Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Performance Tuning”

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 1,657 records · Page 92
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

High-Temperature Heat Transfer Fluid Circulator for Concentrating Solar Power Systems How Mohawk and Velo3Dare Printing the Future of Energy

A high temperature Heat Transfer Fluid circulator compressor with wide turn down ratio aerodynamic performance was designed, fabricated and tested for application with high pressure super Critical CO2 fluid capable of temperatures and pressures to >750°C and 25 MPa respectively. The modular system decoupled the drive motor system from an independent compressor shaft to provide thermal isolation between the high temperature compressor shaft and the lower temperature capable motor elements. While the thermal management approach appears to be viable there are potential losses in the secondary cooling flows that need to be taken into account and fine-tuned during operation in order to achieve desired overall system efficiencies. The system was also shown to be viable and dynamically stable throughout the entire operating speed range, thereby validating the overall system configuration and structural design approach used. Finally, while it was not possible to achieve specific PTC-10 test points, the compressor performance tests provide sufficient data to indicate that the compressor has the ability to achieve the required performance in practice. The two key factors affecting compressor design and cost are the maximum expected temperature and pressure. The maximum allowable temperatures for current state-of-the-art (SOA) motor driven compressor technology is limited to of approximately 150-200°C, but inlet temperatures >750 °C may be encountered. While the optimal efficiency of compression is in the transcritical region, some concepts require inlet pressures to 225 bar.

14 SOLAR ENERGY↗
DOE OSTI2025

Leveraging transfer learning and leaf spectroscopy for leaf trait prediction with broad spatial, species, and temporal applicability

Accurate and reliable prediction of leaf traits is crucial for understanding plant adaptations to environmental variation, monitoring terrestrial ecosystems, and enhancing comprehension of functional diversity and ecosystem functioning. Currently, various approaches (e.g., statistical, physical models) have been developed to estimate leaf traits through hyperspectral remote sensing and leaf spectroscopy. However, the absence of high-performing, transferable, and stable models across various domains of space, plant functional types (PFTs) and seasons hinder our ability to quantify and comprehend spatiotemporal variations in leaf traits. This study proposes robust and highly transferable models for better predicting leaf traits with hyperspectral reflectance. Initially, three datasets were assembled, pairing common leaf traits — chlorophyll (Chla+b), carotenoids (Ccar), leaf mass per area (LAM), equivalent water thickness (EWT) — with leaf spectra measurements collected across diverse geographic locations in the U.S. and Europe, PFTs, and seasons. Measurements were acquired using spectroradiometers (e.g., ASD FieldSpec 3/4/Pro and SVC HR-1024i) with integrating spheres, leaf clips, and contact probes. Here, we then developed transfer learning-based hybrid models that incorporated the domain knowledge of radiative transfer models (RTMs) through pretraining processes and were well-constrained by fine-tuning with field measurements. Through comparison with other state-of-the-art statistical models, including partial-least squares regression (PLSR) and Gaussian Process Regression (GPR), as well as pure physical models, we found that the proposed transfer learning models achieved better predictive performance and higher transferability. Specifically, compared to other statistical models and pure RTMs, the transfer learning model exhibited higher coefficient of determination (R 2 ) values with range of 0.01 to 0.79, lower normalized root mean square error (NRMSE) with range of 0.06 % to 33.25 % in model performance. Additionally, the models exhibited improved transferability, with higher R 2 values range from 0.04 to 0.32, lower NRMSE range from 0.08 % to 30.81 %. The findings underscore that transfer learning models through integrating domain knowledge from RTMs and limited observations, can harness the advantages of both RTMs and statistical models and serve as a promising approach for effectively predicting leaf traits.

59 BASIC BIOLOGICAL SCIENCES↗
NASA NTRS2011

Operational Parameters, Considerations, and Design Decisions for Resource-Constrained Ion Trap Mass Spectrometers

