Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Failure Rate”

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 217 records · Page 12
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Effects of hydrogen partial pressure on crack initiation and growth rate in vintage X52 steel

There is an increasing interest in ensuring compatibility of existing natural gas infrastructure with conveyance of hydrogen, especially for vintage (pre-1970s) pipeline steels. While hydrogen is known to affect the fatigue and fracture resistance of these materials, there is a lack of data on its effects on crack initiation at low hydrogen partial pressure. To address this gap, the current work presents a series of fatigue tests on vintage API Grade X52 steel in pure nitrogen, air and gaseous hydrogen at partial pressures of 1 and 207 bar. Circumferential notch tension (CNT) specimens with direct current potential difference (DCPD) measurements are used to get S–N curves for crack initiation and failure, and compact tension (CT) tests to assess crack growth rates. Results indicate that air accelerates crack initiation, low hydrogen partial pressure affects the crack growth rate, and higher pressures have the largest impact on both. Surprisingly, crack initiation is faster in air than at 1 bar hydrogen partial pressure. The CNT tests also reveal that the orientation of the banded microstructure has no effect on the crack initiation site, but it influences its growth behavior. Moreover, fractography shows that hydrogen also promotes instances of quasi-cleavage and intergranular fracture, linked to embrittlement of the material. Here, these results demonstrate that, while hydrogen may impact the entirety of the fatigue process, its effects on the early stages of damage accumulation become relevant only at larger pressures. More comprehensive studies of crack initiation in hydrogen atmospheres will be critical to future codes and standards for hydrogen infrastructure.

Circumferential notch tension↗
DOE OSTI2021

Effect of local structure and stoichiometry on the dynamic behavior of bi-metal interfaces

Molecular dynamics (MD) simulations are performed to investigate the failure behavior of Al/Al, Al/Fe, and Al/AlFe bi-layers under high strain-rate shock loading. Additional simulations are performed to invesigate solely the tensile response of these systems and to isolate the effects of loading history on damage and failure. Specifically, the compression stage of the shock is excluded from the tensile simulations. This study shows that local structure variation, including the introduction of serrations, plays different roles in controlling damage nucleation depending on the specific loading condition. Under shock loading, the stress for void nucleation and the resulting void distribution is insensitive to an interface structure, whereas under pure tensile loading, the opposite is true, even though the strain rate is comparable. The above difference can be explained based on the corresponding change in the total deformation prior to damage nucleation under shock loading, which is missing from the pure tensile loading. Therefore, it is concluded that whether local interface structure matters under high strain-rate loading depends on the specific loading history and the stress state: the local structure variation matters only when it alters the deformation behavior, and subsequently, damage nucleation and evolution.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Build plate conduction cooling for thermal management of wire arc additive manufactured components

Wire arc additive manufacturing (WAAM) allows for quick, large component manufacturing with fast deposition rates while leveraging readily available wire feedstock that is significantly cheaper than metal powder. However, the increased deposition rate of this process requires enhanced thermal management as failures can occur due to overheating. A common strategy to mitigate overheating is to dwell, or pause, between individual layers; however, this can significantly increase build times and eliminate the advantage of additive manufacturing being able to manufacture components quickly. To help mitigate this issue, this study explores the use of active cooling to maintain process control and to decrease overall build time. In this work, conductive cooling applied to either the bottom or side of the print substrate was explored. Results from this study showed that bottom build plate active cooling can be used to decrease dwell times by up to 50% and decrease cool-down to room temperature after the building process by up to 75%. Results from this study demonstrate that the use of active cooling strategies for WAAM can be used for better thermal control over the process and should be further investigated.

36 MATERIALS SCIENCE↗
DOE OSTI2021

DEM analysis of the effect of interface transition zone on dynamic splitting tensile behavior of high-strength concrete based on multi-phase model

