Search NASA⌕ Search

SEARCH · Search NASA

Results for “model-free”

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 127 records · Page 7
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Imaging Wavepackets in Real-Space & Disentangling Ultrafast Solvation Dynamics with High-Energy Ultrafast X-ray Scattering

The microscopic information of solute and solute-solvent motions can be measured using ultrafast diffuse x-ray scattering, and compared to molecular dynamics simulations, as demonstrated in studies of solvated molecular systems done in X-ray free-electron lasers. However, for the typical photon energies used in these experiments, the photon momentum transfer range was limited to lower Q values where the signals arising from different parts of the studied system overlap significantly. Recently, this limitation was greatly alleviated by extending the photon energy at the Linac Coherent Light-Source (LCLS), significantly increasing the accessible momentum transfer range. This improvement is transformative for ultrafast diffuse x-ray scattering measurements, enabling for the first time to disentangle the solute scattering difference signal, that persists to higher Q values, from the bulk solvent and solute-solvent cross-terms difference signals. In addition, it enhances the prediction ability of modeling and simulation methods such as the hybrid QM/MM approach. The extended Q range also opens the way to resolve in real-space details regarding coherent wavepackets motions beyond their center-of-mass positions. Here, we present the first results on high-energy (18keV) diffuse x-ray scattering in solution, demonstrating high fidelity time-resolved scattering and analysis of the photoexcited model photocatalysts PtPOP ([Pt2(POP)4]4-), and IrDimen ([Ir2(dimen)4]2+) in several solvents. These complexes provide ideal systems for demonstrating the ability of high-Q ultrafast scattering and QM/MM simulations to decompose ultrafast solvation dynamics into specific changes in the solute-solvent pair distribution function with a particular focus on how electronically excited states change the interaction between the solvent and photo-catalytically active metal sites. We introduce an approach to further utilize the high-energy capability and demonstrate a single-shot ultra-wide-angle X-ray Scattering modality using two perpendicular detectors, spanning a scattering angle range of more than 100 degrees, allowing to extend the accessible momentum transfer range up to Q~14 Å-1. We develop a model-free method to invert the x-ray scattering signals and enable the recovery of multiple pair-density motions that happen simultaneously, allowing us to directly measure in real-space nuclear wavepacket motions. . [1] Natan, Adi. "Real-Space Inversion and Super-Resolution of Ultrafast X-ray Scattering using Natural Scattering Kernels." arXiv preprint arXiv:2107.05576 (2021)

Natan, Adi↗
DOE OSTI2024

Achieving Cyber-Resilience for Power Systems using a Learning, Model-Assisted Blockchain Framework

The secure integration and management of distributed energy resources (DER) and power aggregators in the electric grid requires secure communications and a physics-aware Command and Control (C2) strategy. A Blockchain (BC)-based overlay network was developed to provide a security layer for the existing power grid network that mitigates risks in current and legacy network and C2 protocols. By integrating a Model-Assisted Machine Learning (MAML) framework with a Secure Blockchain Overlay Network (SBON) a defense-in-depth strategy was achieved. In our approach, the MAML framework leveraged a smart contract framework to gather network data and learn the dynamics of DER to develop detection strategies for attacks targeting sensors and actuators used by DER. The MAML framework learned dynamical systems models for individual DERs to detect sensor attacks. For DER we utilized a Digital Twin (DT) to accelerate the learning process for a model resistant to stealthy attacks. The project created DT for PV inverters and BESS. The DTs were coupled with a model-assisted, data-driven learning of DER behavior. Specifically, we evaluated architectures for model-based learning with model-free fine-tuning. Additionally, differential privacy techniques were used to obfuscate data, while still allowing the computation of attack detection results based on obfuscated data. The SBON developed leverages a private permissioned blockchain network orchestrated with the Hyperledger Fabric framework. To connect the cyber world, which orchestrates the blockchain fabric, and the physical world where the power network resides, we developed a system implementation to enable the secure interaction of the physical world and the abstracted blockchain.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Unraveling the Pyrolytic Behavior and Kinetics of Single Polymers and Plastic-Rich Municipal Solid Waste Using Thermal Analysis

