Search NASA⌕ Search

SEARCH · Search NASA

Results for “Policy optimization”

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 73 records · Page 4
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

Optimal carbon storage reservoir management through deep reinforcement learning

Model-based optimization plays a central role in energy system design and management. The complexity and high-dimensionality of many process-level models, especially those used for geosystem energy exploration and utilization, often lead to formidable computational costs when the dimension of decision space is also large. This work adopts elements of recently advanced deep learning techniques to solve a sequential decision-making problem in applied geosystem management. Specifically, a deep reinforcement learning framework was formed for optimal multiperiod planning, in which a deep Q-learning network (DQN) agent was trained to maximize rewards by learning from high-dimensional inputs and from exploitation of its past experiences. To expedite computation, deep multitask learning was used to approximate high-dimensional, multistate transition functions. Both DQN and deep multitask learning are pattern based. As a demonstration, the framework was applied to optimal carbon sequestration reservoir planning using two different types of management strategies: monitoring only and brine extraction. Both strategies are designed to mitigate potential risks due to pressure buildup. Results show that the DQN agent can identify the optimal policies to maximize the reward for given risk and cost constraints. Finally, experiments also show that knowledge the agent gained from interacting with one environment is largely preserved when deploying the same agent in other similar environments.

15 GEOTHERMAL ENERGY↗
DOE OSTI2024

Development of algorithms for augmenting and replacing conventional process control using reinforcement learning

Here, this work seeks to allow for the online operation and training of model-free reinforcement learning (RL) agents but limit the risk to system equipment and personnel. The parallel implementation of RL alongside more conventional process control (CPC) allows for the RL algorithm to learn from CPC. The past performance of both methods are assessed on a continuous basis allowing for a transition from CPC to RL and, if needed, transitioning back to CPC from RL. This allows for the RL algorithm to slowly and safely assume control of the process without significant degradation in control performance. It is shown that the RL can derive a near optimal policy even when coupled with a suboptimal CPC. It is also demonstrated that the coupled RL-CPC algorithm learns at a faster rate than traditional RL methods of exploration while the algorithm’s performance does not deteriorate below CPC, even when exposed to an unknown operating condition.

30 DIRECT ENERGY CONVERSION↗
DOE OSTI2024

A safe reinforcement learning algorithm for supervisory control of power plants

Traditional control theory-based methods require tailored engineering for each system and constant fine-tuning. In power plant control, one often needs to obtain a precise representation of the system dynamics and carefully design the control scheme accordingly. Model-free Reinforcement learning (RL) has emerged as a promising solution for control tasks due to its ability to learn from trial-and-error interactions with the environment. It eliminates the need for explicitly modeling the environment’s dynamics, which is potentially inaccurate. However, the direct imposition of state constraints in power plant control raises challenges for standard RL methods. To address this, we propose a chance-constrained RL algorithm based on Proximal Policy Optimization for supervisory control. Our method employs Lagrangian relaxation to convert the constrained optimization problem into an unconstrained objective, where trainable Lagrange multipliers enforce the state constraints. In conclusion, our approach achieves the smallest distance of violation and violation rate in a load-follow maneuver for an advanced Nuclear Power Plant design.

constrained optimization↗
DOE OSTI2021

The surprisingly inexpensive cost of state-driven emission control strategies

Traditionally, analysis of the costs of cutting greenhouse gas emissions has assumed that governments would implement idealized, optimal policies such as uniform economy-wide carbon taxes. Yet actual policies in the real world, especially in large federal governments, are often highly heterogeneous and vary in political support and administrative capabilities within a country. While the benefits of heterogeneous action have been discussed widely for experimentation and leadership, little is known about its costs. Focusing on the United States, we represent plausible variation (by more than a factor of 3) in the stringency of state-led climate policy in a process-based integrated assessment model (GCAM-USA). For a wide array of national decarbonization targets, we find that the nationwide cost from heterogeneous subnational policies is only one-tenth higher than nationally uniform policies. Such results hinge on two critical technologies (advanced biofuels and electricity) for which inter-state trade ameliorates the economic efficiencies that might arise with heterogeneous action.