Mass spectrometers are increasingly finding applications in new and unique areas, often in situations where key operational resources (i.e. power, weight and size) are limited. One such example is the Mars Organic Molecule Analyzer (MOMA). This instrument is a joint venture between NASA and the European Space Agency (ESA) to develop an ion trap mass spectrometer for chemical analysis on Mars. The constraints on such an instrument are significant as are the performance requirements. While the ideal operating parameters for an ion trap are generally well characterized, methods to maintain analytical performance with limited power and system weight need to be investigated and tested. Methods Experiments have been performed on two custom ion trap mass spectrometers developed as prototypes for the MOMA instrument. This hardware consists of quadrupole ion trap electrodes that are 70% the size of common commercial instrumentation. The trapping RF voltage is created with a custom tank circuit that can be tuned over a range of RF frequencies and is driven using laboratory supplies and amplifiers. The entire instrument is controlled with custom Lab VIEW software that allows a high degree of flexibility in the definition of the scan function defining the ion trap experiment. Ions are typically generated via an internal electron ionization source, however, a laser desorption source is also in development for analysis of larger intact molecules. Preliminary Data The main goals in this work have been to reduce the power required to generate the radio frequency trapping field used in an ion trap mass spectrometer. Generally minimizing the power will also reduce the volume and mass of the electronics to support the instrument. In order to achieve optimum performance, commercial instruments typically utilize RF frequencies in the 1 MHz range. Without much concern for power usage, they simply generate the voltage required to access the mass range of interest. In order to reduce the required RF voltage (and power), operation of the ion trap at lower RF frequencies has been investigated. Surprisingly, the performance of the instrument has only been slightly degraded at RF frequencies all the way down to 500 kHz. Mass resolution is relatively stable to this point and depending on the resonant ejection point used, the peak intensity is also quite stable. To date only masses up to m/z 200 have been fully investigated, however, additional studies are planned to verify the performance with higher mass ions. The lower frequency and voltage should reduce the pseudo potential well depth, eventually affecting the trapping efficiency of the instrument -- effect that could manifest itself in significantly limiting the mass range of trapped ions. Other methods to reduce the RF power while maintaining analytical performance are also under investigation. This includes ion ejection at lower q(sub z) values to access a given mass with a lower RF voltage. The loss of mass resolution at lower q(sub eject) points has been measured and current work is underway to leverage scan speed and the use of non-linear resonances in order to counter this trend. The overall trap performance under this range of operating conditions will be presented with a goal of identifying what trade-offs are acceptable.

Danell, Ryan M.↗
NASA NTRS2021

Probing the Origin of the Open Circuit Voltage in Perovskite Quantum Dot Photovoltaics

Perovskite quantum dots (PQDs) have many properties that make them attractive for optoelectronic applications, including expanded compositional tunability and crystallographic stabilization. While they have not achieved the same photovoltaic (PV) efficiencies of top-performing perovskite thin films, they do reproducibly show high open circuit voltage (VOC) in comparison. Further understanding of the VOC attainable in PQDs as a function of surface passivation, contact layers, and PQD composition will further progress the field and may lend useful lessons for non-QD perovskite solar cells. Here, we use photoluminescence-based spectroscopic techniques to understand and identify the governing physics of the VOC in CsPbI3 PQDs. In particular, we probe the effect of the ligand exchange and contact interfaces on the VOC and free charge carrier concentration. The free charge carrier concentration is orders of magnitude higher than in typical perovskite thin films and could be tunable through ligand chemistry. Tuning the PQD A-site cation composition via replacement of Cs+ with FA+ maintains the background carrier concentration but reduces the trap density by up to a factor of 40, reducing the VOC deficit. These results dictate how to improve PQD optoelectronic properties and PV device performance and explain the reduced interfacial recombination observed by coupling PQDs with thin-film perovskites for a hybrid absorber layer.

perovskite quantum dot↗
DOE OSTI2022

Automated Control for Nuclear Thermal Propulsion Start-Up using MOOSE-based Applications