Pyrolysis is a highly promising thermochemical recycling technology for converting heterogenous plastic waste into sustainable fuels in a single step. Therefore, understanding the pyrolysis mechanism is essential for enabling rational reactor design and enhancing efficient recycling techniques. In this study, the thermal degradation behaviors and corresponding kinetics of pure polymers (PE, PP, and PET) and plastic-rich MSW were examined using simultaneous thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC). Experiments were carried out in the temperature range of 30-800°C with variable heating rates from 5°C/min to 20°C/min in an ultra-high purity Argon atmosphere. Our results indicated that the plastic pyrolysis was an endothermic process, with varying decomposition temperature ranges depending on their structure and composition. Various iso-conversional model-free methods (Friedman, Flynn-Wall-Ozawa, Starink, and Kissinger-Akahira-Sunose) were utilized to determine the apparent activation energy of the plastic degradation, which increased in the following order: PET (214 kJ/mol), PP (218 kJ/mol), PE (245 kJ/mol), and MSW (249 kJ/mol). Finally, Criado’s master plots were employed to identify the best-fitting reaction model and the pre-exponential factor was subsequently determined.

Bashir, Muhammad Aamir↗
DOE OSTI2021

Deep Reinforcement Learning for Autonomous Water Heater Control

Electric water heaters represent 14% of the electricity consumption in residential buildings. An average household in the United States (U.S.) spends about USD 400–600 (0.45 ¢/L–0.68 ¢/L) on water heating every year. In this context, water heaters are often considered as a valuable asset for Demand Response (DR) and building energy management system (BEMS) applications. To this end, this study proposes a model-free deep reinforcement learning (RL) approach that aims to minimize the electricity cost of a water heater under a time-of-use (TOU) electricity pricing policy by only using standard DR commands. In this approach, a set of RL agents, with different look ahead periods, were trained using the deep Q-networks (DQN) algorithm and their performance was tested on an unseen pair of price and hot water usage profiles. The testing results showed that the RL agents can help save electricity cost in the range of 19% to 35% compared to the baseline operation without causing any discomfort to end users. Additionally, the RL agents outperformed rule-based and model predictive control (MPC)-based controllers and achieved comparable performance to optimization-based control.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2022

SDSS-IV MaNGA: How the Stellar Populations of Passive Central Galaxies Depend on Stellar and Halo Mass

We analyze spatially resolved and co-added SDSS-IV MaNGA spectra with signal-to-noise ratio ~100 from 2200 passive central galaxies (z ~ 0.05) to understand how central galaxy assembly depends on stellar mass (M*) and halo mass (M h ). We control for systematic errors in M h by employing a new group catalog from Tinker and the widely used Yang et al. catalog. At fixed M*, the strengths of several stellar absorption features vary systematically with M h . Completely model-free, this is one of the first indications that the stellar populations of centrals with identical M* are affected by the properties of their host halos. To interpret these variations, we applied full spectral fitting with the code alf. At fixed M*, centrals in more massive halos are older, show lower [Fe/H], and have higher [Mg/Fe] with 3.5σ confidence. We conclude that halos not only dictate how much M* galaxies assemble but also modulate their chemical enrichment histories. Turning to our analysis at fixed M h , high-M* centrals are older, show lower [Fe/H], and have higher [Mg/Fe] for M h > 10 12 h –1 M⊙ with confidence >4σ. While massive passive galaxies are thought to form early and rapidly, our results are among the first to distinguish these trends at fixed M h . They suggest that high-M* centrals experienced unique early formation histories, either through enhanced collapse and gas fueling or because their halos were early forming and highly concentrated, a possible signal of galaxy assembly bias.

79 ASTRONOMY AND ASTROPHYSICS↗
NASA NTRS1990

Introduction to Fuzzy Set Theory

An introduction to fuzzy set theory is described. Topics covered include: neural networks and fuzzy systems; the dynamical systems approach to machine intelligence; intelligent behavior as adaptive model-free estimation; fuzziness versus probability; fuzzy sets; the entropy-subsethood theorem; adaptive fuzzy systems for backing up a truck-and-trailer; product-space clustering with differential competitive learning; and adaptive fuzzy system for target tracking.

