Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “AUTOS”

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 649 records · Page 36

Visualization of Jet Impingement and Ignition in a Piston-Cylinder Chamber

The transportation sector accounts for almost a third of the United States' energy consumption. We are developing predictive simulations of the complex in-cylinder processes of internal combustion engines to improve performance and decrease pollutant emissions. We provide a first look at the jet impingement and ignition inside a piston-cylinder chamber using volume rendering. Through the visualization, we constructed a complete 3D picture of the jet impingement on the piston-bowl walls leading to the formation of pockets of auto-ignition of the fuel-air mixture. These auto-ignition pockets subsequently lead to flame stabilization, releasing the chemical energy for conversion to useful work. A detailed physical picture of flame stabilization, as provided by the visualization techniques presented in this study, help in the design of next-generation internal combustion engines.

91 SC - Advanced Scientific Computing Reseach (ASC↗

Phase coherence of parametric-decay modes during high-harmonic fast-wave heating in the National Spherical Torus Experiment

Third-order spectral analysis, in particular the auto bicoherence, was applied to probe signals from high-harmonic fast-wave heating experiments in the National Spherical Torus Experiment. Strong evidence was found for parametric decay of the 30 MHz radio-frequency (RF) pump wave, with a low-frequency daughter wave at 2.7 MHz, the local majority-ion cyclotron frequency. The primary decay modes have auto bicoherence values around 0.85, very close to the theoretical value of one, which corresponds to total phase coherence with the pump wave. The threshold RF pump power for onset of parametric decay was found to be between 200 kW and 400 kW.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Deep Learning-based Non-Stationary Bias Correction (NSBC)

This work develops the NSBC (non-stationary bias correction) methodology to correct temperature projection bias from E3SM. The NSBC deep learning framework consists of a three-part architecture: an auto-encoder for compressing the spatial information, an LSTM for predicting annual temperature mean, and a U-Net for capturing the residual bias in temperature. The non-stationary bias correction (NSBC) framework can correct the non-stationarity of the biases of the climate models, which significantly improves the accuracy of future temperature prediction and improves the overestimation of extreme high temperatures that many existing bias correction methods suffer from. Getting started 1. Obtain the historical climate simulation and observation data. The E3SM simulation data are available through https://aims2.llnl.gov/search/cmip6/. The pseudo observations, the Geophysical Fluid Dynamics Laboratory (GFDL)-ESM4 model (Krasting et al., 2018) are available through https://aims2.llnl.gov/search/cmip6/. The spatial resolution of E3SM and pseudo observation datasets are both regridded to a common 1° resolution grid using conservative interpolation. The regridded E3SM and pseudo observation with 1° resolution can be found throught ./data/. 2. Train the Auto-encoder model. Python 0-autoencoder.py 3. Train the LSTM Python 1-LSTM.py 4. Generate the annual mean temperature based on trained LSTM Python 2-generate_annual_mean_LSTM.py 5. Train the U-Net. Python 3-unet.py 6. Evaluation and compared with the baseline Python 4_evaluation.py Is there a deadline approaching that requires the release of yo

Lucas, Donald↗

Quantification and prediction of error in MIMO tests

Dynamic testing of systems is most realistic of real-world conditions when multiple input and multiple output (MIMO) techniques are used. To replicate measured environmental conditions, a series of desired outputs (responses) on the system must be realized by inverting the frequency response functions (FRF) matrix for input estimation. System identification is dependent on the number of inputs and outputs, which affect the type of solutions. In input estimation, the auto spectra of the determined inputs are affected by both the auto spectra and cross spectra of the desired outputs. This research evaluates errors and investigates two sources in an experimental setting. The first significant error source is test-to-test variability in FRFs from system identification tests, quantified by magnitude variability from long-duration tests. The second significant error source is the realization of time histories from frequency domain inputs. The spectral content of the realized random process deviates from the desired content and is also quantified. The structure is a linear three-story frame fixed at the base; two inputs are provided uniaxially with the use of suspended electrodynamic shakers. The proposed model for error sources was found to be effective in predicting errors in a SISO (single-input single-output) and a square (2-input 2-output) MIMO test.