Highlights: • Crushable aggregates with the real geometry shape are generated by using clump-cluster method. • Flat-Joint Model is adopted as the constitutive law. • The tensile strength is almost not affected by ITZ strength when its strength exceeds 0.8 times that of mortar. • A modified formula that can reflect the functional relationship between the ITZ strength and the TDIF is proposed. The tensile strength is one of the main parameters in the design of concrete structures and has an important influence on the cracking performance and durability of concrete materials. As the weak region in concrete, the strength of the interface transition zone (ITZ) directly affects the dynamic mechanical response of concrete. Meanwhile, considering that the tensile performance of concrete is far weaker than its compressive performance, the study of the strength of ITZ is of great significance to the dynamic tensile fracture characteristics and mechanical properties of concrete. In this paper, a method combining 3D scanning technology and Clump-Cluster was proposed to generate a real geometry shape model of crushable aggregate. On this basis, the 3D discrete element model (DEM) of high-strength concrete was constructed using the Flat-Joint Model (FJM). The effects of the ITZ strength on the failure mode, number of microcracks, tensile strength and strain rate effect of high-strength concrete under strain rates from 0.01/s to 100/s were studied by splitting tensile test. The simulation results indicate that the number of microcracks generated decreases with the increase of ITZ strength, and the proportion of ITZ cracks decreases obviously. When the ITZ strength exceeds the critical value of 0.8 times that of mortar, the tensile strength is almost not affected by ITZ strength. In addition, based on the numerical simulation results, a modified formula that can reflect the functional relationship between the ITZ strength and the dynamic increasing factor of the concrete tensile strength (TDIF) is proposed, and the rationality of the TDIF modified formula is verified.

36 MATERIALS SCIENCE↗
DOE OSTI2020

The Effect of η Phase Precipitates on the Creep Behavior of Alloy 263 and Variants

Phase stability is an important design parameter for Ni-based superalloys to be used in future advanced ultra-supercritical (AUSC) power plants as exposure times in this type of environment are considerable. In this investigation, microstructures based on candidate alloy 263 were obtained with varying amounts of η precipitates using isothermal exposure at 800°C for times ranging from 1,000 h to 10,000 h. The effect of η phase stability on the creep properties was determined using creep specimens isothermally aged at 800°C for 8 h, 3,000 h, 5,000 h and 10,000 h prior to creep screening. The creep life was found to exponentially decrease with increasing density of η phase while the elongation to failure was found to increase. Furthermore, the minimum creep rate was related to the density of η phase; a relationship that did not depend on the alloy formulation. Modification to the Ti and Al concentrations slowed down the γ' to η transformation while doubling the γ' fraction after standard heat treatment. By modifying the Ti and Al content, and thereby improving γ' stability over η, the creep lives of specimens isothermally aged for up to 5,000 h were greater than that of the nominal alloy in its standard aged condition.

36 MATERIALS SCIENCE↗
DOE OSTI2020

The Completed SDSS-IV extended Baryon Oscillation Spectroscopic Survey: Large-scale structure catalogues for cosmological analysis

ABSTRACT We present large-scale structure catalogues from the completed extended Baryon Oscillation Spectroscopic Survey (eBOSS). Derived from Sloan Digital Sky Survey (SDSS) IV Data Release 16 (DR16), these catalogues provide the data samples, corrected for observational systematics, and random positions sampling the survey selection function. Combined, they allow large-scale clustering measurements suitable for testing cosmological models. We describe the methods used to create these catalogues for the eBOSS DR16 Luminous Red Galaxy (LRG) and Quasar samples. The quasar catalogue contains 343 708 redshifts with 0.8 < z < 2.2 over 4808 deg2. We combine 174 816 eBOSS LRG redshifts over 4242 deg2 in the redshift interval 0.6 < z < 1.0 with SDSS-III BOSS LRGs in the same redshift range to produce a combined sample of 377 458 galaxy redshifts distributed over 9493 deg2. Improved algorithms for estimating redshifts allow that 98 per cent of LRG observations result in a successful redshift, with less than one per cent catastrophic failures (Δz > 1000 km s−1). For quasars, these rates are 95 and 2 per cent (with Δz > 3000 km s−1). We apply corrections for trends between the number densities of our samples and the properties of the imaging and spectroscopic data. For example, the quasar catalogue obtains a χ2/DoF = 776/10 for a null test against imaging depth before corrections and a χ2/DoF= 6/8 after. The catalogues, combined with careful consideration of the details of their construction found here-in, allow companion papers to present cosmological results with negligible impact from observational systematic uncertainties.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2020

Probabilistic Assessment of Damage from Same Shock Response Spectra due to Variations in Damping

