Search NASA⌕ Search

SEARCH · Search NASA

Results for “Learning control”

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 775 records · Page 43
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Assimilating partial observation to enhance feedback control of stochastic dynamical systems

Here, in this paper, we present a novel methodology to tackle feedback optimal control problems in scenarios where the exact state of the controlled process is unknown. It integrates data assimilation techniques and optimal control solvers to manage partial observation of the state process, a common occurrence in practical scenarios. Traditional stochastic optimal control methods assume full state observation, which is often not feasible in real-world fluid dynamics control problems. Our approach underscores the significance of utilizing observational data to inform control policy design. Specifically, we introduce a kernel learning backward stochastic differential equation (SDE) filter to enhance data assimilation efficiency and propose a sample-wise stochastic optimization method within the stochastic maximum principle framework. We demonstrate the efficacy and accuracy of our method in the control of advection-diffusion-reaction flow problem and the Dubins airplane maneuvering problem with model uncertainty.

data driven↗
DOE OSTI2022

Machine Learning for Automated Metadata Assignment in Buildings: Cooperative Research and Development (Final Report, CRADA Number CRD-18-00767)

RealTerm Energy and NREL have identified a shared vision to evaluate opportunities to facilitate the organization and assignment of metadata to building control system (BCS) data via industry-informed machine learning (ML). Manual metadata assignment is labor intensive and costly, slowing down any Energy Management and Information System (EMIS) deployment in the building space. This project aims to develop methodologies to accurately assign this metadata and significantly decrease the level of effort associated with deploying EMIS. The objective of this project is to identify/design methodologies to assign metadata to HVAC control points automatically. The identified methodologies will be programmed in analytics algorithms so they can ingest a list of points and produce a detailed tagging following the Haystack classification nomenclature. To validate the efficacy of each methodology, tagging results will be compared utilizing a list of points extracted from RealTerm's building database-as well as data extracted from the NREL campus via the Intelligent Campus program-enabling testing against large datasets with real world challenges. The developed methodologies may leverage building manager/operator input on a limited basis to add context to the classifying algorithms. The partnership aims to advance global efforts in areas related to the DOE missions through improving operational performance of commercial buildings. It is well documented that buildings fall out of commission after they are occupied, wasting significant energy and incurring associated costs simply due to poor operational performance. Emerging EMIS technologies that perform continuous commissioning help to address this issue, yet integration of these systems can be labor intensive both for the technology vendor and the building owner/operator. This project will enable more efficient and cost-effective analytics for buildings, enabling improvement in building operations at lower cost points.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗
DOE OSTI2023

Traffic Signal Optimization by Integrating Reinforcement Learning and Digital Twins

Machine learning (ML) methods, especially reinforcement learning (RL), have been widely considered for traffic signal optimization in intelligent transportation systems. Most of these ML methods are centralized, lacking in scalability and adaptability in large traffic networks. Further, it is challenging to train such ML models due to the lack of training platforms and/or the cost of deploying and training in a real traffic networks. This paper presents an approach for the integration of decentralized graph-based multi-agent reinforcement learning (DGMARL) with a Digital Twin (DT) to optimize traffic signals for the reduction of traffic congestion and network-wide fuel consumption related to stopping. Specifically, the DGMARL agents learn traffic state patterns and make decisions regarding traffic signal control with assistance from a Digital Twin module, which simulates and replicates the traffic behaviors of a real traffic network. The proposed approach was evaluated using PTV-Vissim [1], a microscopic traffic simulation platform. PTV-Vissim is also the simulation engine of the DT, enabling emulation and optimization of the traffic signals on the MLK Smart Corridor in Chattanooga, Tennessee. Compared to an actuated signal control baseline approach, experiment results show that Eco_PI, a developed performance measure capturing the impact of stops on fuel consumption, was reduced by 44.27% in a 24-hour and an average of 29.88% in a PM-peak-hour scenario.

