Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “fault 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 1,189 records · Page 66
All sourcesNASA NTRSDOE OSTI
NASA NTRS2014

Advancing Autonomous Operations for Deep Space Vehicles

Starting in Jan 2012, the Advanced Exploration Systems (AES) Autonomous Mission Operations (AMO) Project began to investigate the ability to create and execute "single button" crew initiated autonomous activities [1]. NASA Marshall Space Flight Center (MSFC) designed and built a fluid transfer hardware test-bed to use as a sub-system target for the investigations of intelligent procedures that would command and control a fluid transfer test-bed, would perform self-monitoring during fluid transfers, detect anomalies and faults, isolate the fault and recover the procedures function that was being executed, all without operator intervention. In addition to the development of intelligent procedures, the team is also exploring various methods for autonomous activity execution where a planned timeline of activities are executed autonomously and also the initial analysis of crew procedure development. This paper will detail the development of intelligent procedures for the NASA MSFC Autonomous Fluid Transfer System (AFTS) as well as the autonomous plan execution capabilities being investigated. Manned deep space missions, with extreme communication delays with Earth based assets, presents significant challenges for what the on-board procedure content will encompass as well as the planned execution of the procedures.

Haddock, Angie T.↗
DOE OSTI2025

Vibroacoustic Process Monitoring of 5-Stage Centrifugal Contactors

Vibroacoustic monitoring was investigated to aid in informing process operators with parameters regarding aqueous separation techniques. These techniques have a wide range of applications including nuclear fuel reprocessing. A small-scale solvent extraction system was set up with five magnetic drive contactors and two piston pumps to collect realistic operation. Vibration sensors were placed on three of the contactors and both pumps, and a microphone was placed near the system. Contactor rotational speed and pump flow rate were varied with known values to compare against measured vibroacoustic signatures. Spectral analysis allowed for the determination of contactor rotational speed and the ratio of organic to aqueous flow rate, an important parameter for ensuring optimal extraction. Vibration metrics such as kurtosis and crest factor provided additional details regarding individual rotational speeds and have potential in examination of faults for predictive maintenance. Overall, results indicate the capability of vibroacoustic monitoring to improve operator awareness and process outcomes.

11 - NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
NASA NTRSDate not supplied

Analysis of a Landing System for Planetary Payloads Utilizing Passive Energy Absorbing Composite Structure

Delivery of a payload from space to a planetary surface currently requires the development of an application specific landing system to protect the payload from forces imparted during impact with the planet surface. Often, active energy attenuating systems such as retro-rockets, deployable parachutes, and airbags are utilized within these landing systems to reduce landing impact energy. Unfortunately, these active systems come at a cost; active energy attenuating systems are susceptible to system faults which may limit or completely negate their energy attenuating capability. Additionally, components needing to be stowed such as fuel, parachutes, and airbags increase design complexity, cost, and weight. To overcome these limitations, this study examines the potential of passive energy attenuation through energy absorbing structural design and composite materials to mitigate landing loads for small payload planetary delivery. Researchers at the National Aeronautics and Space Administration (NASA) Langley Research Center (LaRC) have conducted extensive research into developing energy absorbing structures and components for the attenuation of impact energy under various loading conditions including aircraft crash and spacecraft impact. The current study leverages this research to design a lightweight planetary delivery system which utilizes unique outer mold line (OML) geometry and passive energy absorbing structural design to limit landing loads across potential planetary surface environments. The OML geometry is designed to control impact orientation and provide self-righting capabilities for slopped impact surfaces. The internal structure is composed of composite material structures arranged to provide energy absorption which is robust to impact angle and impact velocity. The developed planetary delivery design concept will be evaluated using finite element (FE) model analysis. Simulations of landing impacts with representative soil surface environments will be used to characterize the energy absorbing capabilities of the landing system. Sensitivity of predicted impact force to landing environment, impact angle, and impact velocity will be assessed to identify capabilities and limitations of the initial structural design. Results will be used to determine the feasibility of a lightweight composite structure to passively absorb landing energy for robust planetary payload delivery.