Interpretation of field data from shock tests and subsequent assessment of product safety margins via laboratory testing are based on the shock response spectra (SRS). The SRS capture how a single degree of freedom (SDOF) structure responds to the shock at differing frequencies and, therefore, no longer contain the duration or other temporal parameters pertaining to the shock. A single duration can often be included in the technical specification or in the recreation of acceleration vs. time history from the specified SRS; however, there is little basis for that beyond technical judgment. The loss of such temporal information can result in the recreated SRS being the same while its effect on a system or component can be different. This paper attempts to quantify this deficiency as well as propose a simple method of capturing damping from shock waves that can allow the original waveform to be more accurately reconstructed from the SRS. In this study the decay rate associated with various frequencies that comprise the overall shock was varied. This variation in the decay rate leads to a variation in the acceleration vs. time history, which can be correlated to a “Damage Index” that captures the fatigue damage imparted to the object under shock. Several waveforms that have the same SRS but varying rates of decay for either high- or low-frequency components of the shock were investigated. The resulting variation in stress cycles and Damage Index is discussed in the context of the lognormal distribution of fatigue failure data. It is proposed that, along with the SRS, the decay rate is also captured to minimize the discrepancy between field data and representative laboratory tests.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Fracture Network Localization Preceding Catastrophic Failure in Triaxial Compression Experiments on Rocks

We quantify the spatial distribution of fracture networks throughout six in situ X-ray tomography triaxial compression experiments on crystalline rocks at confining stresses of 5–35 MPa in order to quantify how fracture development controls the final macroscopic failure of the rock, a process analogous to those that control geohazards such as earthquakes and landslides. Tracking the proportion of the cumulative volume of fractures with volumes >90th percentile to the total fracture volume, ∑ v 90 / v t o t indicates that the fracture networks tend to increase in localization toward these largest fractures for up to 80% of the applied differential stress. The evolution of this metric also matches the evolution of the Gini coefficient, which measures the deviation of a population from uniformity. These results are consistent with observations of localizing low magnitude seismicity before large earthquakes in southern California. In both this analysis and the present work, phases of delocalization interrupt the general increase in localization preceding catastrophic failure, indicating that delocalization does not necessarily indicate a reduction of seismic hazard. However, the proportion of the maximum fracture volume to the total fracture volume does not increase monotonically. Experiments with higher confining stress tend to experience greater localization. To further quantify localization, we compare the geometry of the largest fractures, with volumes >90th percentile, to the best fit plane through these fractures immediately preceding failure. The r 2 scores and the mean distance of the fractures to the plane indicate greater localization in monzonite than in granite. The smaller mean mineral diameter and lower confining stress in the granite experiments may contribute to this result. Tracking these various metrics of localization reveals a close association between macroscopic yielding and the acceleration of fracture network localization. Near yielding, ∑ v 90 / v t o t and the Gini coefficient increase while the mean distance to the final failure plane decreases. Macroscopic yielding thus occurs when the rate of fracture network localization increases.

58 GEOSCIENCES↗
DOE OSTI2024

A systems genetic analysis identifies putative mechanisms and candidate genes regulating vessel traits in poplar wood

Wood is the water conducting tissue of tree stems. Like most angiosperm trees, poplar wood contains water-conducting vessel elements whose functional properties affect water transport and growth rates, as well as susceptibility to embolism and hydraulic failure during water stress and drought. Here we used a unique hybrid poplar pedigree carrying genomically characterized chromosomal insertions and deletions to undertake a systems genomics analysis of vessel traits. We assayed gene expression in wood forming tissues from clonal replicates of genotypes covering dosage quantitative trait loci with insertions and deletions, genotypes with extreme vessel trait phenotypes, and control genotypes. A gene co-expression analysis was used to assign genes to modules, which were then used in integrative analyses to identify modules associated with traits, to identify putative molecular and cellular processes associated with each module, and finally to identify candidate genes using multiple criteria including dosage responsiveness. These analyses identified known processes associated with vessel traits including stress response, abscisic acid and cell wall biosynthesis, and in addition identified previously unexplored processes including cell cycle and protein ubiquitination. We discuss our findings relative to component processes contributing to vessel trait variation including signaling, cell cycle, cell expansion, and cell differentiation.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