Yousef, Odey↗

A numerical study of ignition and flame development characteristics in GCI combustion using large eddy simulations and chemical explosive mode analysis

Here, this work investigates the ignition and flame development processes of low reactivity fuel combustion under compression ignition conditions based on the large eddy simulation approach. The chemical explosive mode analysis (CEMA) is employed to characterize the local combustion features, including gas-liquid fuel zone, auto-ignition, diffusion-assisted, extinction, cool flame and post-ignition zone, among which auto-ignition and post-ignition are found to play a key role in the overall heat release process. The local flame propagation modes in gasoline compression ignition (GCI) are determined by quantifying the relative magnitude of diffusion/chemistry at a representative progress variable in the pre-ignition zone. The results show that autoignition fronts and deflagration waves exist simultaneously in the ignition and intense high temperature heat release (HTHR) stages, but autoignition fronts dominate. In addition, the chemical kinetic processes of four heat release periods are analyzed. The heat release during the ignition period is found to be dominated by the reactions CH 3 + H (+M) <=> CH 4 (+M) and CH 3 CHO + H <=> CH 2 CHO + H 2 . The reaction CH 2 OH + OH <=> CH 2 O + H 2 O always plays an important role in the heat releases during the other three combustion stages including intense HTHR, moderate HTHR and post-combustion.

33 ADVANCED PROPULSION SYSTEMS↗

Comparative structural analysis of Bru1 region homeologs in Saccharum spontaneum and S. officinarum

Here, sugarcane is a major sugar and biofuel crop, but genomic research and molecular breeding have lagged behind other major crops due to the complexity of auto-allopolyploid genomes. Sugarcane cultivars are frequently aneuploid with chromosome number ranging from 100 to 130, consisting of 70-80 % S. officinarum , 10-20 % S. spontaneum , and 10 % recombinants between these two species. Analysis of a genomic region in the progenitor autoploid genomes of sugarcane hybrid cultivars will reveal the nature and divergence of homologous chromosomes. As a result, to investigate the origin and evolution of haplotypes in the Bru1 genomic regions in sugarcane cultivars, we identified two BAC clones from S. spontaneum and four from S. officinarum and compared to seven haplotype sequences from sugarcane hybrid R570. The results clarified the origin of seven homologous haplotypes in R570, four haplotypes originated from S. officinarum , two from S. spontaneum and one recombinant.. Retrotransposon insertions and sequences variations among the homologous haplotypes sequence divergence ranged from 18.2 % to 60.5 % with an average of 33. 7 %. Gene content and gene structure were relatively well conserved among the homologous haplotypes. Exon splitting occurred in haplotypes of the hybrid genome but not in its progenitor genomes. Tajima's D analysis revealed that S. spontaneum hapotypes in the Bru1 genomic regions were under strong directional selection. Numerous inversions, deletions, insertions and translocations were found between haplotypes within each genome. In conclusion, this is the first comparison among haplotypes of a modern sugarcane hybrid and its two progenitors. Tajima's D results emphasized the crucial role of this fungal disease resistance gene for enhancing the fitness of this species and indicating that the brown rust resistance gene in R570 is from S. spontaneum . Species-specific InDel, sequences similarity and phylogenetic analysis of homologous genes can be used for identifying the origin of S. spontaneum and S. officinarum haplotype in Saccharum hybrids. Comparison of exon splitting among the homologous haplotypes suggested that the genome rearrangements in Saccharum hybrids S. officinarum would be sufficient for proper genome assembly of this autopolyploid genome. Retrotransposon insertions and sequences variations among the homologous haplotypes sequence divergence may allow sequencing and assembling the autopolyploid Saccharum genomes and the auto-allopolyploid hybrid genomes using whole genome shotgun sequencing.

09 BIOMASS FUELS↗

Utah FORGE: Updated Seismic Event Catalogue from the April, 2022 Stimulation of Well 16A(78)-32

These are revised catalogs, related to the April, 2022 well 16A(78)-32 stimulation (phases 1,2, & 3), provided by Geo Energie Suisse (GES) that include additional events at the start of Stage 1 and some tidying up of some locations. These catalogs also include events for additional events that were auto-located to provide a larger dataset for statistical analyses, like b-value calculations. The actual auto-locations have been removed to prevent spurious location plots being created. Times are recorded in UTC (Coordinate Universal Time), and the coordinate reference system is UTM Zone 12N, NAD83.