Crashworthiness↗
NASA NTRS1974

Tectonic analysis of Baja California and Parras shear belt in Mexico

The author has identified the following significant results. Geological correlation of terrain across the Gulf of California using ERTS-1 imagery revealed significant similarities between Isla Tiburon, Isla Angel de la Guarda, and the San Carlos Range in mainland Mexico. ERTS-1 imagery was used to check the validity of the existence of major trans-Baja fault zones. ERTS-1 imagery also shows that high albedo sediments similar to known late Tertiary marine sediments are widespread in southern and middle Baja and extend in places to the eastern side of the Peninsula. Major faults in northern Mexico and across the border in the United States were mapped, and ample evidence was found that the Parras and parts of the Texas lineament are belts of major transverse shear faults in areas outside the supposed limit of the Texas and Parras lineaments. A fundamental concept which may help explain many complexities in the tectonic development is beginning to emerge: The southwestern part of North America was torn by massive left-lateral shear of transverse trend (east-west) during the compressive stage of the late Mesozoic and early Cenozoic. This tectonic style has changed into tensional rifting (Basin and Range) and right-lateral shear later in the Cenozoic and Quaternary.

Abdel-Gawad, M.↗
DOE OSTI2022

Enabling machine learning-ready HPC ensembles with Merlin

With the growing complexity of computational and experimental facilities, many scientific researchers are turning to machine learning (ML) techniques to analyze large scale ensemble data. With complexities such as multi-component workflows, heterogeneous machine architectures, parallel file systems, and batch scheduling, care must be taken to facilitate this analysis in a high performance computing (HPC) environment. Here, we present Merlin, a workflow framework to enable large ML-friendly ensembles of scientific HPC simulations. By augmenting traditional HPC with distributed compute technologies, Merlin aims to lower the barrier for scientific subject matter experts to incorporate ML into their analysis. As a producer–consumer workflow model, Merlin enables multi-machine, cross-batch job, dynamically allocated yet persistent workflows capable of utilizing surge-compute resources. Key features of Merlin are a flexible HPC-centric interface, low per-task overhead, multi-tiered fault recovery, and a hierarchical sampling algorithm that allows for $\mathscr{O}$(N) task execution and $\mathscr{O}$(N ln N) task queuing to ensembles of millions of tasks. In addition to Merlin’s design, we test the algorithm’s performance in an HPC center and demonstrate the ability to enqueue 40 million simulations in 100 s, with a 30 millisecond per-task overhead that is independent of ensemble size. Finally, we describe some example applications that Merlin has enabled on leadership-class HPC resources, such as the ML-augmented optimization of nuclear fusion experiments and the calibration of infectious disease models to study the progression of and possible mitigation strategies for COVID-19.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Resiliency in numerical algorithm design for extreme scale simulations

Here this work is based on the seminar titled ‘Resiliency in Numerical Algorithm Design for Extreme Scale Simulations’ held March 1–6, 2020, at Schloss Dagstuhl, that was attended by all the authors. Advanced supercomputing is characterized by very high computation speeds at the cost of involving an enormous amount of resources and costs. A typical large-scale computation running for 48 h on a system consuming 20 MW, as predicted for exascale systems, would consume a million kWh, corresponding to about 100k Euro in energy cost for executing 10 23 floating-point operations. It is clearly unacceptable to lose the whole computation if any of the several million parallel processes fails during the execution. Moreover, if a single operation suffers from a bit-flip error, should the whole computation be declared invalid? What about the notion of reproducibility itself: should this core paradigm of science be revised and refined for results that are obtained by large-scale simulation? Naive versions of conventional resilience techniques will not scale to the exascale regime: with a main memory footprint of tens of Petabytes, synchronously writing checkpoint data all the way to background storage at frequent intervals will create intolerable overheads in runtime and energy consumption. Forecasts show that the mean time between failures could be lower than the time to recover from such a checkpoint, so that large calculations at scale might not make any progress if robust alternatives are not investigated. More advanced resilience techniques must be devised. The key may lie in exploiting both advanced system features as well as specific application knowledge. Research will face two essential questions: (1) what are the reliability requirements for a particular computation and (2) how do we best design the algorithms and software to meet these requirements? While the analysis of use cases can help understand the particular reliability requirements, the construction of remedies is currently wide open. One avenue would be to refine and improve on system- or application-level checkpointing and rollback strategies in the case an error is detected. Developers might use fault notification interfaces and flexible runtime systems to respond to node failures in an application-dependent fashion. Novel numerical algorithms or more stochastic computational approaches may be required to meet accuracy requirements in the face of undetectable soft errors. These ideas constituted an essential topic of the seminar. The goal of this Dagstuhl Seminar was to bring together a diverse group of scientists with expertise in exascale computing to discuss novel ways to make applications resilient against detected and undetected faults. In particular, participants explored the role that algorithms and applications play in the holistic approach needed to tackle this challenge. This article gathers a broad range of perspectives on the role of algorithms, applications and systems in achieving resilience for extreme scale simulations. The ultimate goal is to spark novel ideas and encourage the development of concrete solutions for achieving such resilience holistically.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2021