Kosko, Bart↗
NASA NTRS1995

A hybrid neurocomputing/numerical strategy for nonlinear structural analysis

A hybrid neurocomputing/numerical strategy is presented for geometrically nonlinear analysis of structures. The strategy combines model-free data processing capabilities of computational neural networks with a Pade approximants-based perturbation technique to predict partial information about the nonlinear response of structures. In the hybrid strategy, multilayer feedforward neural networks are used to extend the validity of solutions by using training samples produced by Pade approximations to the Taylor series expansion of the response function. The range of validity of the training samples is taken to be the radius of convergence of Pade approximants and is estimated by setting a tolerance on the diverging approximants. The norm of residual vector of unbalanced forces in a given element is used as a measure to assess the quality of network predictions. To further increase the accuracy and the range of network predictions, additional training data are generated by either applying linear regression to weight matrices or expanding the training data by using predicted coefficients in a Taylor series. The effectiveness of the hybrid strategy is assessed by performing large-deflection analysis of a doubly-curved composite panel with a circular cutout, and postbuckling analyses of stiffened composite panels subjected to an in-plane edge shear load. In all the problems considered, the hybrid strategy is used to predict selective information about the structural response, namely the total strain energy and the maximum displacement components only.

Szewczyk, Z. Peter↗
NASA NTRS1998

Decision Processes in Discrimination: Fundamental Misrepresentations of Signal Detection Theory

In the first part of this article, I describe a new approach to studying decision making in discrimination tasks that does not depend on the technical assumptions of signal detection theory (e.g., normality of the encoding distributions). Applying these new distribution-free tests to data from three experiments, I show that base rate and payoff manipulations had substantial effects on the participants' encoding distributions but no effect on their decision rules, which were uniformly unbiased in equal and unequal base rate conditions and in symmetric and asymmetric payoff conditions. In the second part of the article, I show that this seemingly paradoxical result is readily explained by the sequential sampling models of discrimination. I then propose a new, "model-free" test for response bias that seems to more properly identify both the nature and direction of the biases induced by the classical bias manipulations.

Balakrishnan, J. D.↗
NASA NTRS2005

Influence of Coronal Abundance Variations

The PI of this project was Jeff Scargle of NASA/Ames. Co-I's were Alma Connors of Eureka Scientific/Wellesley, and myself. Part of the work was subcontracted to Eureka Scientific via SAO, with Vinay Kashyap as PI. This project was originally assigned grant number NCC2-1206, and was later changed to NCC2-1350 for administrative reasons. The goal of the project was to obtain, derive, and develop statistical and data analysis tools that would be of use in the analyses of high-resolution, high-sensitivity data that are becoming available with new instruments. This is envisioned as a cross-disciplinary effort with a number of "collaborators" including some at SA0 (Aneta Siemiginowska, Peter Freeman) and at the Harvard Statistics department (David van Dyk, Rostislav Protassov, Xiao-li Meng, Epaminondas Sourlas, et al). We have developed a new tool to reliably measure the metallicities of thermal plasma. It is unfeasible to obtain high-resolution grating spectra for most stars, and one must make the best possible determination based on lower-resolution, CCD-type spectra. It has been noticed that most analyses of such spectra have resulted in measured metallicities that were significantly lower than when compared with analyses of high- resolution grating data where available (see, e.g., Brickhouse et al., 2000, ApJ 530,387). Such results have led to the proposal of the existence of so-called Metal Abundance Deficient, or "MAD" stars (e.g., Drake, J.J., 1996, Cool Stars 9, ASP Conf.Ser. 109, 203). We however find that much of these analyses may be systematically underestimating the metallicities, and using a newly developed method to correctly treat the low-counts regime at the high-energy tail of the stellar spectra (van Dyk et al. 2001, ApJ 548,224), have found that the metallicities of these stars are generally comparable to their photospheric values. The results were reported at the AAS (Sourlas, Yu, van Dyk, Kashyap, and Drake, 2000, BAAS 196, v32, #54.02), and at the conference on Statistical Challenges in Modem Astronomy (Sourlas, van Dyk, Kashyap, Drake, and Pease, 2003, SCMA 111, Eds. E.D.Feigelson, G.J.Babu, New York:Springer, p489-490). We also described the limitations of one of the most egregiously misused and misapplied statistical tests in astrophysical literature, the F-test for verifying model components (Protassov, van Dyk, Connors, Kashyap, and Siemiginowska, 2002, ApJ, 571,545). Indeed, a search through the ApJ archives turned up 170 papers in the 5 previous years that used the F-test explicitly in some form or the other, and with the vast majority of them not using it correctly! Indeed, looking at just 4 issues of the ApJ in 2001, we found 13 instances of its use, of which nine were demonstrably incorrect. Clearly, it is difficult to understate the importance of this issue. We also worked on speeding up Bayes Blocks and Sparse Bayes Blocks algorithms to make them more tractable for large searches. We also supported staistics students and postdocs in both explicit physics- model-based (spectra with tens of thousands of atomic lines) and "model-free" -- i.e. non-parametric or semi-parametric -- algorithms. Work on using more of the latter is just beginning; while using multi-scale methods for Poisson imaging has come to hition. In fact, "An Image Restoration Technique with Error Estimates", by D. Esch, A. Connors, M. Karovska, and D. van Dyk, was published by ApJ (Esch et a1.2004, ApJ, 610, 1213). The code has been delivered to M. Karovska for CXC; and is available for beta-testing upon request. The other large project we worked on was on the self-consistent modeling of logN-logs curves in the Poisson limit. logN-logs curves are a fundamental tool in the study of source populations, luminosity functions, and cosmological parameters. However, their determination is hampered by statistical effects such as the Eddington bias, incompleteness due to detection efficiency, faint source flux fluctuations, etc. We have develed a new and powerful method using the full Poisson machinery that allows us to model the logN-logs distribution of X-ray sources in a self-consistent manner. Because we properly account for all the above statistical effects, our modeling is valid over the full range of the data, and not just for strong sources, as is normally done. Using a Bayesian approach and modeling the fluxes with known functional forms such as simple or broken power-laws, and conditioning the expected photon counts on the fluxes, the background contamination, effective area, detector vignetting, and detection probability, we can delve deeply into the low counts regime and extend the usefulness of medium sensitivity surveys such as ChAMP by orders of magnitude. The built-in flexibility of the algorithm also allows a simultaneous analysis of multiple datasets. We have applied this analysis to a set a Chandra observations (Sourlas, Kashyap, Zezas, van Dyk, 2004, HEAD #8, #16.32)

