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 307 records · Page 17
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

A Machine Learning-Assisted Framework to Control Thermally Anisotropic Building Envelopes in Residential Buildings

To curb the energy consumption of buildings and their related CO2 emissions, Oak Ridge National Laboratory (ORNL) has developed the thermally anisotropic building envelope (TABE) —a multi-layer design comprising insulation materials and metal foils connected to thermal loops. In this study, a machine learning-assisted framework was developed to control the TABE in residential buildings to reduce the computation load for future optimal rule-based control and application. First, a 2D finite element model was established in COMSOL to calculate the hourly heat flux through exterior walls installed with the TABE. Then, TABE wall heat fluxes were simulated for various given indoor and outdoor boundary conditions, as well as thermal loops fluid temperatures and flow rates. Since the finite element simulations are computationally expensive, an artificial neural network (ANN) was then trained to use as a proxy of the finite element (COMSOL) modeling. Finally, the trained ANN model was coupled with the EnergyPlus model to predict the energy consumption of a US Department of Energy prototype single-family house installed with the TABE. An optimal simple rule-based control was determined from predefined rules for a case study. The results demonstrate that the developed machine learning–assisted framework can reduce 99.9% of the computation time while efficiently managing residential building energy for installed TABE walls.

Shen, Zhenglai↗
NASA NTRS2010

Thermoelectric Converter for Loop Heat Pipe Temperature Control: Experience and Lessons Learned

This paper describes the theoretical background and implementation methodology of using a thermoelectric converter (TEC) for operating temperature control of a loop heat pipe (LHP). In particular, experimental results from ambient and thermal vacuum tests of an LHP are presented for illustrations. The most commonly used state-of-the-art method to control the LHP operating temperature is to cold bias its compensation chamber (CC) and use an electrical heater to maintain the CC at the desired set point temperature. Although effective, this approach has its shortcomings in that the electrical heater can only provide heating to the CC, and the required power can be large under certain conditions. An alternative method is to use a TEC, which is capable of providing both heating and cooling to the CC. In this method, one side of the TEC is attached to the CC, and the other side is connected to the evaporator via a thermal strap. Using a bipolar power supply and a control algorithm, a TEC can function as a heater or a cooler, depending on the direction of the current flow. Extensive ground tests of several LHPs have demonstrated that a TEC can provide very tight temperature control for the CC. It also offers several additional advantages: (1) The LHP can operate at temperatures below its natural operating temperature at low heat loads; (2) The required heater power for a TEC is much less than that for an electrical heater; and (3) It enhances the LHP start-up success. Although the concept of using a TEC for LHP temperature control is simple, there are many factors to be considered in its implementation for space applications because the TEC is susceptible to the shear stress and yet has to sustain the dynamic load under the spacecraft launch environment. The added features that help the TEC to withstand the dynamic load will inevitably affect the TEC thermal performance. Some experiences and lessons learned are addressed in this paper.

Ku, Jentung↗
DOE OSTI2020

Deep Reinforcement Learning for Residential HVAC Control with Consideration of Human Occupancy

The Artificial Intelligence (AI) development described herein uses model-free Deep Reinforcement Learning (DRL) to minimize energy cost during residential heating, ventilation, and air conditioning (HVAC) operation. Building cooling loads and HVAC operation are difficult to accurately model due to complexity, lack of measurements and data, and model specific performance, so online machine learning is used to allow for real-time readjustment in performance. Energy costs for the multi-zone cooling unit shown in this work are minimized by scheduling on/off commands around dynamic prices. By taking advantage of precooling events that take place when the price is low, the agent is able to reduce operational cost without violating user comfort. The DRL controller was tested in simulation where the learner achieved a 43.89% cost reduction when compared to traditional, fixed-setpoint operation. The system is now ready for the next phase of testing in a live, real-time home environment.