Multi-Agent Reinforcement Learning, Digital Twin, ↗
DOE OSTI2021

High temperature oxidation of corrosion resistant alloys from machine learning

Parabolic rate constants, k p , were collected from published reports and calculated from corrosion product data (sample mass gain or corrosion product thickness) and tabulated for 75 alloys exposed to temperatures between ~800 and 2000 K (~500–1700 °C; 900–3000°F). Data were collected for environments including lab air, ambient and supercritical carbon dioxide, supercritical water, and steam. Materials studied include low- and high-Cr ferritic and austenitic steels, nickel superalloys, and aluminide materials. A combination of Arrhenius analysis, simple linear regression, supervised and unsupervised machine learning methods were used to investigate the relations between composition and oxidation kinetics. The supervised machine learning techniques produced the lowest mean standard errors. The most significant elements controlling oxidation kinetics were Ni, Cr, Al, and Fe, with Mo and Co composition also found to be significant features. The activation energies produced from the machine learning analysis were in the correct distributions for the diffusion constants for the oxide scales expected to dominate in each class.

Materials Science↗
DOE OSTI2021

Nontargeted vs. Targeted vs. Smart Load Shifting Using Heat Pump Water Heaters

Deployment of CTA-2045–enabled devices is increasing in the U.S. market. These devices allow utilities or third-party aggregators to control appliance energy use in homes, and could also be applied to end uses in small commercial buildings. This study focuses on a field study using CTA-2045–enabled water heaters to shift electric load off the peak and toward periods when renewable resources are more prevalent (e.g., near noon for solar resources and near midnight for wind resources). The following load shifting strategies were compared to understand effects on the aggregate load-shifting capabilities of Heat Pump Water Heaters (HPWHs) and on consumer hot water supply: non-targeted (traditional), targeted (grouped, with different shifting schedules) and “smart” (adaptive control commands). The results of this study show that targeted and smart control strategies yield significantly more load-shifting potential from a population of water heaters than the non-targeted approach without sacrificing hot water supply to occupants. However, as control commands become more aggressive, aggregators may face challenges in meeting consumer hot water demand. Furthermore, the findings and lessons learned can benefit electric utilities and inform updates to manufacturer controls and communications standards. The data collected may also be useful for developing and validating HPWH models.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2023

The Uniform Methods Project: Smart Thermostat Evaluation Protocol

A smart thermostat is an internet-connected device that controls home heating, ventilation, and air-conditioning (HVAC) equipment and can automatically adjust temperature set points to optimize performance and achieve energy savings. Smart thermostat features often include two way communication, occupancy detection (such as geofencing and occupancy sensors), schedule learning, and seasonal optimization algorithms. Smart thermostats can control most conventional HVAC systems, including central air conditioners, heat pumps, and forced air furnaces. Several types of residential utility programs offer smart thermostats as replacements measures. Working with smart thermostat vendors, utilities can offer separate optimization programs to produce energy savings beyond those achieved by installing a smart thermostat. From an evaluation perspective, smart thermostat programs have several noteworthy features. First, the energy savings from a smart thermostat may change over the life of the device. As a smart thermostat is connected to the internet, original equipment manufacturers can update the thermostat software to improve the thermostat's energy efficiency. Likewise, users can adjust the thermostat settings and schedules over time in response to changes in weather, thermal comfort, energy prices, or preferences for energy efficiency. Additionally, many thermostat manufacturers offer seasonal optimization programs that recommend changes or make minor, automated adjustments to the thermostat settings to improve energy efficiency. These opt-in programs are now standard offerings for many smart thermostat manufacturers and provided at no additional cost to users. The potential for software updates and continuous optimization and the evolving nature of user interactions mean future energy savings may differ from first-year savings and the energy savings of smart thermostats may need to be evaluated more than once. Second, smart thermostats often have small unit energy savings relative to a home's total energy consumption, especially in comparison to whole- home retrofit programs. This can make it difficult to detect the smart thermostat savings in billing or advanced metering infrastructure (AMI) meter consumption data. For example, as cooling loads in many regions average about 20% of annual electricity consumption, smart thermostat savings of 10% of cooling energy use would equate to a 2% reduction in home electricity consumption. Evaluators should use regression analysis of whole-home billing consumption or advanced metering infrastructure (AMI) meter consumption data to evaluate smart thermostat savings because, as explained at greater length below , these data are usually available to evaluators and regression can control for the impacts of weather and other potentially confounding factors on a home's energy consumption. Finally, as with other energy efficiency programs, participation in smart thermostat programs is self-selective. As discussed at greater length below , smart thermostat participants tend to be, among other things, younger, higher-income, and more likely to adopt electric vehicles (EVs) and internet connected devices than nonparticipants. These differences are often unobservable to the evaluator and correlated with a home's energy consumption, creating the potential for bias in estimating savings. Due to the small unit savings of thermostats, errors and biases from self-selection that may not be very consequential when evaluating a whole- home retrofits (e.g., ±2% of home electricity consumption) can have a major impact when evaluating the savings and cost-effectiveness of smart thermostat programs. A percentage point change in the estimated savings could affect the cost-effectiveness of a program. This means it is important for evaluators to assess and to minimize the potential for error from selection bias in estimating smart thermostat program savings. The Uniform Methods Project provides model protocols for determining energy savings and demand reductions that result from specific energy efficiency measures implemented through state and utility programs. In most cases, the measure protocols are based on a particular option identified by the International Performance Verification and Measurement Protocol ; however, this work provides a more detailed approach to implementing that option. Each chapter is written by technical experts in collaboration with their peers, reviewed by industry experts, and subject to public review and comment. The UMP protocols can be used by utilities, program administrators, public utility commissions, evaluators, and other stakeholders for both program planning and evaluation.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2020