15 GEOTHERMAL ENERGY↗

Market Barriers and Drivers for the Next Generation Fault Detection and Diagnostic Tools

Commercial buildings in the U.S. consume as much as 30% excess energy compared to buildings that operate fault free and efficiently. Fault detection and diagnostic (FDD) platforms help to continually identify operational inefficiencies and maintain low-carbon performance. However, the recommendations generated by FDD tools need to be implemented by technicians, resulting in delays or lost savings opportunities. Recent research advances showed fault AUTOcorrection integrating with commercial FDD offerings filled this gap. Seven innovative AUTOcorrection algorithms were integrated into two FDD platforms and deployed across four buildings. The enhanced tools successfully correct faults focusing on incorrectly programmed schedules, override not released, control hunting, rogue zone, and suboptimal setpoints. Although its technical efficacy has been proven in the field, fault AUTO-correction is still early in the deployment cycle and opportunities and barriers need to be understood to reach its full potential in market transformation. This paper broadly introduces the new technology that automatically corrects HVAC faults. The authors describe in detail technology potential, market barriers, and enablers for scalability based on field testing results and interviews with the FDD providers and facility managers. The interviewees agreed that AUTO-correction can reduce the extent to which savings are dependent upon human intervention, scale building operators’ ability to act on FDD findings (especially for facilities with small operation teams), and achieve significant savings. To enable scalable deployment, future efforts are needed to overcome the barriers such as cybersecurity and accountability concerns from building operators and standardization of control parameters used in building automation systems.

Pritoni, Marco↗

Indiana: Indiana's Clean Energy Resources and Economy [Brochure]

Strong in clean energy manufacturing, renewables, and efficiency technologies, Indiana ranks in the top third of states in wind energy potential, with a capacity of more than 148,227 megawatts—enough to generate more than four times the state’s current electricity consumption. The four utility-scale wind power plants currently operating in Indiana generate more than 1,000 megawatts—up from virtually zero in 2005. Several dozen Indiana companies produce components, tools, or materials for the wind energy industry. Indiana’s automotive manufacturing industry employed more than 84,000 people and produced more than $\$$8 billion worth of goods in 2010. Several of Indiana’s auto and auto part manufacturers have entered or are preparing to enter the growing clean transportation sector. Indiana is one of 20 states with a binding energy savings target. The annual savings target starts at 0.3% in 2010 and will ramp up to 2% in 2019. By 2019, this effort is expected to save more than 2.3 million megawatt hours—$\$$177 million worth of electricity—per year statewide. Indiana utility companies also sponsor 43 rebate programs for energy efficiency and renewable energy, as well as a loan program for residential energy efficiency improvements.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Exposure Evaluation for Benzene, Lead and Noise in Vehicle and Equipment Repair Shops

An exposure assessment was performed at the equipment and vehicle maintenance repair shops operating at the U. S. Department of Energy Hanford site, in Richland, Washington. The maintenance shops repair and maintain vehicles and equipment used in support of the Hanford cleanup mission. There are three general mechanic shops and one auto body repair shop. The mechanics work on heavy equipment used in construction, cranes, commercial motor vehicles, passenger-type vehicles in addition to air compressors, generators, and farm equipment. Services include part fabrication, installation of equipment, repair and maintenance work in the engine compartment, and tire and brake services. Work performed at the auto body shop includes painting and surface preparation which involves applying body filler and sanding. 8-hour time-weighted-average samples were collected for benzene and noise exposure and task-based samples were collected for lead dust work activities involving painted metal surfaces. Benzene samples were obtained using 3M™ 3520 sampling badges and were analyzed for additional volatile organic compounds. These compounds were selected based on material safety data sheet information for the aerosol products used by the mechanics for each day of sampling. The compounds included acetone, ethyl ether, toluene, xylene, VM&P naphtha, methyl ethyl ketone, and trichloroethylene. Laboratory data for benzene, VM&P naphtha, methyl ethyl ketone and trichloroethylene were all below the reporting detection limit. Airborne concentrations for acetone, ethyl ether, toluene and xylene were all less than 10% of their occupational exposure limit. The task-based samples obtained for lead dusts were submitted for a metal scan analysis to identify other metals that might be present. Laboratory results for lead dusts were all below the reporting detection limit and airborne concentration for the other metals observed in the samples were less than 10% of the occupational exposure limit. Noise dosimetry sampling was performed on a random basis and was representative of the different work activities within the four shops. Twenty three percent of the noise samples exceeded the occupational exposure limit of 85 decibels for an 8-hour time-weightedaverage. Work activities where noise levels were higher included use of impact wrenches and grinding wheels.

