastic deformation problems. This nonlinear boundary-value problem is discretized by adaptive finite element methods. The automated h-adaptive mesh refinements are driven by error indicators, based on selected principal variables in the problem (stresses, non-elastic strains, damage, etc.). In the time domain, adaptive time-stepping is used, combined with a predictor-corrector time marching algorithm. The time selection is controlled by required time accuracy. In order to take into account strong temperature dependency of material parameters, the nonlinear structural solution a coupled with thermal analyses (one-way coupling). Several test examples illustrate the importance and benefits of adaptive mesh refinements in accurate prediction of damage levels and failure time.

Min, J. B.↗
NASA NTRS1977

Treatment of low strains and long hold times in high temperature metal fatigue by strainrange partitioning

A procedure for treating creep-fatigue for low strainranges and long hold times is outlined. A semi-experimental approach, wherein several cycles of the imposed loading is actually applied to a specimen in order to determine the stable hysteresis loop, can be very useful in the analysis. Because such tests require only a small fraction of the total failure time, they are not inherently prohibitive if experimental equipment is available. The need for accurate constitutive equations is bypassed because the material itself acts to translate the imposed loading into the responsive hysteresis loops. When strainrange partitioning has been applied in such cases very good results have been obtained.

Manson, S. S.↗
NASA NTRS2005

Verification of the Multi-Axial, Temperature and Time Dependent (MATT) Failure Criterion

An extensive test and analytical effort has been completed by the Space Shuttle's Reusable Solid Rocket Motor (KSKM) nozzle program to characterize the failure behavior of two epoxy adhesives (TIGA 321 and EA946). As part of this effort, a general failure model, the "Multi-Axial, Temperature, and Time Dependent" or MATT failure criterion was developed. In the initial development of this failure criterion, tests were conducted to provide validation of the theory under a wide range of test conditions. The purpose of this paper is to present additional verification of the MATT failure criterion, under new loading conditions for the adhesives TIGA 321 and EA946. In many cases, the loading conditions involve an extrapolation from the conditions under which the material models were originally developed. Testing was conducted using three loading conditions: multi-axial tension, torsional shear, and non-uniform tension in a bondline condition. Tests were conducted at constant and cyclic loading rates ranging over four orders of magnitude. Tests were conducted under environmental conditions of primary interest to the RSRM program. The temperature range was not extreme, but the loading ranges were extreme (varying by four orders of magnitude). It should be noted that the testing was conducted at temperatures below the glass transition temperature of the TIGA 321 adhesive. However for the EA946, the testing was conducted at temperatures that bracketed the glass transition temperature.

Richardson, David E.↗