A machine learning model for predicting the minimum miscibility pressure of CO 2 and crude oil system based on a support vector machine algorithm approach

CO 2 enhanced oil recovery (EOR) is a potential way for carbon capture, utilization and storage (CCUS). Though, the effect of CO 2 injection is greatly influenced by the reservoir conditions. Typically, Minimum miscible pressure (MMP) is selected as one of the key parameters for the screening and evaluation of prospective CO 2 flooding. Conventional slim tube test is both accurate and widely accepted but it is inefficient. Existing empirical formulas for MMPs are easy to be used but have been proved inaccurate and unreliable. Machine learning-based methods have great advantages in predicting MMP. However, only predication accuracy is discussed for most models without the screening of the main control factors and further validation of the model reliability. In this paper, a new prediction model based on support vector machine (SVM) was developed for pure/impure CO 2 and crude oil system. This study was based on 147 sets of MMP data from the literature with full information on reservoir temperature, oil composition and gas composition. The main control factors were screened by several statistical methods. Unlike the conventional prediction models that verified by only prediction accuracy, learning curve and single factor control variable analysis are further validated to obtain the optimum model.

02 PETROLEUM↗
DOE OSTI2023

AttentionFire_v1.0: interpretable machine learning fire model for burned-area predictions over tropics

Abstract. African and South American (ASA) wildfires account for more than 70 % of global burned areas and have strong connection to local climate for sub-seasonal to seasonal wildfire dynamics. However, representation of the wildfire–climate relationship remains challenging due to spatiotemporally heterogenous responses of wildfires to climate variability and human influences. Here, we developed an interpretable machine learning (ML) fire model (AttentionFire_v1.0) to resolve the complex controls of climate and human activities on burned areas and to better predict burned areas over ASA regions. Our ML fire model substantially improved predictability of burned areas for both spatial and temporal dynamics compared with five commonly used machine learning models. More importantly, the model revealed strong time-lagged control from climate wetness on the burned areas. The model also predicted that, under a high-emission future climate scenario, the recently observed declines in burned area will reverse in South America in the near future due to climate changes. Our study provides a reliable and interpretable fire model and highlights the importance of lagged wildfire–climate relationships in historical and future predictions.

58 GEOSCIENCES↗
DOE OSTI2023

2023 Roadmap on molecular modelling of electrochemical energy materials

New materials for electrochemical energy storage and conversion are the key to the electrification and sustainable development of our modern societies. Molecular modelling based on the principles of quantum mechanics and statistical mechanics as well as empowered by machine learning techniques can help us to understand, control and design electrochemical energy materials at atomistic precision. Therefore, this roadmap, which is a collection of authoritative opinions, serves as a gateway for both the experts and the beginners to have a quick overview of the current status and corresponding challenges in molecular modelling of electrochemical energy materials for batteries, supercapacitors, CO 2 reduction reaction, and fuel cell applications.