Cold Climate Degradation: An Analysis of Double-Axis Tracked, E-W Vertical, and Fixed-Tilt Photovoltaic Deployments in Alaska

As countries around the world transition towards renewable energy, there is increasing interest in using photovoltaic (PV) technologies to help decarbonize remote northern communities due to their scalability and affordability. However, a major barrier towards large-scale adoption of PV in cold climates is performance uncertainty under extreme environmental conditions including snowfall, freeze-thaw cycles, and high wind loads. Existing literature on PV degradation rates in the North is relatively limited, with published degradation rates varying between -0.2%/year (Sweden) to -1.3%/year (Scotland). At this workshop, we will present preliminary results on the long-term performance of two diverse photovoltaic sites located in Fairbanks, Alaska at 64.8 degrees N: a monofacial Al-BSF double-axis tracking site maintained by the Cold Climate Housing Research Center (CCHRC), and a bifacial PERC/SHJ E-W vertical and south-facing fixed-tilt site maintained by the Alaska Center Energy and Power (ACEP). CCHRC data has been collected over a period of 15 years, while ACEP site data has been collected over 4 years. Using the degradation analysis tool, RdTools, we will present annual system degradation rates, seasonal performance ratio, and identify potential cold climate failure mechanisms for commercially available PV technologies. This analysis will add to existing literature by directly comparing the performance of multiple PV configurations in Alaska.

bifacial↗
DOE OSTI2025

AGR TRISO Fuel Performance Modeling in FY-25

This report summarizes the activities performed in FY-25 to support fuel performance modeling for the Advanced Gas Reactor (AGR) Fuel Development and Qualification Program. This includes implementation of a new uranium oxycarbide (UCO) kernel swelling rate model, inner pyrolytic carbon (IPyC) cracking behavior and failure predictions, development of new UCO and silicon carbide (SiC) cesium diffusion parameters, and the thermomechanical behavior of particle layers using experimental data from micro-tensile strength testing.

11 - NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2022

Rethinking PCMI Failures Under Transient Conditions

Under PCMI conditions, the cladding strain is driven by either pellet swelling during normal operation or pellet thermal expansion in transients. The loading in PCMI is displacement controlled because friction between the pellet and the cladding results in a strain compatibility at the pellet/cladding interface. Failure is known to occur under PCMI loading conditions during operation when normal power changes result in release of iodine (and other fission products) causing stress corrosion cracking where the crack-propagation rate is dependent upon the cladding stress but slow relative to strictly mechanical failures. Failures after normal power changes are often referred to as failure by pellet-cladding interaction or PCI to distinguish it from PCMI. Such failures are considered displacement-controlled failures, as noted above, and occur with limited plastic strain. This is different to what is typically thought of as PCMI failure under transient conditions, either Anticipated Operational Occurrences (AOOs) or Reactivity Insertion Accidents (RIAs), which is generally assumed to occur once the cladding reaches a mechanical strain limit. But there are two features in the typical PCMI failure in an RIA that are not consistent with the PCMI assumptions. The first is the ductile shear failure feature in cross section and the second is a large crack opening post failure (e.g., see OI-11 failure in NSRR program). The ductile feature is indicative of a plastic instability that requires very large local strains, and the large crack opening must be the result of a release of energy, both of which are not expected if stress in the cladding was only driven by thermal expansion in the pellet. It is concluded that there must be a mechanism (likely related to rod internal pressure) that eliminates strain compatibility between the pellet and cladding – allowing strain localization to occur in what are known as PCMI transient failures. In some sense, these PCMI failures may actually be early burst failures that occur when thermal expansion causes uniform strains greater than a defined limit and fission gas pressure eliminate the strain compatibility

Cantonwine, Paul↗
DOE OSTI2022

Performance of photosensors in a high-rate environment for gas Cherenkov detectors

We report the solenoidal large intensity device (SoLID) at Jefferson Lab will push the boundaries of luminosity for a large-acceptance detector, which necessitates the use of a light-gas threshold Cherenkov counter for online event selection. Due to the high luminosity, the single-photon background rate in this counter can exceed 160 kHz/cm 2 at the photosensors. Therefore, it is essential to validate the high-rate limits of the planned photosensors and readout electronics in order to mitigate the risk of failure. We report on the design and an early set of studies carried out using a small Cherenkov counter in a high-rate environment up to 60 kHz/cm 2 , in Hall C at Jefferson Lab. Commercially available multi-anode photomultipliers (MaPMT) and low-cost large-area picosecond photodetectors (LAPPD) were tested using the JLab FADC250 modules for readout. The test beam results show that the MaPMT array and the internal stripline LAPPD can detect and identify single-electron and pair-production events in high-rate environments. Due to its higher quantum efficiency, the MaPMT array provided a better separation between the single-electron and the pair-production events compared to the internal stripline LAPPD. A GEANT4 simulation confirms the beam test performance of our counter device.