Peng, Wei↗
DOE OSTI2025

A Reinforcement Learning Approach to Augment Conventional PID Control in Nuclear Power Plant Transient Operation

The ability of nuclear reactors to operate their power conversion cycles more flexibly will enhance their value to energy grids with variable pricing. Current nuclear control systems are typically classical controllers that are often based on proportional-integral-derivative (PID) control. This paper presents a method of augmenting the existing PID control for difficult transient operations in nuclear power plants using a reinforcement learning–derived feedforward signal applied in real time. The agents, which are trained on a test thermal load-following problem, are designed to improve steam generator outlet temperature control for a range of fast load-following scenarios covering ramp rates from 9%/min to 15%/min. Several reinforcement learning algorithms were initially investigated for the training of the feedforward agents with deep Q-learning (DQN) and proximal policy optimization (PPO) networks, which were found to be the most promising. The DQN controllers utilize discrete actions, giving them a better disturbance rejection at steady state but inconsistent response to initial temperature deviations. In contrast, PPO-trained agents, which take continuous actions except for a dead zone around zero, were shown to have the best combination of high disturbance rejection at steady state and good tracking of the desired temperature value. The ability of the PPO agent was also examined, with the average time of decision making found to be on the order of 1 ms. The fault properties of the controller under the loss of the reinforcement learning agent feedforward signal were also examined. The controller showed strong performance in situations of “no-signal” faults. but was less good at handling “stuck-at” faults, where the feedforward signal remains at a set value. In both cases, however, the PID was able to successfully maintain stability, eventually returning the system to a steady state. It is hoped that this work will allow for the proposed control architecture to be examined for more difficult control problems such that it may eventually be used to adapt existing nuclear plants for more aggressive load-following on grids of the future.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2021

Gaming the beamlines—employing reinforcement learning to maximize scientific outcomes at large-scale user facilities

Abstract Beamline experiments at central facilities are increasingly demanding of remote, high-throughput, and adaptive operation conditions. To accommodate such needs, new approaches must be developed that enable on-the-fly decision making for data intensive challenges. Reinforcement learning (RL) is a domain of AI that holds the potential to enable autonomous operations in a feedback loop between beamline experiments and trained agents. Here, we outline the advanced data acquisition and control software of the Bluesky suite, and demonstrate its functionality with a canonical RL problem: cartpole. We then extend these methods to efficient use of beamline resources by using RL to develop an optimal measurement strategy for samples with different scattering characteristics. The RL agents converge on the empirically optimal policy when under-constrained with time. When resource limited, the agents outperform a naive or sequential measurement strategy, often by a factor of 100%. We interface these methods directly with the data storage and provenance technologies at the National Synchrotron Light Source II, thus demonstrating the potential for RL to increase the scientific output of beamlines, and layout the framework for how to achieve this impact.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Learning to Unknot

We introduce natural language processing into the study of knot theory, as made natural by the braid word representation of knots. We study the UNKNOT problem of determining whether or not a given knot is the unknot. After describing an algorithm to randomly generate N-crossing braids and their knot closures and discussing the induced prior on the distribution of knots, we apply binary classification to the UNKNOT decision problem. We find that the Reformer and shared-QK Transformer network architectures outperform fully-connected networks, though all perform at 95% accuracy. Perhaps surprisingly, we find that accuracy increases with the length of the braid word, and that the networks learn a direct correlation between the confidence of their predictions and the degree of the Jones polynomial. Finally, we utilize reinforcement learning (RL) to find sequences of Markov moves and braid relations that simplify knots and can identify unknots by explicitly giving the sequence of unknotting actions. Trust region policy optimization (TRPO) performs consistently well, reducing 80% of the unknots with up to 96 crossings we tested to the empty braid word, and thoroughly outperformed other RL algorithms and random walkers. Studying these actions, we find that braid relations are more useful in simplifying to the unknot than one of the Markov moves.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
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↗
DOE OSTI2021