25 ENERGY STORAGE↗
DOE OSTI2024

Integrate Latimer Controls' Solution into RTAC (CRADA Final Report, CRD-23-24672)

Latimer Controls, Inc. was awarded two vouchers under the Department of Energy's American-Made Solar Prize Round 6 to conduct collaborative research at a national laboratory. The National Renewable Energy Laboratory (NREL) was selected as a partner to assist Latimer Controls in the performance evaluation of its photovoltaic (PV) control software. This collaboration focuses on developing a hardware-in-the-loop (HIL) testbed at NREL, which will be used to test and validate the Latimer PV control technology in a realistic yet de-risked environment. Both Latimer and NREL teams will work together to analyze the collected test data, derive insights, and disseminate the scientific findings. Recent studies underscore the potential of solar energy as a zero-marginal-cost and zero-emission flexibility resource within the bulk power system, particularly when integrated with advanced control systems. To enhance the performance of such systems, Latimer Controls has developed leading-edge technologies, including machine learning (ML) algorithms and hierarchical inverter set-point allocation methods. These innovations are designed to estimate the operational headroom of large PV plants for grid integration and control. However, comprehensive validation under real-world conditions remains necessary. To address this gap, the concurrent CRADA project proposes the real-world application and validation of the Latimer Control solution within a HIL environment. Initially, the Latimer algorithm was developed and tested within MATLAB Simulink, a platform suitable for research-level simulations and iterative development. However, transitioning this technology to a real solar site as an industry-ready solution necessitates implementation in a format compatible with widely used solar power plant controllers. In this additional CRADA work, the MATLAB Simulink-based logic will be translated into Structured Text, a programming language compliant with IEC 61131 standards, which is commonly used for custom logic implementations in industry-leading programmable logic controllers (PLCs), such as the Schweitzer SEL real-time automation controller (RTAC). This transition will facilitate the deployment of the Latimer Control solution in real-world solar power plants, thereby advancing the technology towards commercialization.

14 SOLAR ENERGY↗
DOE OSTI2022

Learning infinite-horizon average-reward restless multi-action bandits via index awareness

We consider the online restless bandits with average-reward and multiple actions, where the state of each arm evolves according to a Markov decision process (MDP), and the reward of pulling an arm depends on both the current state of the corresponding MDP and the action taken. Since finding the optimal control is typically intractable for restless bandits, existing learning algorithms are often computationally expensive or with a regret bound that is exponential in the number of arms and states. In this paper, we advocate \textit{index-aware reinforcement learning} (RL) solutions to design RL algorithms operating on a much smaller dimensional subspace by exploiting the inherent structure in restless bandits. Specifically, we first propose novel index policies to address dimensionality concerns, which are provably optimal. We then leverage the indices to develop two low-complexity index-aware RL algorithms, namely, (i) GM-R2MAB, which has access to a generative model; and (ii) UC-R2MAB, which learns the model using an upper confidence style online exploitation method. We prove that both algorithms achieve a sub-linear regret that is only polynomial in the number of arms and states. A key differentiator between our algorithms and existing ones stems from the fact that our RL algorithms contain a novel exploitation that leverages our proposed provably optimal index policies for decision-makings.

Xiong, Guojun↗
DOE OSTI2024

Automated Calibration for Rapid Optical Spectroscopy Sensor Development for Online Monitoring

An automated platform has been developed to assist researchers in the rapid development of optical spectroscopy sensors to quantify species from spectral data. This platform performs calibration and validation measurements simultaneously. Real-time, in situ monitoring of complex systems through optical spectroscopy has been shown to be a useful tool; however, building calibration models requires development time, which can be a limiting factor in the case of radiological or otherwise hazardous systems. While calibration time can be reduced through optimized design of experiments, this study approached the challenge differently through automation. The ATLAS (Automated Transient Learning for Applied Sensors) platform used pneumatic control of stock solutions to cycle flow profiles through desired calibration concentrations for multivariate model construction. Additionally, the transients between desired concentrations based on flow calculations were used as validation measurements to understand model predictive capabilities. This automated approach yielded an incredible 76% reduction in model development time and a 60% reduction in sample volume versus estimated manual sample preparation and static measurements. The ATLAS system was demonstrated on two systems: a three-lanthanide system with Pr/Nd/Ho representing a use case with significant overlap or interference between analyte signatures and an alternate system containing Pr/Nd/Ni to demonstrate a use case in which broad-band corrosion species signatures interfered with more distinct lanthanide absorbance profiles. Both systems resulted in strong model prediction performance (RMSEP < 9%). Lastly, ATLAS was demonstrated as a tool to simulate process monitoring scenarios (e.g., column separation) in which models can be further optimized to account for day-to-day changes as necessary (e.g., baseline correction). Ultimately, ATLAS offers a vital tool to rapidly screen monitoring methods, investigate sensor fusion, and explore more complex systems (i.e., larger numbers of species).