Mckee, Evan↗
DOE OSTI2021

A Cyber-Physical System for Freeway Ramp Meter Signal Control Using Deep Reinforcement Learning in a Connected Environment

Freeway bottlenecks such as on-ramp merging areas account for about 40% of recurring freeway congestion. It is generally agreed that building more roads and adding more lanes to existing infrastructure does not solve the congestion problem, and so dynamic traffic control measures offer a more cost-effective alternative. Ramp meters, traffic signal devices that regulate traffic flow entering freeways, are among the most effective measures to mitigate congestion at on-ramp merging areas on freeways. The confluence of deep reinforcement learning (RL) and connectivity provides a possible solution to advance ramp meter signal control. Deep RL is a group of machine-learning methods that enables an agent learning from the environment to improve its performance. In this study, three deep RL methods-proximal policy optimization (PPO), Ape-X deep Q-network (DQN), and asynchronous advantage actor-critic agents (A3C)-are explored for ramp meter signal control to maximize vehicle speed and traffic throughput, as well as to minimize energy consumption and emissions at freeway on-ramp merging areas in a connected environment. The low computational requirement and scalability of deep RL for deployment make it a powerful optimization tool for time-sensitive applications such as ramp meter signal control. The results of this study show that deep RL methods yield superior performance to both a fixed-time controller and ALINE A, a state-of-the-art feedback controller.

ADVANCED PROPULSION SYSTEMS,MATHEMATICS AND COMPUT↗
NASA NTRS1991

On the integration of reinforcement learning and approximate reasoning for control

The author discusses the importance of strengthening the knowledge representation characteristic of reinforcement learning techniques using methods such as approximate reasoning. The ARIC (approximate reasoning-based intelligent control) architecture is an example of such a hybrid approach in which the fuzzy control rules are modified (fine-tuned) using reinforcement learning. ARIC also demonstrates that it is possible to start with an approximately correct control knowledge base and learn to refine this knowledge through further experience. On the other hand, techniques such as the TD (temporal difference) algorithm and Q-learning establish stronger theoretical foundations for their use in adaptive control and also in stability analysis of hybrid reinforcement learning and approximate reasoning-based controllers.

Berenji, Hamid R.↗
NASA NTRS1994

The Network Operations Control Center upgrade task: Lessons learned

This article synthesizes and describes the lessons learned from the Network Operations Control Center (NOCC) upgrade project, from the requirements phase through development and test and transfer. At the outset, the NOCC upgrade was being performed simultaneously with two other interfacing and dependent upgrades at the Signal Processing Center (SPC) and Ground Communications Facility (GCF), thereby adding a significant measure of complexity to the management and overall coordination of the development and transfer-to-operations (DTO) effort. Like other success stories, this project carried with it the traditional elements of top management support and exceptional dedication of cognizant personnel. Additionally, there were several NOCC-specific reasons for success, such as end-to-end system engineering, adoption of open-system architecture, thorough requirements management, and use of appropriate off-the-shelf technologies. On the other hand, there were several difficulties, such as ill-defined external interfaces, transition issues caused by new communications protocols, ambivalent use of two sets of policies and standards, and mistailoring of the new JPL management standard (due to the lack of practical guidelines). This article highlights the key lessons learned, as a means of constructive suggestions for the benefit of future projects.

Sherif, J. S.↗
DOE OSTI2022

Deep reinforcement learning with online data augmentation to improve sample efficiency for intelligent HVAC control

Deep Reinforcement Learning (DRL) has started showing success in real-world applications such as building energy optimization. Much of the research in this space utilized simulated environments to train RL-agent in an offline mode. Very few research have used DRL-based control in real-world systems due to two main reasons: 1) sample efficiency challenge---DRL approaches need to perform a lot of interactions with the environment to collect sufficient experiences to learn from, which is difficult in real systems, and 2) comfort or safety related constraints---user's comfort must never or at least rarely be violated. In this work, we propose a novel deep Reinforcement Learning framework with online Data Augmentation (RLDA) to address the sample efficiency challenge of real-world RL. We used a time series Generative Adversarial Network (TimeGAN) architecture as a data generator. We further evaluated the proposed RLDA framework using a case study of an intelligent HVAC control. With a ≈28% improvement in the sample efficiency, RLDA framework lays the way towards increased adoption of DRL-based intelligent control in real-world building energy management systems.

Kurte, Kuldeep↗
DOE OSTI2024

Machine learning surrogate for charged particle beam dynamics with space charge based on a recurrent neural network with aleatoric uncertainty

In this work, we develop a machine learning (ML) model with aleatoric uncertainty for the low energy beam transport (LEBT) region of the LANSCE linear accelerator in which we model the transport of a space-charge-dominated 750 keV proton beam through a lattice of 22 quadrupole magnets. Our ML model is developed based on data generated by a Kapchinsky–Vladimirsky (KV) envelope model of beam transport. We show that a recurrent neural network can be used as a dynamical surrogate model for fast prediction of the LEBT beam envelope. Furthermore, we endow the model with the prediction of aleatoric uncertainty and compare three different approaches. We demonstrate that the ML-based uncertainty quantification models are well calibrated and produce good estimates of the regions where the model is less certain about its predictions. This ML framework is a necessary step in the development of a real-time virtual diagnostic tool with uncertainty quantification that can be integrated into more complex downstream tasks (e.g., adaptive control or learning flexible control policies via reinforcement learning) for improved efficiency in beam operations. In future work, we plan to expand on this preliminary study by considering more realistic envelope models that include longitudinal momentum spread and dispersive effects in bending magnets, as well as particle tracking codes with 3D space charge (such as and ). Published by the American Physical Society 2024