This report presents a Griffin/Bison/RELAP-7 numerical model of a prototypical NTP system that features fuel assemblies arranged in rings, and which was designed to simulate rapid startup transients. The physics modeled include full-core neutronics, assembly-wise heat conduction, and conjugate heat transfer, with the balance of plant mainly imposed through boundary conditions. In addition, various forms of automated reactivity control were deployed by using the MOOSE to autonomously drive the model and simulate the reactor transitioning from assumed initial conditions to nominal power in a fraction of a minute. To generate the cross-sections of the neutronics model, and in an effort to simultaneously account for the tremendous axial temperature gradients in the reactor and to limit the number of state points required for cross-section generation, the average component temperatures and hydrogen densities in the cooling channels were correlated to the average fuel and moderator temperatures, and fixed axial profiles were derived for nominal conditions and then used during the transient. With this approximation, a tractable cross-section library tabulated with fuel/moderator temperatures and CD angles was generated using Serpent. The full-core SPH correction procedure and the CD decusping technology in Griffin, respectively, ensure preservation of the multiplication factor and reaction rates at state points, along with a reasonably accurate reactivity worth between tabulated CD angles, despite using a coarse mesh. Feedback from other physics was calculated by modeling one representative fuel assembly per ring, along with the corresponding fuel and moderator cooling channels. To limit power overshoots during startup, another layer of multiphysics coupling was added to the model in order to automatically control the drums. Two different technologies presented herein showed outstanding performance in this regard: (1) a novel hybrid PID controller based on both power and reactivity signals, and (2) a PGC that relies on kinetics parameters and reactivity coefficients to predict future behavior and adjust the desired signal accordingly. A challenging benchmark was devised, featuring a power demand curve that exponentially increases by a factor of 500 within 30 seconds, then levels out after that. Both control approaches create a simulated power curve that closely follows the power demand curve and limits power overshoots to 1% or less. While the former approach requires more tuning of the internal parameters, the latter requires additional knowledge of the reactivity feedback coefficients and rates of change of the corresponding variables, including fuel and moderator temperature, which could be difficult to dynamically measure for a real NTP system. Fortunately, some inaccuracy in these quantities will not drastically degrade the PGC performance. Subsequently, a more realistic startup sequence was considered, in which the mass flow rate and outlet pressures are ramped up to model bootstrap and thrust build-up phases prior to reaching steady-state conditions, demonstrating the ability of the hybrid PID and PGCs to handle such transients, with both types of controllers exhibiting very similar behavior. Nevertheless, a significant chamber temperature overshoot was observed, caused by the demanded power signal and assumed mass flow rate. This issue could be mitigated by deploying a reactor controller that follows the chamber temperature signal and actuates both the control valves and drums (rather than using a power signal based solely on the drums to control reactivity). Enhancement of the hydrogen fluid properties available in MOOSE, as well as a better understanding of prototypical initial conditions, are also needed to further enhance this startup model. Finally, a study was performed to model decay heat post-shutdown, and to prepare for extending this model to predict shutdown behavior and post-shutdown pulsed cooling requirements.

33 ADVANCED PROPULSION SYSTEMS↗
NASA NTRS2015

Significant Attenuation of Lightly Damped Resonances Using Particle Dampers

When equipment designs must perform in a broad band vibration environment it can be difficult to avoid resonances that affect life and performance. This is especially true when an organization seeks to employ an asset from a heritage design in a new, more demanding vibration environment. Particle dampers may be used to provide significant attenuation of lightly damped resonances to assist with such a deployment of assets by including only a very minor set of modifications. This solution may be easier to implement than more traditional attenuation schemes. Furthermore, the cost in additional weight to the equipment can be very small. Complexity may also be kept to a minimum, because the particle dampers do not require tuning. Attenuating the vibratory response with particle dampers may therefore be simpler (in a set it and forget it kind of way) than tuned mass dampers. The paper will illustrate the use of an "equivalent resonance test jig" that can assist designers in verifying the potential resonance attenuation that may be available to them during the early trade stages of the design. An approach is suggested for transforming observed attenuation in the jig to estimated performance in the actual service design. KEY WORDS: Particle Damper, Performance in Vibration Environment, Damping, Resonance, Attenuation, Mitigation of Vibration Response, Response Estimate, Response Verification.

Smith, Andrew↗
NASA NTRS1990

High performance millimeter-wave microstrip circulators and isolators

Millimeter wave systems, phased array antennas, and high performance components all require wideband circulators (and isolators) to perform diplexing and switching, to improve isolation and Voltage Standing Wave Ratio (VSWR), and to construct IMPATT diode reflection amplifiers. Presently, most of the millimeter-wave circulators and isolators are available in the configurations of waveguide or stripline, both of which suffer from the shortcomings of bulky size/weight, narrow bandwidth, and poor compatibility with monolithic millimeter-wave integrated circuits (MMIC). MMW microstrip circulators/isolators can eliminate or improve these shortcomings. Stub-tuned microstrip circulator configuration were developed utilizing the electromagnetic fields perturbation technique, the adhesion problems of microstrip metallization on new ferrite substrate were overcome, the fabrication, assembly, packaging techniques were improved, and then successfully designed, fabricated a Ka band circulator which has isolation and return loss of greater than 16dB, insertion loss less than 0.7dB. To assess the steady and reliable performance of the circulator, a temperature cycling test was done over the range of -20 to +50 C for 3 continuous cycles and found no significant impact or variation of circulator performance.