Multiscale Assessment of Caprock Integrity for Geologic Carbon Storage in the Pennsylvanian Farnsworth Unit, Texas, USA

Leakage pathways through caprock lithologies for underground storage of CO2 and/or enhanced oil recovery (EOR) include intrusion into nano-pore mudstones, flow within fractures and faults, and larger-scale sedimentary heterogeneity (e.g., stacked channel deposits). To assess multiscale sealing integrity of the caprock system that overlies the Morrow B sandstone reservoir, Farnsworth Unit (FWU), Texas, USA, we combine pore-to-core observations, laboratory testing, well logging results, and noble gas analysis. A cluster analysis combining gamma ray, compressional slowness, and other logs was combined with caliper responses and triaxial rock mechanics testing to define eleven lithologic classes across the upper Morrow shale and Thirteen Finger limestone caprock units, with estimations of dynamic elastic moduli and fracture breakdown pressures (minimum horizontal stress gradients) for each class. Mercury porosimetry determinations of CO2 column heights in sealing formations yield values exceeding reservoir height. Noble gas profiles provide a “geologic time-integrated” assessment of fluid flow across the reservoir-caprock system, with Morrow B reservoir measurements consistent with decades-long EOR water-flooding, and upper Morrow shale and lower Thirteen Finger limestone values being consistent with long-term geohydrologic isolation. Together, these data suggest an excellent sealing capacity for the FWU and provide limits for injection pressure increases accompanying carbon storage activities.

58 GEOSCIENCES↗
NASA NTRS1973

New aspects on the tectonic of the Alps and the Apennines revealed by ERTS-1 data

An analysis of the tectonic situation in the Alps and Apennine Mountains based on ERTS-1 imagery is presented. It is stated that the ERTS-1 imagery reveals connections between shearing systems of more than regional importance which could not be determined previously. The tracing of locally known fault distances by ERTS-1 imagery is discussed.

Bodechtel, J.↗
NASA NTRS1989

Electrical performance characteristics of high power converters for space power applications

The first goal of this project was to investigate various converters that would be suitable for processing electric power derived from a nuclear reactor. The implementation is indicated of a 20 kHz system that includes a source converter, a ballast converter, and a fixed frequency converter for generating the 20 kHz output. This system can be converted to dc simply by removing the fixed frequency converter. This present study emphasized the design and testing of the source and ballast converters. A push-pull current-fed (PPCF) design was selected for the source converter, and a 2.7 kW version of this was implemented using three 900 watt modules in parallel. The characteristic equation for two converters in parallel was derived, but this analysis did not yield any experimental methods for measuring relative stability. The three source modules were first tested individually and then in parallel as a 2.7 kW system. All tests proved to be satisfactory; the system was stable; efficiency and regulation were acceptable; and the system was fault tolerant. The design of a ballast-load converter, which was operated as a shunt regulator, was investigated. The proposed power circuit is suitable for use with BJTs because proportional base drive is easily implemented. A control circuit which minimizes switching frequency ripple and automatically bypasses a faulty shunt section was developed. A nonlinear state-space-averaged model of the shunt regulator was developed and shown to produce an accurate incremental (small-signal) dynamic model, even though the usual state-space-averaging assumptions were not met. The nonlinear model was also shown to be useful for large-signal dynamic simulation using PSpice.

Stuart, Thomas A.↗
NASA NTRS1992

A demonstration of an intelligent control system for a reusable rocket engine

An Intelligent Control System for reusable rocket engines is under development at NASA Lewis Research Center. The primary objective is to extend the useful life of a reusable rocket propulsion system while minimizing between flight maintenance and maximizing engine life and performance through improved control and monitoring algorithms and additional sensing and actuation. This paper describes current progress towards proof-of-concept of an Intelligent Control System for the Space Shuttle Main Engine. A subset of identifiable and accommodatable engine failure modes is selected for preliminary demonstration. Failure models are developed retaining only first order effects and included in a simplified nonlinear simulation of the rocket engine for analysis under closed loop control. The engine level coordinator acts as an interface between the diagnostic and control systems, and translates thrust and mixture ratio commands dictated by mission requirements, and engine status (health) into engine operational strategies carried out by a multivariable control. Control reconfiguration achieves fault tolerance if the nominal (healthy engine) control cannot. Each of the aforementioned functionalities is discussed in the context of an example to illustrate the operation of the system in the context of a representative failure. A graphical user interface allows the researcher to monitor the Intelligent Control System and engine performance under various failure modes selected for demonstration.