47 OTHER INSTRUMENTATION↗
DOE OSTI2026

ACT University Collaboration Proposal: Role of manufacturing defects on material failure under dynamic loading for developing enhanced failure models and theories (Final Report)

Research on the tensile behavior of additively manufactured 316L stainless steel coupon specimens at increasing strain rates was conducted over the past year at Penn State. Dynamic loading rates into the 1000 1/s loading rates were performed on specimens with nearly 1.32” gage lengths. Stress-strain plots show ductile and plastic behavior well beyond 12% strain with necked specimens, having smaller effective gage lengths showing up to 56% ultimate strain. Additional tests performed on compact tension specimens helped with simulation work to understand the deformation behavior. Using finite elements, it was possible to determine the feasibility of a comprehensive experimental validation study towards an improved model for AM material failure – specifically, the Bai-Wierzbicki approach which accounts for lode angle and triaxiality. A non-significant number of tests are projected for eight specific failure nodes, with additional replicates to provide strain-rate capabilities to the existing formulation. One such correction factor is explored for quasi-static, notched specimens. Finally, stress intensity factor was explored using a set of compact-tension specimens which would also provide useful validation data for any simulation work.

36 MATERIALS SCIENCE↗
DOE OSTI2020

Challenges and Practical Solutions for Conducting Structural MARSSIM Surveys in Elevated Radon Environments - 20413

The Multi-Agency Radiation Survey and Site Investigation Manual (MARSSIM) [1] provides comprehensive guidance for the development and implementation of radiological surveys conducted to demonstrate compliance with a dose- or risk-based cleanup goal during final status survey. The MARSSIM methodology follows an integrated strategy combining direct measurements and sampling to assess uniform residual contamination, and scanning surveys to identify localized areas of elevated activity. Building surveys assessing residual surface contamination present unique challenges from reliance on field measurements that may be subject to less than ideal environmental conditions. Additionally, limited characterization and excessive conservatism incorporated into the conceptual site model may result in derived concentration guideline levels (DCGL) and, as a consequence of relatively inefficient field instrumentation, corresponding attributable count rates generally comparable to background. The presence of elevated ambient radon concentrations and associated progeny deposition both exacerbate these intrinsic survey challenges and introduce a series of technical, logistic, and administrative obstacles that can wreak havoc on the statistical legitimacy of each component of the MARSSIM process. Short-lived radon progeny including polonium-218, lead-214, bismuth-214, and polonium-214 produce excessive alpha and beta radiation that may decrease instrument sensitivity through an increase in background count rates, and, at worst, may result in incorrect survey unit failure through the simulation of residual surface activity. Aside from the obvious solution of exhaustive radon mitigation, the primary objectives are to distinguish between radon progeny deposition and actual surface activity, and collect sufficient defensible data. The following is a technical discussion of the consequences of performing building MARSSIM surveys in the presence of elevated ambient radon atmospheres, conceptual examples, and an overview of practical strategies to minimize the effects of radon interference while collecting sufficient defensible data to complete the survey. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2020

Failure modes of protection layers produced by atomic layer deposition of amorphous TiO 2 on GaAs anodes