Shih, Ming↗
DOE OSTI2020

Tuning spin-orbit coupling towards enhancing photocurrent in hybrid organic-inorganic perovskites by using mixed organic cations

The presence of the heavy elements leads to strong spin-orbit coupling (SOC) in hybrid organic-inorganic perovskites (HOIPs), which plays an important role in the photovoltaic performance of HOIPs by changing the populations between bright and dark states. The organic cation is a critical composition and affects SOC via the interaction with the inorganic cage. In this work, we use magneto-photocurrent measurement to explore the SOC effect upon using mixed organic cations (methylammonium (MA) and formamidinium (FA)) with different dipole moments in Pb based perovskites. Magneto-photocurrent measurements showed that the internal magnetic parameter B0 is decreased from 402.41 mT to 180.18 mT and 104.58 mT with decreasing the MA/FA ratio. This provides evidence that changing the internal dipole moment by using mixed organic cations can essentially change the SOC in HOIPs. Simultaneously, the photocurrent is increased from 21.46 mA/cm 2 to 22.60 mA/cm 2 and 23.08 mA/cm 2 when the MA/FA ratio is changed from 1:0 to 0.9:0.1 and 0.7:0.3. Above all, our results indicate that the change in SOC by different organic cations can also be a factor affecting the performance of perovskite solar cells.

36 MATERIALS SCIENCE↗
NASA NTRS1979

Lidar temperature profiling - Performance simulations of Mason's method

In Mason's method (1975) atmospheric temperatures are inferred from a measure of the Boltzmann distribution of rotational states in one of the vibrational bands of O2. Differential absorption is measured using three tunable, narrowband pulse lasers. The outputs of two are tuned to wavelengths at the centers of absorption lines at either end of a particular branch in the band; the third wavelength is in a region of no absorption. The temperature-altitude profile can be calculated from the ratio of the two line absorption coefficients plus a priori knowledge of the line parameters. In the present paper, computer simulations of various lidar configurations are made, using different line pairs in the atmospheric bands of O2 (approximately 630, 690, and 760 nm). Simulated results are presented for temperature profiles measured from a Space Shuttle lidar.

Schwemmer, G. K.↗
NASA NTRS1994

Tuning CAS Application using AIMS: An Automated Instrumentation and Monitoring System

To bring together NASA's scientists and engineers and their counterparts in industry, other government agencies, and academia working in the Computational AeroSciences (CAS) field. This workshop is part of the technology transfer plan of the High Performance Computing and Communications Program (HPCCP). Specific objectives of this Workshop are to: (1) communicate the goals and objectives of HPCCP in the area of CAS; (2) promote and disseminate CAS technology within the appropriate technical communities, including NASA, industry, academia, and other government labs; (3) help promote synergy among CAS scientists; and (4) permit feedback from peer researchers in issues pacing the CAS field in general and the HPCCP CAS program in particular.

Mehra, P.↗
DOE OSTI2024

Analyzing the effect of doping concentration in split-well resonant-phonon terahertz quantum cascade lasers

The effect of doping concentration on the temperature performance of the novel split-well resonant-phonon (SWRP) terahertz quantum-cascade laser (THz QCL) scheme supporting a clean 4-level system design was analyzed using non-equilibrium Green’s functions (NEGF) calculations. Experimental research showed that increasing the doping concentration in these designs led to better results compared to the split-well direct-phonon (SWDP) design, which has a larger overlap between its active laser states and the doping profile. However, further improvement in the temperature performance was expected, which led us to assume there was an increased gain and line broadening when increasing the doping concentration despite the reduced overlap between the doped region and the active laser states. Through simulations based on NEGF calculations we were able to study the contribution of the different scattering mechanisms on the performance of these devices. We concluded that the main mechanism affecting the lasers’ temperature performance is electron-electron (e-e) scattering, which largely contributes to gain and line broadening. Interestingly, this scattering mechanism is independent of the doping location, making efforts to reduce overlap between the doped region and the active laser states less effective. Optimization of the e-e scattering thus could be reached only by fine tuning of the doping density in the devices. By uncovering the subtle relationship between doping density and e-e scattering strength, our study not only provides a comprehensive understanding of the underlying physics but also offers a strategic pathway for overcoming current limitations. This work is significant not only for its implications on specific devices but also for its potential to drive advancements in the entire THz QCL field, demonstrating the crucial role of e-e scattering in limiting temperature performance and providing essential knowledge for pushing THz QCLs to new temperature heights.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2024