Musgrave, Jeffrey L.↗
NASA NTRS1993

Formal verification of algorithms for critical systems

We describe our experience with formal, machine-checked verification of algorithms for critical applications, concentrating on a Byzantine fault-tolerant algorithm for synchronizing the clocks in the replicated computers of a digital flight control system. First, we explain the problems encountered in unsynchronized systems and the necessity, and criticality, of fault-tolerant synchronization. We give an overview of one such algorithm, and of the arguments for its correctness. Next, we describe a verification of the algorithm that we performed using our EHDM system for formal specification and verification. We indicate the errors we found in the published analysis of the algorithm, and other benefits that we derived from the verification. Based on our experience, we derive some key requirements for a formal specification and verification system adequate to the task of verifying algorithms of the type considered. Finally, we summarize our conclusions regarding the benefits of formal verification in this domain, and the capabilities required of verification systems in order to realize those benefits.

Rushby, John M.↗
NASA NTRS1995

Investigation of air transportation technology at Ohio University: 1993-1994

The Joint University Program (JUP) offers students, faculty, and staff of the Avionics Engineering Center the opportunity to conduct basic research relating to the safety and efficiency of the National Airspace System. During the 1993-1994 year research was continued in a variety of GPS-related technologies and also in a hybrid data uplink system. (1) Processing techniques developed for the GPS attitude determination have been applied to centimeter-level kinematic GPS aircraft positioning for flight reference systems and aircraft autoland and taxiing operations. This resulted in the first real-time kinematic GPS autoland. (2) The fault detection and isolation (FDI) algorithm developed under the Joint University Program has been fully adopted by RTCA Special Committee 159 as the baseline algorithm for sole-means navigation of GPS integrated with GLONASS, LORAN-C, or baro-altimeter. (3) A Satellite Coverage Research Analysis Model (SCRAM) has been developed which provides outage areas, outage dynamics, and availability information for satellite-based navigation systems. RTCA Special Committee 159 has adopted this model for use in their work. (4) A basic receiver structure has been simulated and tested for use with the hybrid data uplink. This uplink utilizes the phase of an existing AM carrier to transmit digital data, thus both amplitude and phase modulation are applied, resulting in a hybrid carrier. The receiver simulation is directly applicable to real-time implementation as a result of the development platform.

Lilley, Robert W.↗
NASA NTRS2013

Spinoff 2012

Topics covered include: Water Treatment Technologies Inspire Healthy Beverages; Dietary Formulas Fortify Antioxidant Supplements; Rovers Pave the Way for Hospital Robots; Dry Electrodes Facilitate Remote Health Monitoring; Telescope Innovations Improve Speed, Accuracy of Eye Surgery; Superconductors Enable Lower Cost MRI Systems; Anti-Icing Formulas Prevent Train Delays; Shuttle Repair Tools Automate Vehicle Maintenance; Pressure-Sensitive Paints Advance Rotorcraft Design Testing; Speech Recognition Interfaces Improve Flight Safety; Polymers Advance Heat Management Materials for Vehicles; Wireless Sensors Pinpoint Rotorcraft Troubles; Ultrasonic Detectors Safely Identify Dangerous, Costly Leaks; Detectors Ensure Function, Safety of Aircraft Wiring; Emergency Systems Save Tens of Thousands of Lives; Oxygen Assessments Ensure Safer Medical Devices; Collaborative Platforms Aid Emergency Decision Making; Space-Inspired Trailers Encourage Exploration on Earth; Ultra-Thin Coatings Beautify Art; Spacesuit Materials Add Comfort to Undergarments; Gigapixel Images Connect Sports Teams with Fans; Satellite Maps Deliver More Realistic Gaming; Elemental Scanning Devices Authenticate Works of Art; Microradiometers Reveal Ocean Health, Climate Change; Sensors Enable Plants to Text Message Farmers; Efficient Cells Cut the Cost of Solar Power; Shuttle Topography Data Inform Solar Power Analysis; Photocatalytic Solutions Create Self-Cleaning Surfaces; Concentrators Enhance Solar Power Systems; Innovative Coatings Potentially Lower Facility Maintenance Costs; Simulation Packages Expand Aircraft Design Options; Web Solutions Inspire Cloud Computing Software; Behavior Prediction Tools Strengthen Nanoelectronics; Power Converters Secure Electronics in Harsh Environments; Diagnostics Tools Identify Faults Prior to Failure; Archiving Innovations Preserve Essential Historical Records; Meter Designs Reduce Operation Costs for Industry; Commercial Platforms Allow Affordable Space Research; Fiber Optics Deliver Real-Time Structural Monitoring; Camera Systems Rapidly Scan Large Structures; Terahertz Lasers Reveal Information for 3D Images; Thin Films Protect Electronics from Heat and Radiation; Interferometers Sharpen Measurements for Better Telescopes; and Vision Systems Illuminate Industrial Processes.