54 ENVIRONMENTAL SCIENCES↗

Building Surface Science Capacity to Serve the Automobile Industry in Southeastern Michigan, final report

This project, “Building Surface Science Capacity to Serve the Automobile Industry in Southeastern Michigan” was carried out in two phases: (1) the 2009 – 2012 renovation of space in the new EMU Science Complex, which included the Surface Science Laboratory (SSL), a very vigorous research lab at EMU that carries on a variety of research projects to serve the auto and other industries in Michigan; and (2) the 2013 purchase of several pieces of equipment to further enhance the research capability of the SSL. The funding granted by the DoE was proposed to “renovate the space in the Science Complex to include SSL and purchase equipment for tribological and electrochemical impedance measurements in the lab, thus SSL will serve the auto and other industries in Michigan better.” We believe we have fully accomplished the mission.

77 NANOSCIENCE AND NANOTECHNOLOGY↗

Fuel Flexible, Low Emission Catalytic Combustor for Opportunity Fuel Applications

Limited fuel resources, increasing energy demand and stringent emission regulations are drivers to evaluate process off-gases or process waste streams as fuels for power generation. Often these process waste streams have low energy content and/or highly reactive components. Operability of low energy content fuels in gas turbines leads to issues such as unstable and incomplete combustion. On the other hand, fuels containing higher-order hydrocarbons lead to flashback and auto-ignition issues. Due to above reasons, these fuels cannot be used directly without modifications or efficiency penalties in gas turbine engines. To enable the use of these wide variety of fuels in gas turbine engines a rich catalytic lean burn (RCL®) combustion system was developed and tested in a subscale high pressure (10 atm.) rig. The RCL® injector provided stability and extended turndown to low Btu fuels due to catalytic pre-reaction. Previous work has shown promise with fuels such as blast furnace gas (BFG) with LHV of 85 Btu/ft3 successfully combusted. This program extends on this work by further modifying the combustor to achieve greater catalytic stability enhancement. Fuels containing low energy content such as weak natural gas with a Lower Heating Value (LHV) of 6.5 MJ/m3 (180 Btu/ft3 to natural gas fuels containing higher hydrocarbon (e.g ethane) with LHV of 37.6 MJ/m3 (1010 Btu/ft3) were demonstrated with improved combustion stability; an extended turndown (defined as the difference between catalytic and non-catalytic lean blow out) of greater than 250oF was achieved with CO and NOx emissions lower than 5 ppm corrected to 15% O2. In addition, for highly reactive fuels the catalytic region preferentially pre-reacted the higher order hydrocarbons with no events of flashback or auto-ignition allowing a stable and safe operation with low NOx and CO emissions.

Eteman, Shahrokh↗

Improving Fatigue Performance of AHSS Welds

Reported herein is technical progress on a U.S. Department of Energy CRADA project with industry cost-share aimed at developing the technical basis and demonstrate the viability of innovative in-situ weld residual stresses mitigation technology that can substantially improve the weld fatigue performance and durability of auto-body structures. The developed technology would be cost-effective and practical in high-volume vehicle production environment. Enhancing weld fatigue performance would address a critical technology gap that impedes the widespread use of advanced high-strength steels (AHSS) and other lightweight materials for auto body structure light-weighting. This means that the automotive industry can take full advantage of the AHSS in strength, durability and crashworthiness without the concern of the relatively weak weld fatigue performance. The project comprises both technological innovations in weld residual stress mitigation and due-diligence residual stress measurement and fatigue performance evaluation. Two approaches were investigated. The first one was the use of low temperature phase transformation (LTPT) weld filler wire, and the second focused on novel thermo-mechanical stress management technique. Both technical approaches have resulted in considerable improvement in fatigue lives of welded joints made of high-strength steels. Synchrotron diffraction measurement confirmed the reduction of high tensile weld residual stresses by the two weld residual stress mitigation techniques.

42 ENGINEERING↗

A New Method for Low Cost Production of Titanium Alloys for Reducing Energy Consumption of Mechanical Systems