47 OTHER INSTRUMENTATION↗
DOE OSTI2023

Building thermal dynamics modeling with deep transfer learning using a large residential smart thermostat dataset

Understanding thermal dynamics and obtaining the computational model of residential buildings enable its scaled application in energy retrofits, control optimization and decarbonization. In this paper, we present a deep learning approach to model building thermal dynamics with smart thermostat data collected from residential buildings, with the goal to investigate model generalizability. In the first stage, we developed and compared different Deep Learning architectures including Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM) models and CNN-LSTM to predict indoor air temperature in a multi-step time horizon. In the second stage, we implemented a Transfer Learning (TL) process, which aims to improve the prediction performance on a new set of buildings (targets), exploiting the knowledge of related or similar buildings (sources). Different TL strategies and source model identification methods were investigated. The study showed that the CNN-LSTM performed the best among the architectures compared, with an average Mean Absolute Error (MAE) of 0.26 °C for one-hour-ahead (twelve 5-min future steps) predictions. Furthermore, the results showed that freezing the LSTM layer and fine-tuning the other layers of the CNN-LSTM achieved the best performance among four TL strategies, which further improved the performance with respect to a machine learning approach by 10%, and proving the effectiveness and generalizability of the proposed approach. A comparison of three different source model identification methods showed that randomly selecting source models constrained by similar building characteristics can provide good TL performance while retaining simplicity comparing with other quantitative source identification methods.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2025

A Neural Optimizer With Decision-Focused Learning for Optimal Energy Storage Operation

Here, this article introduces a neural optimizer-based framework for optimizing battery energy storage system (BESS) control for grid services, including demand charge and energy cost reduction. By leveraging decision-focused learning (DFL), the proposed framework ensures seamless integration and adaptation, significantly enhancing control performance. A patch time-series transformer is employed for peak load forecasting, incorporating aleatoric uncertainty quantification to account for forecasting uncertainties within the decision-making process. The framework utilizes a solver-in-the-loop approach to generate optimal BESS actions, which are then used to train the neural optimizer-based agent. By co-optimizing both BESS operational modes and output power within the NN, the system achieves improved performance and robustness. After initial training, the forecasting and control models are jointly fine-tuned to account for forecasting errors, further improving decision precision and efficiency through DFL. Case studies are performed to validate the performance of the framework using multiple real-world datasets, demonstrating superior performance in monthly peak load forecasting compared to state-of-the-art models. In addition, the results are compared against existing decision-making approaches. The results demonstrate a reduction in monthly peak forecasting error by approximately 15% across various performance measures and achieve an optimization gap for BESS operation that is about three times smaller compared to existing methods.

Kim, Hyeonjin [Pacific Northwest National Laborato↗
DOE OSTI2024

Fracture Networks Imaging in CO2 Injection Zones in IBDP Site: An Unsupervised Machine Learning Application with Multiple Datasets

Poster presented at the 17th International Conference on Greenhouse Gas Control Technologies GHGT-17 held in Calgary, Canada, October 20-24, 2024. This poster highlights the integration of unsupervised machine learning (ML) techniques as a transformative tool for advancing understanding of CO2 injection into reservoirs that could potentially contribute to optimizing injection strategies and reservoir management, ultimately bolstering the efficacy and sustainability of CO2 storage.

Kumar, Abhash↗
DOE OSTI2012

Utah Wind Power Educational Consortium for Workforce Development (Final Scientific/Technical Report)

The funds from this project were used to purchase tools and instrumentation to help replicate actual on-the-job wind energy scenarios which provided the students with the practical or applied components of wind energy jobs. This project enhanced the educational experiences provided for the students in terms of engineering and science components of wind energy by using electronics, control systems, and electro-mechanical instrumentation to help students learn standardized wind-specific craftsman skills. In addition the tools and instrumentation helped the students learn the safety necessary to work in the wind industry.

17 WIND ENERGY↗