Source record↗
DOE OSTI2023

From Inside Out: How the Buried Interface, Shell Defects, and Surface Chemistry Conspire to Determine Optical Performance in Nonblinking Giant Quantum Dots

“Giant” or core/thick‐shell quantum dots (gQDs) are an important class of solid‐state quantum emitter characterized by strongly suppressed blinking and photobleaching under ambient conditions, and reduced nonradiative Auger processes. Together, these qualities provide distinguishing and useful functionality as single‐ and ensemble‐photon sources. For many applications, operation at elevated temperatures and under intense photon flux is desired, but performance is strongly dependent on the synthetic method employed for thick‐shell growth. Here, a comprehensive analysis of gQD structural properties “from the inside out” as a function of shell‐growth method is reported: successive ionic layer adsorption and reaction (SILAR) and high‐temperature continuous injection (HT‐CI), or sequential combinations of the two. Key correlations across synthesis methods, structural features (interfacial alloying, stacking‐fault density and surface‐ligand identity), and performance metrics (quantum yield, single‐gQD photoluminescence under thermal/photo stress, charging behavior and quantum‐optical properties) are identified. Surprisingly, it is found that interfacial alloying is the strongest indicator of gQD stability under stress, but this parameter is not the determining factor for Auger suppression. Furthermore, quantum yield is strongly influenced by surface chemistry and can approach unity even in the case of high shell‐defect density, while introduction of zinc‐blende stacking faults increases the likelihood that a gQD exhibits charged‐state emission.

77 NANOSCIENCE AND NANOTECHNOLOGY↗
DOE OSTI2021

Event-modeled Risk Assessment Using Linked Diagrams

Event Modeling Risk Assessment using Linked Diagrams (EMRALD) is a software tool developed at INL for researching the capabilities of dynamic PRA (Probabilistic Risk Assessment). In order to promote the effective use of dynamic PRA by the general community, EMRALD focuses on the following key aspects: Simplifying the modeling process by providing a structure that corresponds to traditional PRA modeling methods Providing a user interface (UI) that makes it easy for the user to model and visualize complex interactions Allowing the user to couple with other analysis applications such as physics based simulations. This includes one-way communication for most applications and two-way loose coupling for customizable applications Providing the sequence and timing of events that lead to the specified outcomes when calculating results Traditional aspects of components with basic events, fault trees, and event trees are all captured in a dynamic framework of state diagrams, which are displayed.

Prescott, SteveR↗
DOE OSTI2022

High-Precision Characterization of Seismicity from the 2022 Hunga Tonga-Hunga Ha'apai Volcanic Eruption

The earthquake swarm accompanying the January 2022 Hunga Tonga-Hunga Ha'apai (HTHH) volcanic eruption includes a large number of posteruptive moderate-magnitude seismic events and presents a unique opportunity to use remote monitoring methods to characterize and compare seismic activity with other historical caldera-forming eruptions. We compute improved epicentroid locations, magnitudes, and regional moment tensors of seismic events from this earthquake swarm using regional to teleseismic surface-wave cross correlation and waveform modeling. Precise relative locations of 91 seismic events derived from 59,047 intermediate-period Rayleigh- and Love-wave cross-correlation measurements collapse into a small area surrounding the volcano and exhibit a southeastern time-dependent migration. Regional moment tensors and observed waveforms indicate that these events have a similar mechanism and exhibit a strong positive compensated linear vector dipole component. Precise relative magnitudes agree with regional moment tensor moment magnitude ($M_w$) estimates while also showing that event sizes and frequency increase during the days after the eruption followed by a period of several weeks of less frequent seismicity of a similar size. The combined information from visual observation and early geologic models indicate that the observed seismicity may be the result of a complex series of events that occurred after the explosive eruption on 15 January, possibly involving rapid resupply of the magma chamber shortly after the eruption and additional faulting and instability in the following weeks. In addition, we identify and characterize an $M_w$ 4.5 event five days before the paroxysmal explosion on 15 January, indicating that additional seismic events preceding the main eruption could have been identified with improved local monitoring. As a result, our analysis of the HTHH eruption sequence demonstrates the value of potentially utilizing teleseismic surface-wave cross correlation and waveform modeling methods to assist in the detailed analysis of remote volcanic eruption sequences.