43 PARTICLE ACCELERATORS↗
DOE OSTI2023

Machine Learning-Based Model Predictive Control of Two-Time-Scale Systems

In this study, we present a general form of nonlinear two-time-scale systems, where singular perturbation analysis is used to separate the dynamics of the slow and fast subsystems. Machine learning techniques are utilized to approximate the dynamics of both subsystems. Specifically, a recurrent neural network (RNN) and a feedforward neural network (FNN) are used to predict the slow and fast state vectors, respectively. Moreover, we investigate the generalization error bounds for these machine learning models approximating the dynamics of two-time-scale systems. Next, under the assumption that the fast states are asymptotically stable, our focus shifts toward designing a Lyapunov-based model predictive control (LMPC) scheme that exclusively employs the RNN to predict the dynamics of the slow states. Additionally, we derive sufficient conditions to guarantee the closed-loop stability of the system under the sample-and-hold implementation of the controller. A nonlinear chemical process example is used to demonstrate the theory. In particular, two RNN models are constructed: one to model the full two-time-scale system and the other to predict solely the slow state vector. Both models are integrated within the LMPC scheme, and we compare their closed-loop performance while assessing the computational time required to execute the LMPC optimization problem.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Report for 2.2.1 Task 5: Develop and Document a State-Based Alarm System for a Nuclear Power Plant Control Room Using Machine Learning

Idaho National Laboratory (INL) has contracted IFE to support human factors research and the development of leading-edge technology to support control room operators as the U.S. fleet undergoes modernization and digitalization in legacy plants and control rooms. As the nuclear industry starts to shift to more digital controls and systems, these upgrades provide more information to the main control room in the form of digital signals and values. Also, an increase in the number of alarms points is provided by the vendors. New digital control rooms shift from main control boards to soft controls, and alarm panels to a single display. This change to a single display has created a problem of alarm waterfall, a situation in which the addition of new alarm points causes an overload of information to the operator during either a plant disturbance or other than normal full power operations. IFE’s goal is to find a workable solution to assist operators in both handling and understanding incoming alarms during emergency situations under these waterfall conditions.

99 GENERAL AND MISCELLANEOUS↗
NASA NTRS1989

RSRA/X-Wing flight control system development - Lessons learned

The X-Wing, in concept, marries the efficiencies of a helicopter and fixed wing aircraft through the use of a four-bladed wing/rotor that can be rotated or stopped in flight. The RSRA/X-Wing flight test program was a technology demonstration of this concept which, after three successful flights, was discontinued in late 1987. In spite of many technical challenges in this program, such as the use of circulation control, the fabrication of a large all-composite rotor, the development of an advanced, quadruplex digital flight control system, and the need for higher harmonic control, no major technical problems had been encountered at the time of the stop-work order. This paper addresses the issues of flight control system development and focuses on lessons learned. As with other such programs, software development was the most consuming issue. Other subjects of discussion include the problems of balancing program goals with technical goals, software- and hard-ware-related problems, safety issues, and system testing.

Corliss, Lloyd D.↗
NASA NTRSDate not supplied

Simultaneous Trajectory Tracking Control and Online Mass Estimation for a Regolith Excavating Robot via Integral Concurrent Learning

An adaptive controller is developed for a regolith excavation robot to determine the mass of excavated material and to account for the effects of gravity and friction while on the surface of other celestial bodies. A data-based integral concurrent learning (ICL) parameter update law accounts for and estimates the unknown mass, gravity, and friction parameters. A Lyapunov-based analysis proves that the trajectory tracking error and the parameter estimate errors exponentially converge to zero. An estimation of the mass of regolith excavated by the robot is calculated from the estimated parameters. A simulation study is performed to show the performance of the developed technique. Simulation results show that for 3 kilograms of excavated material, the mass estimate has an error of 4.9 grams.