Non-Invasive Biophysical Techniques to Monitor the Structural Plasticity of the Photosynthetic Machinery of Live Diatom Cells

The photosynthetic performance of diatoms depends largely on the organization and structural flexibility of their thylakoid membranes, the densely packed, highly organized membrane vesicles in which light reactions of photosynthesis occur. Different regulatory mechanisms that fine tune the photosynthetic functions affect the organization of the photosynthetic machinery at different levels of structural complexity, from the level of individual protein complexes to the macroarray of membrane proteins and the remodeling of the entire thylakoid membrane system. To monitor these reorganizations, non-invasive techniques are of special value. In this chapter, we focus our attention on three of these techniques, which have been demonstrated to provide unique and useful information on the structure and structural and functional plasticity of live diatom cells: (i) circular dichroism (CD) spectroscopy, which has provided unique information on the chiral (macro-)organization of protein complexes and on their rapid, reversible reorganizations, fine-tuning the light-harvesting processes, as well as on variations in the short-range excitonic interactions in the antenna complexes; (ii) small-angle neutron scattering (SANS), which has been used to determine the periodic organization of the thylakoid membranes and to monitor reversible ultrastructural changes on the time-scale of minutes, induced by variations in the environmental conditions such as changes in temperature or light intensity; and (iii) electrochromic shift absorbance transients (ΔA ECS ), a spectroscopic tool which has been shown to be capable of identifying distinct functional groups of the light-harvesting carotenoid fucoxanthin in different diatoms and in cells exposed to different light intensities. Future use of these techniques will most certainly contribute to the deeper understanding of key regulatory mechanisms of photosynthesis in diatoms.

Szabo, Milan↗
DOE OSTI2021

A generalized machine learning workflow to visualize mechanical discontinuity

Accurate detection and mapping of mechanical discontinuity in materials has widespread industrial and research applications. Herein, we developed a generalized machine-learning framework for visualizing single mechanical discontinuity embedded in material of any composition, velocity, density, porosity, and size with limited data. The proposed visualization of discontinuity requires accurate estimations of the length, location, and orientation of the embedded discontinuity by processing multipoint wave-transmission measurements. k-Wave simulator is used to create a large dataset of elastic waveforms recorded during multi-point wave-transmission measurements through materials containing single mechanical discontinuity. k-Wave simulator considers the wave attenuation, dispersion, and mode conversion in wave motion. Discrete wavelet transform (DWT) and statistical feature extraction are essential for data preprocessing prior to the data-driven model development. DWT also minimizes the effect of noise. Using hyper-parameter tuning and cross validation, gradient boosting regression can visualize the mechanical discontinuity with an accuracy of 0.85, in terms of coefficient of determination. A double-layered neural network-based regression has better performance with an accuracy of 0.95. Use of convolutional neural network converts the predictive task from a waveform processing to an image processing problem. Convolutional neural network achieved a generalization performance of 0.91. The proposed generalized workflow requires robust simulation of wave propagation, signal processing, feature engineering, and model evaluation. Sensors closest to the source and those located opposite the source are the most significant for the desired visualization. Notably, the sensors closest to the source capture the non-linear associations, whereas the sensor on the border opposite to the source capture the linear associations between the measured waveforms and the properties of the mechanical discontinuity.

42 ENGINEERING↗
DOE OSTI2020

In Silico Discovery of Covalent Organic Frameworks for Carbon Capture

We screen a database of more than 69 000 hypothetical covalent organic frameworks (COFs) for carbon capture using parasitic energy as a metric. To compute CO 2 –framework interactions in molecular simulations, we develop a genetic algorithm to tune the charge equilibration method and derive accurate framework partial charges. Nearly 400 COFs are identified with parasitic energy lower than that of an amine scrubbing process using monoethanolamine; more than 70 are better performers than the best experimental COFs and several perform similarly to Mg-MOF-74. We analyze the effect of pore topology on carbon capture performance to guide the development of improved carbon capture materials.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2019