This project investigated an innovative manufacturing process intended to minimize the cost of production of titanium materials and components, and increase the adoption of Ti components for energy consuming applications, such as automobiles. A key innovation of the proposed manufacturing approach is a novel Ti powder sintering technology for making titanium materials with ultrafine grain microstructure in the as-sintered state with minimum, or an absence, of post-sintering processes. The new sintering technology is termed Hydrogen Sintering and Phase Transformations (HSPT), and constitutes a promising manufacturing technology that can be used to produce titanium (Ti) materials and components in a near-net-shape form, thus also minimizing machining costs. Our objective was to meet, or possibly surpass, the mechanical property levels for ASTM B348 Grade 5 for wrought Ti-6Al-4V. Although specific applications call for varying mechanical property requirements, ASTM B348 was created for the demanding applications of the aerospace industry, and is the established standard for Ti-6Al-4V. While the primary goal was to meet, or exceed this standard, the team also had the goal of demonstrating this could be done at a significantly lower cost of production. Interim goals of the project were to fully develop this novel sintering process, and provide sufficient baseline testing to make the method practical and attractive to industry. By optimizing the process parameters for the sintering of titanium hydride (TiH 2 ) powders in a hydrogen atmosphere and controlling the phase transformations during and after sintering, the HSPT process was expected to reduce the energy consumption, and thus cost, of making Ti alloys and fabricating Ti components. The process was designed such that no high temperature melting is required for producing Ti alloys; little or no post-sintering processing is needed for producing desired microstructures (and therefore enhanced mechanical properties), and finally, minimum machining is needed to fabricate finished Ti components. An energy analysis within this report provides more detail, but calculated values indicate that the HSPT process is less than half as energy intensive as conventional wrought processing, while producing mechanical properties that are comparable. In addition to the energy savings anticipated from the industrial production of Ti components, a second prong of energy savings resides in the use phase of components produced, primarily from use in the transportation sector. Titanium has a number of material qualities appropriate for the auto industry, particularly low mass and corrosion resistance. By reducing the weight of automobiles and other vehicles, energy costs and CO 2 production will be reduced over the lifetime of the vehicles, and components in corrosive environments on vehicles, such as exhaust systems and other under carriage parts, may not have to be replaced during a vehicle’s lifetime. Our analysis indicates that by replacing only 5.6 kg of steel parts in an auto with Ti components across the entire US fleet would save approximately 486 million gallons of gasoline per year. This correlates to a reduction of 3.6 million metric tons of CO 2 per year. The potential for replacing many more of the steel parts in automobiles with lighter weight titanium components is clear. The project was very successful overall, meeting all milestones and surpassing project goals in terms of mechanical properties and microstructures produced. In addition to tensile properties, fatigue properties were emphasized in the project work. Powder metallurgy processes often have porosity to some degree in their final microstructure, and porosity is a well-known cause of crack initiation and low fatigue performance. Although many automobile applications do not undergo fatigue stress regimes, many others do encounter cyclic stress, and design criteria in the latter case require good fatigue properties. Production and testing of HSPT parts showed excellent tensile properties and fracture toughness, and fatigue properties that exceeded all previously reported powder metallurgy Ti methods, overlapping with wrought processed values. Fatigue limits exceeded 500 MPa and tensile strength exceeded 1,000 MPa while maintaining good ductility. Microstructures produced during the project period easily surpassed pre-project expectations. In addition to producing very fine grains in the as-sintered state (without post sintered thermo-mechanical work), porosity was reduced and industrially relevant microstructures previously undemonstrated in any other powder metallurgy titanium method were produced using HSPT materials. These microstructures, both bi-modal and globularized, were produced with simple post-sinter heat treatments, but without the need for energy intensive mechanical work. The employed heat treatments expanded the available mechanical property range (tensile strength vs. ductility) of the HSPT system in Ti-6Al-4V. The project has resulted in the publication, thus far, of five refereed journal articles and five conference proceedings papers, as well as a patent application, two dissertations and a master’s thesis. Two additional journal articles are currently under review, and at least three others are currently in preparation, with several additional students anticipated to graduate within the coming year. Presentations and papers were a particular focus of the second half of the project, once significant experimentation had been performed and analyzed. As part of our efforts to disseminate information of our results, the Ti research teams within Prof. Fang’s and Prof. Chandran’s research groups had a strong presence at the 13th World Conference on Ti, August 16-20, 2015, in San Diego. Several research groups in the US and in Europe are now performing experiments using the HSPT process. Accompanying efforts to bring HSPT to the Ti community at large, and industry in particular, work has continued with our partners and with other interested industrial Ti users and producers, including Boeing and GKN (a major powder metallurgy parts manufacturer). Commercialization has been a central focus of the final phase of the project, and Reading Alloys signed a provisional licensing agreement in summer of 2015. They are currently seeking an appropriate customer with which to pursue initial parts manufacturing efforts. Other licensing options and partners are continuing to be pursued. The promise of lightweight, strong and corrosion resistant Ti alloys with long fatigue lifetimes for automobile or transportation applications has been the vision of the metal industry since titanium came to the attention of scientists and engineers. The sole limitation of realizing these goals has been cost, which is primarily a function of energy used in production. The HSPT process was shown through this work to be capable of realizing this goal, and facilitating the practical use of titanium in US automotive and other industries.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Aerial Assessment of Liquid in Clouds at Oliktok (Final Campaign Report)