Scargle, Jeffrey D.↗
NASA NTRS2002

Automated Escape Guidance Algorithms for An Escape Vehicle

An escape vehicle was designed to provide an emergency evacuation for crew members living on a space station. For maximum escape capability, the escape vehicle needs to have the ability to safely evacuate a station in a contingency scenario such as an uncontrolled (e.g., tumbling) station. This emergency escape sequence will typically be divided into three events: The fust separation event (SEP1), the navigation reconstruction event, and the second separation event (SEP2). SEP1 is responsible for taking the spacecraft from its docking port to a distance greater than the maximum radius of the rotating station. The navigation reconstruction event takes place prior to the SEP2 event and establishes the orbital state to within the tolerance limits necessary for SEP2. The SEP2 event calculates and performs an avoidance burn to prevent station recontact during the next several orbits. This paper presents the tools and results for the whole separation sequence with an emphasis on the two separation events. The fust challenge includes collision avoidance during the escape sequence while the station is in an uncontrolled rotational state, with rotation rates of up to 2 degrees per second. The task of avoiding a collision may require the use of the Vehicle's de-orbit propulsion system for maximum thrust and minimum dwell time within the vicinity of the station vicinity. The thrust of the propulsion system is in a single direction, and can be controlled only by the attitude of the spacecraft. Escape algorithms based on a look-up table or analytical guidance can be implemented since the rotation rate and the angular momentum vector can be sensed onboard and a-priori knowledge of the position and relative orientation are available. In addition, crew intervention has been provided for in the event of unforeseen obstacles in the escape path. The purpose of the SEP2 burn is to avoid re-contact with the station over an extended period of time. Performing this maneuver properly requires knowledge of the orbital state, which is obtained during the navigation state reconstruction event. Since the direction of the delta-v of the SEPI maneuver is a random variable with respect to the Local Vertical Local Horizontal (LVLH) coordinate system, calculating the required SEP2 burn is a challenge. This problem was solved using a neural network as a model-free function approximation technique.