Amorphous titanium dioxide (a-TiO 2 ) films formed by atomic layer deposition can serve as protective coatings for semiconducting photoanodes in water-splitting cells using strongly alkaline aqueous electrolytes. Herein, we experimentally examine the mechanisms of failure for p + -GaAs anodes coated with a-TiO 2 films (GaAs/a-TiO 2 ). Galvanic displacement of exposed GaAs by Au allowed imaging of pinholes in the a-TiO 2 coatings, and enabled collection of quantitative and statistical data associated with pinhole defects. A combination of imaging, electrochemical measurements, and quantitative analyses of corrosion products indicated that extrinsic pinholes were present in the a-TiO 2 films before electrochemical operation. During electrochemical operation these pinholes led to pitting corrosion of the underlying GaAs substrate. The dominant source of pinholes was the presence of atmospheric particulate matter on the GaAs surface during deposition of the a-TiO 2 layer. The pinhole density decreased substantially when the thickness of the a-TiO 2 coating increased beyond 45 nm, and approached zero when the thickness of the film exceeded 112 nm. The density of pinholes in films thinner than 45 nm decreased when the a-TiO 2 coating was deposited in an environmentally controlled cleanroom. Pinhole-free GaAs/a-TiO 2 devices were also tested via chronoamperometry to quantify the rate of pinhole formation during electrochemistry. The time-to-failure increased with thickness, suggesting that the failure mechanism may involve diffusion or migration through the film. However, other mechanisms may also contribute to the degradation of thicker films (>112 nm). Nevertheless, as previously hypothesized, extrinsic pinhole defects formed during deposition and testing control the short-term protective performance of the a-TiO 2 film for GaAs anodes evolving O 2 from water.

25 ENERGY STORAGE↗
DOE OSTI2023

High-Temperature Corrosion of Chromia-Forming Ni-Based Alloys in CO 2 Containing Impurities

Future technologies require structural materials resistant to environmental degradation in high-temperature CO 2 -rich environments. In this work, we exposed several commercially available Ni-based alloys (230, 263, 282, 617, 625, and 740H) to atmospheric pressure gases intended to simulate the compositions expected in future direct-fired supercritical CO 2 power cycles. The alloys were exposed to 95% CO 2 + 4% H 2 O + 1% O 2 and the same gas containing 0.1% SO 2 at temperatures of 600, 650, 700, 750, and 800 °C for 2500 h. With minor exceptions, chromia scales formed on all alloys at all temperatures in the SO 2 -free gas, yielding parabolic growth rates that followed an Arrhenius temperature dependence. Behavior in the SO 2 -containing gas was more complex. Generally, the alloys performed well at temperatures of 650, 750, and 800 °C. While some alloys further performed relatively well across the whole temperature range, several of the alloys experienced chromia failure resulting in non-protective duplex oxide scales and high oxidation rates, at temperatures of 600 and 700 °C. Deviation from protective behavior was associated with internal sulfide formation and, additionally for the case of 600 °C, external sulfate formation. Extensive carburization accompanied growth of the non-protective duplex oxide scales, which made it more difficult for the alloy to recover after initiation of the sulfur-induced accelerated corrosion process. The thermodynamic and kinetic factors influencing the accelerated corrosion in the presence of sulfur are discussed. The results suggest that caution is required when assessing compatibility of Ni-based alloys for CO 2 -based systems when sulfur-based impurities are expected.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Machine Learning Predicts the Timing and Shear Stress Evolution of Lab Earthquakes Using Active Seismic Monitoring of Fault Zone Processes

Abstract Machine learning (ML) techniques have become increasingly important in seismology and earthquake science. Lab‐based studies have used acoustic emission data to predict time‐to‐failure and stress state, and in a few cases, the same approach has been used for field data. However, the underlying physical mechanisms that allow lab earthquake prediction and seismic forecasting remain poorly resolved. Here, we address this knowledge gap by coupling active‐source seismic data, which probe asperity‐scale processes, with ML methods. We show that elastic waves passing through the lab fault zone contain information that can predict the full spectrum of labquakes from slow slip instabilities to highly aperiodic events. The ML methods utilize systematic changes in P‐wave amplitude and velocity to accurately predict the timing and shear stress during labquakes. The ML predictions improve in accuracy closer to fault failure, demonstrating that the predictive power of the ultrasonic signals improves as the fault approaches failure. Our results demonstrate that the relationship between the ultrasonic parameters and fault slip rate, and in turn, the systematically evolving real area of contact and asperity stiffness allow the gradient boosting algorithm to “learn” about the state of the fault and its proximity to failure. Broadly, our results demonstrate the utility of physics‐informed ML in forecasting the imminence of fault slip at the laboratory scale, which may have important implications for earthquake mechanics in nature.

58 GEOSCIENCES↗