Max L. Greene↗
DOE OSTI2023

Dynamic Model Agnostic Reliability Evaluation of Machine-Learning Models Integrated in Instrumentation & Control Systems

In recent years, the field of machine learning (ML), specifically neural networks, has grown significantly and has spurred research in its applicability to digital instrumentation and control systems (DI&C). While ML models have shown promise in operational contexts, the trustworthiness of using such algorithms has not been adequately assessed. Failures of ML integrated systems are not well understood, and the lack of comprehensive risk modeling can degrade the trustworthiness in these systems. In recent reports by the National Institute for Standards and Technology (NIST) [1] and the Nuclear Regulatory Commission (NRC) [2], they indicate that trustworthiness in ML is a critical barrier and will play a vital role in the safe, accountable, and secure operation of intelligent systems. Thus, in this work, we demonstrate a dynamic model-agnostic method to quantify the relative reliability of AI/ML predictions by incorporating out-of-distribution (OOD) detection on the training dataset. It is well documented that most ML algorithms excel at interpolation (or near-interpolation) tasks but experience significant performance degradation at extrapolation. The method, referenced as the Laplacian distributed decay for reliability (LADDR), determines the difference between the operational and training datasets which can used to the relative reliability of AI/ML predictions. LADDR is then demonstrated on a feedforward neural network based digital twin used for the prediction of safety significant factors during a loss-of-flow transient. LADDR is used to demonstrate how training data can be used as evidence to support the relative reliability of ML/AI predictions enhancing the overall trustworthiness of the system.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2021

Adaptive Machine Learning for Robust Diagnostics and Control of Time-Varying Particle Accelerator Components and Beams

Machine learning (ML) is growing in popularity for various particle accelerator applications including anomaly detection such as faulty beam position monitor or RF fault identification, for non-invasive diagnostics, and for creating surrogate models. ML methods such as neural networks (NN) are useful because they can learn input-output relationships in large complex systems based on large data sets. Once they are trained, methods such as NNs give instant predictions of complex phenomenon, which makes their use as surrogate models especially appealing for speeding up large parameter space searches which otherwise require computationally expensive simulations. However, quickly time varying systems are challenging for ML-based approaches because the actual system dynamics quickly drifts away from the description provided by any fixed data set, degrading the predictive power of any ML method, and limits their applicability for real time feedback control of quickly time-varying accelerator components and beams. In contrast to ML methods, adaptive model-independent feedback algorithms are by design robust to un-modeled changes and disturbances in dynamic systems, but are usually local in nature and susceptible to local extrema. In this work, we propose that the combination of adaptive feedback and machine learning, adaptive machine learning (AML), is a way to combine the global feature learning power of ML methods such as deep neural networks with the robustness of model-independent control. We present an overview of several ML and adaptive control methods, their strengths and limitations, and an overview of AML approaches.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Learning Provably Stable Local Volt/Var Controllers for Efficient Network Operation

Here this paper develops a data-driven framework to synthesize local Volt/Var control strategies for distributed energy resources (DERs) in power distribution grids (DGs). Aiming to improve DG operational efficiency, as quantified by a generic optimal reactive power flow (ORPF) problem, we propose a two-stage approach. The first stage involves learning the manifold of optimal operating points determined by an ORPF instance. To synthesize local Volt/Var controllers, the learning task is partitioned into learning local surrogates (one per DER) of the optimal manifold with voltage input and reactive power output. Since these surrogates characterize efficient DG operating points, in the second stage, we develop local control schemes that steer the DG to these operating points. We identify the conditions on the surrogates and control parameters to ensure that the locally acting controllers collectively converge, in a global asymptotic sense, to a DG operating point agreeing with the local surrogates. We use neural networks to model the surrogates and enforce the identified conditions in the training phase. AC power flow simulations on the IEEE 37-bus network empirically bolster the theoretical stability guarantees obtained under linearized power flow assumptions. The tests further highlight the optimality improvement compared to prevalent benchmark methods.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2023

Multi-Task Reinforcement Learning for Distribution System Voltage Control With Topology Changes

This letter proposes a multi-task deep reinforcement learning (DRL) approach for distribution system voltage regulation considering topology changes via PV smart inverter control. The key idea is to encode the topology as an additional state for the DRL and leverage the multi-task learning scheme for joint learning of all task control policies. Unlike other DRL-based methods, our approach is robust to different topologies. Comparison results on the modified IEEE 123-node system demonstrate the enhanced robustness of the proposed method.

24 POWER TRANSMISSION AND DISTRIBUTION↗