Flanary, Ronald↗
NASA NTRS2012

Particle Filtering for Model-Based Anomaly Detection in Sensor Networks

A novel technique has been developed for anomaly detection of rocket engine test stand (RETS) data. The objective was to develop a system that postprocesses a csv file containing the sensor readings and activities (time-series) from a rocket engine test, and detects any anomalies that might have occurred during the test. The output consists of the names of the sensors that show anomalous behavior, and the start and end time of each anomaly. In order to reduce the involvement of domain experts significantly, several data-driven approaches have been proposed where models are automatically acquired from the data, thus bypassing the cost and effort of building system models. Many supervised learning methods can efficiently learn operational and fault models, given large amounts of both nominal and fault data. However, for domains such as RETS data, the amount of anomalous data that is actually available is relatively small, making most supervised learning methods rather ineffective, and in general met with limited success in anomaly detection. The fundamental problem with existing approaches is that they assume that the data are iid, i.e., independent and identically distributed, which is violated in typical RETS data. None of these techniques naturally exploit the temporal information inherent in time series data from the sensor networks. There are correlations among the sensor readings, not only at the same time, but also across time. However, these approaches have not explicitly identified and exploited such correlations. Given these limitations of model-free methods, there has been renewed interest in model-based methods, specifically graphical methods that explicitly reason temporally. The Gaussian Mixture Model (GMM) in a Linear Dynamic System approach assumes that the multi-dimensional test data is a mixture of multi-variate Gaussians, and fits a given number of Gaussian clusters with the help of the wellknown Expectation Maximization (EM) algorithm. The parameters thus learned are used for calculating the joint distribution of the observations. However, this GMM assumption is essentially an approximation and signals the potential viability of non-parametric density estimators. This is the key idea underlying the new approach.

Solano, Wanda↗
NASA NTRS2014

GCR Environmental Models I: Sensitivity Analysis for GCR Environments

Accurate galactic cosmic ray (GCR) models are required to assess crew exposure during long-duration missions to the Moon or Mars. Many of these models have been developed and compared to available measurements, with uncertainty estimates usually stated to be less than 15%. However, when the models are evaluated over a common epoch and propagated through to effective dose, relative differences exceeding 50% are observed. This indicates that the metrics used to communicate GCR model uncertainty can be better tied to exposure quantities of interest for shielding applications. This is the first of three papers focused on addressing this need. In this work, the focus is on quantifying the extent to which each GCR ion and energy group, prior to entering any shielding material or body tissue, contributes to effective dose behind shielding. Results can be used to more accurately calibrate model-free parameters and provide a mechanism for refocusing validation efforts on measurements taken over important energy regions. Results can also be used as references to guide future nuclear cross-section measurements and radiobiology experiments. It is found that GCR with Z>2 and boundary energies below 500 MeV/n induce less than 5% of the total effective dose behind shielding. This finding is important given that most of the GCR models are developed and validated against Advanced Composition Explorer/Cosmic Ray Isotope Spectrometer (ACE/CRIS) measurements taken below 500 MeV/n. It is therefore possible for two models to very accurately reproduce the ACE/CRIS data while inducing very different effective dose values behind shielding.

Slaba, Tony C.↗
NASA NTRSDate not supplied

Spectral Mass-Gauging of Propellant Tanks

An overview of our recent results on the development of Spectral Mass-Gauging (SMG) technology for model-free gauging of propellants in microgravity applications will be presented. The technology is based on application a rigorous result from spectral theory – the Weyl’s Law – which relates the counting function of natural modes in a resonator with its volume. Development of the SMG includes theory of acoustic response of propellant tank, hardware and procedure characterization and optimization, development of data pre-processing approaches and software for automatic mode identification and counting. Main accomplishments in each field of the technology development will be presented. SMG has been tested recently in 1-g on a flight tank filled with water or LN2. We will present results of the tests and discuss their implications for the technology development. The presentation will conclude with a summary of the next steps in the technology maturation.

Mass-gauging↗
NASA NTRSDate not supplied

Optimization of Airport Runway Configuration with Forecast-Augmented Offline Reinforcement Learning

Runway configuration Management (RCM) governs the optimal utilization of runways based on variables such as traffic and meteorological conditions, making it a daunting task in air traffic management due to its dependency on volatile operational and environmental factors. This paper improves upon our previous work [1] on using offline model-free reinforcement learning for creating a Runway Configuration Assistance (RCA) decision-support tool. A novel integration of forecast data from LAMP (Localized Aviation Model Output Statistics Program) and TAF (Terminal Area Forecast) is introduced, enhancing the tool’s accuracy and also its adaptability to quick wind changes. The performance is evaluated using two major US airports, Charlotte Douglas International Airport (CLT) and Denver International Airport (DEN). To counter scalability issues presented by the addition of discrete forecast variables, we transitioned to a continuous state space model, ensuring scalability and inclusion of longer forecast data. The results of our experiments reflect significant improvements in the RCA tool’s prediction accuracy.