Aerial Assessment of Liquid in Clouds at Oliktok (AALCO) Intensive Operation Period (IOP) began in October, 2016 and ended in October, 2017 at the ARM Mobile Facility-3 (AMF-3) at Oliktok Point, Alaska. The operations tested super-cooled liquid water sensors (SLWCs), leaf-wetness sensors, radiosondes, and a distributed temperature sensor (DTS) on tethered balloon system (TBS) platforms throughout the period. An auto-reeler system, a helikite, and a aerostat were tested. When conditions were optimal, the aerostat was preferred to the helikite and the auto-reeler. It was found the SLWCs had better transmission and sensitivity to relay information about the near-surface cloudy boundary layer than the leaf-wetness sensors. The DTS was also found to give useful information about the atmospheric column and deployment is condition-dependent. Results from the SLWCs and DTS are being compared with high resolution Large Eddy Simulations (LES) in the System for Atmospheric Modeling (SAM).

54 ENVIRONMENTAL SCIENCES↗

Performance of solar leasing for low- and middle-income customers in Connecticut

Policymakers are increasingly interested in expanding access to rooftop photovoltaic systems. This study analyzes the financial performance of a Connecticut Green Bank (CGB) solar leasing program, run in partnership with PosiGen, that targets low- and moderate-income customers. We show that this program has successfully reached underserved customers and has reasonable repayment rates given the credit characteristics of the participants. The CGB/PosiGen program reaches many more underserved customers than other PV financing programs in Connecticut.For example, the majority of CGB/PosiGen participants (58%) live in census tracts that have a median income of less than 80% of the area median income (AMI). In contrast, only 9% of participants in the other CGB solar financing programs live in these census tracts. Furthermore, the majority of CGB/PosiGen participants (56%) have FICO scores that would generally be considered non-prime (<670), whereas only 2% of participants in the other programs have similarly low scores. Credit, not income, is the primary factor that explains participants’ financial performance. Overall, we find that delinquency and annualized losses are higher for PosiGen (2.3% and 0.9%) than for other CGB programs (1.4% and 0.1%). Across the CGB programs, lower credit scores are associated with higher rates of delinquency and loss. Therefore, participants’ lower credit scores explain much of the program’s higher rates of delinquency and annualized losses. PosiGen leases perform competitively with market-rate solar and non-solar leases and loans. When compared to securities backed by market-rate PV loans and leases with similar amounts of seasoning, we find that the PosiGen leases have higher delinquency rates but comparable gross loss rates. The similarity in losses is notable given that loss rates for PosiGen leases were higher than those of the other CGB leases and loans. Rather than the PosiGen leases having high loss rates, other CGB leases and loans have unusually low loss rates. We also compare PosiGen to non-solar benchmarks, including indices of auto and consumer loans. We find that the PosiGen leases have significantly less delinquency than non-prime auto loans and have performance comparable to many consumer loans. The U.S. Department of Energy’s Solar Energy Technologies Office supported this research.

14 SOLAR ENERGY↗

Building Datasets and Training Methods for ML Based Magnet Quench Detection

Detecting quenches in superconducting (SC) magnets during training is a challenging process that involves capturing physical events that occur at different frequencies and appear as various signal features. These events may be correlated across instrumentation type, thermal cycle, and ramp. These events together build a more complete picture of continuous processes occurring in the magnet, and may allow us to flag potential precursors for quench detection. We present our work on building an automatic machine learning (ML) based quench detection system. We build upon our existing work on unsupervised auto-encoders for acoustic sensors and quench antenna (QA) by first establishing a supervised ML training pipeline. We show the results of an event tagging, analysis, and simulation framework on our QA and acoustic data which are used concurrently to build a training dataset for a supervised implementation. We then show how this supervised training can be used as a prior in a semi-supervised framework and compare this to the unsupervised neural network auto-encoder performance.This allows us to have a more concrete understanding of the performance of our algorithms relative to physical events occurring in the magnet, and also provides a baseline software tool to generically evaluate our quench prediction autoencoders under completely unsupervised, supervised, and semi-supervised training conditions.

Khan, Maira [Fermilab]↗