58 GEOSCIENCES↗
DOE OSTI2024

Photovoltaic Cable Connectors: A Comparative Assessment of the Present State of the Industry [Slides]

The consequences of failure for balance of systems (BoS) components (such as PV cable connectors) include offline module string(s); low system voltage; arc, ground, insulation, and over-temperature faults; triggered fuse(s); system shutdown; and fire. The degradation modes for connectors are studied here through an industry survey and its subsequent examination, which are compared to field-degraded specimens. 117 specimens were obtained from a variety of locations and climates or accelerated tests. A failure analysis for connectors from PV installations was developed (and applied to 54 specimens) including nondestructive examinations (photography, a custom resistance-current scan, and X-ray computed tomography) and destructive examinations (featuring milling of the external plastic, extraction of the internal convolute spring, and potting and polishing in cross-section). Surface and through-thickness composition of the metal pins and springs was quantified using scanning electron microscopy with energy-dispersive X-ray spectroscopy. Fourier-transform infrared spectroscopy was used to verify the base polymer materials and compare the chemical structure of the connector body, bushing, end nut, and o-ring. Thermogravimetric analysis and differential scanning calorimetry were used to further verify the degradation of the same polymeric components. Updated from 2023 NIST/UL Workshop on Photovoltaic Materials Durability (website: https://events.ul.com/WPMogn?rt=aAuoWsl4E0KaORLCMeOgfA) and 2024 PVRW workshop (https://pvrw.nrel.gov/past-proceedings).

14 SOLAR ENERGY↗
DOE OSTI2026

Monitoring of Liquid Metal Reactor Heater Zones with Recurrent Neural Network Learning of Temperature Time Series

Advanced high-temperature fluid reactors (ARs), such as sodium fast reactors (SFRs) and molten salt cooled reactors (MSCRs) utilize high-temperature fluids at ambient pressure. To melt the fluid during reactor startup and prevent fluid freezing during cooldown, the thermal–hydraulic systems of such ARs include heater zones consisting of specific heaters with controllers, temperature sensors, and thermal insulation. The failure of heater zones due to insulation material degradation or improper installation, resulting in parasitic heat losses, can lead to fluid freezing. The detection of faults using a heat-transfer model is difficult because of a lack of knowledge of the experimental details. Data-driven machine learning of heater zone temperature time series offers a viable alternative. In this study, we benchmarked the performance of recurrent neural networks (RNNs) in an analysis of heat-up transient temperature time series of heater zones installed on a liquid sodium vessel. The RNN models include long short-term memory (LSTM) and gated recurrent unit (GRU) networks, as well as their bi-directional variants, BiLSTM and BiGRU. Anomalous temperature points were designated using a percentile-based threshold applied to residual fluctuations in the detrended temperature time series. Additionally, the impact of the exponentially weighted moving average (EWMA) method on detection accuracy was examined. The RNN models’ performance was assessed using precision, recall, and F 1 score metrics. Results demonstrated that RNN models effectively detect anomalies in temperature time series with the best models for each heater zone achieving F 1 scores of over 93%. To explain the variations in RNN model performance across different heater zones, we used Kullback–Leibler (KL) divergence to quantify the relative entropy between training and testing data, and the Detrended Fluctuation Analysis (DFA) to assess long-range temporal correlations. For datasets with strong long-range correlations and minimal relative entropy between training and testing data, GRU is the best-performing model. When the data exhibits weaker long-term correlations and a significant relative entropy between training and testing distributions, BiGRU shows the best performance. For the data sets with intermediate values of both KL divergence and DFA, the best performance is obtained with LSTM and BiLSTM, respectively.

gated recurrent unit↗