Sumanth Nethi↗
DOE OSTI2022

Lattice QCD and the Computational Frontier

The search for new physics requires a joint experimental and theoretical effort. Lattice QCD is already an essential tool for obtaining precise model-free theoretical predictions of the hadronic processes underlying many key experimental searches, such as those involving heavy flavor physics, the anomalous magnetic moment of the muon, nucleon-neutrino scattering, and rare, second-order electroweak processes. As experimental measurements become more precise over the next decade, lattice QCD will play an increasing role in providing the needed matching theoretical precision. Achieving the needed precision requires simulations with lattices with substantially increased resolution. As we push to finer lattice spacing we encounter an array of new challenges. They include algorithmic and software-engineering challenges, challenges in computer technology and design, and challenges in maintaining the necessary human resources. In this white paper we describe those challenges and discuss ways they are being dealt with. Overcoming them is key to supporting the community effort required to deliver the needed theoretical support for experiments in the coming decade.

Boyle, Peter↗
DOE OSTI2023

Primal-Dual Differentiable Programming for Distribution System Critical Load Restoration: Preprint

Swift and reliable critical load restoration (CLR) can help make a distribution system resilient towards extreme events. To optimally achieve that, alongside practical concerns such as limiting online computational burden, some studies leverage model-free reinforcement learning (RL) to train control policies. Despite the advantages provided by RL algorithms, these approaches suffer from two issues: 1) the lack of a proper mechanism for constraint enforcement, and 2) poor sample efficiency. Therefore, in this paper, a primal-dual differentiable programming (PDDP) method is developed for guiding the training leading to a constraint-satisfying policy. Additionally, the model-based nature of the proposed method aims at improving sample efficiency. The experiment on a CLR problem demonstrates that PDDP can effectively train a control policy that both achieves desirable performance and satisfies required constraints.

differentiable programming↗
DOE OSTI2023

A Novel Multi-Agent Deep Reinforcement Learning-enabled Distributed Power Allocation Scheme for mmWave Cellular Networks

We consider the power allocation problem over shared spectrum for millimeter-Wave (mmWave) cellular downlink. Existing approaches usually find sub-optimal solutions by solving a non-convex optimization which leads to scalability issues due to centralized control. Therefore, distributed and adaptive approaches are desirable. Recently, model-free Deep Reinforcement Learning (DRL) has achieved success in such wireless resource management tasks. By modeling the radio environment as a Markov Decision Process (MDP) with the base stations (BSs) being the agents, power allocation can be automated at the agent level with comparable throughput performance to conventional centralized schemes. The multi-agent setting presents new challenges as the radio environment is impacted by the joint actions of the agents and is no longer stationary from any individual agent’s perspective. Existing literature bypasses this non-stationarity violation by ignoring it which may cause performance degradation. To tackle this issue, we propose a distributed continuous power allocation scheme based on a modified version of multi-agent Deep Deterministic Policy Gradient (MADDPG) that is tailored for the distributed multiple-agent setting. The proposed scheme employs a centralized-training distributed-execution framework where Q-functions are trained over subsets of BSs while each BS determines its transmit power based only on its own local observation. It admits constant per-BS communication and computation complexity and is thus scalable to large networks. Numerical evaluation shows that the proposed scheme adapts well to a wide range of interference conditions and can achieve comparable or better performance than several state-of-the-art non-learning approaches.

99 GENERAL AND MISCELLANEOUS↗
DOE OSTI2024

Reinforcement Learning to Enhance Optimal Operation of Resilient Community Energy Systems

This paper presents a novel model-free multi-agent Reinforcement Learning (RL) control method to enhance the resilience of community energy systems in island mode, which coordinates multiple objectives without the necessity of identifying system models that require expert knowledge. Specifically, a community-level coordinator agent is designed to allocate renewable energy resources among different buildings, and multiple building-level agents are developed to optimize load schedules based on limited energy resources and requirements of building loads and occupants’ comfort. In a two-day evaluation, our RL approach demonstrated a similar performance against MPC without requiring system models and formulation of optimization problems as required in MPC.

ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATION↗