Improving the Electronic Transporting Property for Flexible Field-Effect Transistors with Naphthalene Diimide-Based Conjugated Polymer through Branching/Linear Side-Chain Engineering Strategy

n-Type organic/polymeric semiconductors with high electron mobilities are highly demanded for future flexible organic circuits. Except for developing a new conjugated backbone, recent studies show that side-chain engineering also plays an indispensable role in boosting the charge-transporting property. In this paper, we report a new polymer PNDI2T-DTD with a representative n-type naphthalene diimide (NDI)–bithiophene backbone for high-performance n-type flexible thin-film transistors through branching/linear side-chain engineering strategy. Serving as the flexible side chains, the linear/branching side-chain pattern is found to be effective in tuning the preaggregation behavior in solution and the packing ordering of polymeric chains, resulting in the improvement of thin-film crystallinity. We report the electron mobility of the thin film of PNDI2T-DTD on flexible substrates can reach 1.52 cm 2 V –1 s –1 , which is approximately five times higher than that of PNDI2T-DT with the same conjugated backbone and only branching alkyl chains.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Dark matter trigger for early dark energy coincidence

Current cosmological measurements present a persistent tension in the value of the current cosmic expansion rate, the Hubble constant, as inferred from cosmic microwave background (CMB) and large-scale structure (LSS) data compared to that inferred from the classical distance ladder. Early dark energy (EDE), whose cosmological role is localized in time around the epoch of matter-radiation equality just prior to the release of the CMB photons, is designed to resolve this "Hubble tension". However, the model introduces a new coincidence problem: Why should the EDE dynamics occur near matter-radiation equality if EDE is decoupled from both matter and radiation? The resolution of this problem may lie in an early dark sector (EDS), wherein the dark matter mass is dependent on the EDE scalar field. In this work, we construct such an EDS model and show that it naturally resolves the EDE coincidence problem at the background level without any fine-tuning of the coupling to dark matter or of the initial conditions. When fitting to current cosmological data, including that from the local distance ladder, CMB, and LSS, our EDS maximum-likelihood model performs comparably to EDE for resolving the Hubble tension. However, fitting the Planck CMB data requires a specific range of initial field positions to balance the scalar field fluctuations that drive acoustic oscillations, providing testable differences with other EDE models and a platform for future model-building.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2026

PowerModelsGAT-AI: Physics-Informed Graph Attention for Multi-System Power Flow With Continual Learning

Solving the alternating current power flow equations in real time is essential for secure grid operation, yet classical Newton–Raphson solvers can be slow under stressed conditions. Existing graph neural networks for power flow are typically trained on a single system and often degrade on different systems. We present PowerModelsGAT-AI, a physics-informed graph attention network that predicts bus voltages and generator injections. The model uses bus-type-aware masking to handle different bus types and balances multiple loss terms, including a power-mismatch penalty, using learned weights. We evaluate the model on 14 benchmark systems (4 to 6,470 buses) and train a unified model on 13 of these under contingency conditions with up to two branch outages, achieving an average normalized mean absolute error of 0.89% for voltage magnitudes and R 2 >0.99 for voltage angles. We also show continual learning: when adapting a base model to a new 1,354-bus system, standard fine-tuning causes severe forgetting with error increases exceeding 1000% on base systems, while our experience replay and elastic weight consolidation strategy keeps error increases below 2% and in some cases improves base-system performance. Interpretability analysis shows that learned attention weights correlate with physical branch parameters (susceptance: r=0.38 ; thermal limits: r=0.22 ), and feature importance analysis supports that the model captures established power flow relationships.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2024

Robust and Simple ADMM Penalty Parameter Selection

We present a new method for online selection of the penalty parameter for the alternating direction method of multipliers (ADMM) algorithm. ADMM is a widely used method for solving a range of optimization problems, including those that arise in signal and image processing. In its standard form, ADMM includes a scalar hyperparameter, known as the penalty parameter, which usually has to be tuned to achieve satisfactory empirical convergence. In this work, we develop a framework for analyzing the ADMM algorithm applied to a quadratic problem as an affine fixed point iteration. Using this framework, we develop a new method for automatically tuning the penalty parameter by detecting when it has become too large or small. We analyze this and several other methods with respect to their theoretical properties, i.e., robustness to problem transformations, and empirical performance on several optimization problems. Our proposed algorithm is based on a theoretical framework with clear, explicit assumptions and approximations, is theoretically covariant/invariant to problem transformations, is simple to implement, and exhibits competitive empirical performance.

42 ENGINEERING↗