Hybrid Imitation Learning for Real-Time Service Restoration in Resilient Distribution Systems

Self-healing capability is a critical factor for a resilient distribution system, which requires intelligent agents to automatically perform service restoration online, including network reconfiguration and reactive power dispatch. Here, the article proposes the imitation learning framework for training such an agent, where the agent will interact with an expert built based on the mixed-integer program to learn its optimal policy, and therefore significantly improve the training efficiency compared with exploration-dominant reinforcement learning (RL) methods. This significantly improved training efficiency makes the training problem under N-k scenarios tractable. A hybrid policy network is proposed to handle tie-line operations and reactive power dispatch simultaneously to further improve the restoration performance. The 33-bus and 119-bus systems with N-k disturbances are employed to conduct the training. The results indicate that the proposed method outperforms traditional RL algorithms such as the deep-Q network.

42 ENGINEERING↗
DOE OSTI2019

GPLADD: Quantifying Trust in Government and Commercial Systems A Game-Theoretic Approach

Trust in a microelectronics-based system can be characterized as the level of confidence that a system is free of subversive alterations made during system development, or that the development process of a system has not been manipulated by a malicious adversary. Trust in systems has become an increasing concern over the past decade. This report introduces a novel game-theoretic framework, called GPLADD (Graph-based Probabilistic Learning Attacker and Dynamic Defender), for analyzing and quantifying system trustworthiness at the end of the development process, through the analysis of risk of development-time system manipulation. GPLADD represents attacks and attacker-defender contests over time. Here, time is an explicit constraint and allows incorporating the informational asymmetries between the attacker and defender into analysis. GPLADD includes an explicit representation of attack steps via multi-step attack graphs, attacker and defender strategies, and player actions at different times. GPLADD allows quantifying the attack success probability over time and the attacker and defender costs based on their capabilities and strategies. This ability to quantify different attacks provides an input for evaluation of trust in the development process. We demonstrate GPLADD on an example attack and its variants. We develop a method for representing success probability for arbitrary attacks and derive an explicit analytic characterization of success probability for a specific attack. We present a numeric Monte Carlo study of a small set of attacks, quantify attack success probabilities, attacker and defender costs, and illustrate the options the defender has for limiting the attack success and improving trust in the development process.

37 attack↗
DOE OSTI2024

PPO And Friends

PPO and Friends (PPOAF) is a pytorch implementation of proximal policy optimization for single- and multi-agent reinforcement learning (the PPO), along with several optimizations and add-ons (the Friends) to enable efficient MPI-parallelized model training on HPC clusters.

Maguire, AlisterO↗
DOE OSTI2017

University of Rhode Island Regional Earth Systems Center

The primary objective of this program was to establish the URI Regional Earth System Center (“Center”) that would enhance overall societal wellbeing (health, financial, environmental) by utilizing the best scientific information and technology to achieve optimal policy decisions with maximum stakeholder commitment for energy development, coastal environmental management, water resources protection and human health protection, while accelerating regional economic growth. The Center was to serve to integrate existing URI institutional strengths in energy, coastal environmental management, water resources, and human wellbeing. This integrated research, educational and public/private sector outreach Center was to focus on local, state and regional resources. The centerpiece activity of the Center was in the development and implementation of integrated assessment models (IAMs) that both ‘downscaled’ global observations and interpolated/extrapolated regional observations for analyzing the complexity of interactions among humans and the natural climate system to further our understanding and, ultimately, to predict the future state of our regional earth system. The Center was to begin by first ‘downscaling’ existing global earth systems management tools for studying the causes of local, state and regional climate change and potential social and environmental consequences, with a focus on the regional resources identified above. The Center would ultimately need to address the full feedbacks inherent in the nonlinear earth systems by quantifying the “upscaled” impacts of those regional changes on the global earth system. Through an interacting suite of computer simulations that are informed by observations from the nation’s evolving climate observatories, the Center activities integrates climate science, technology, economics, and social policy into forecasts that will inform solutions to pressing issues in regional climate change science, ‘green economy’ investment and climate policy. These project objectives were designed as part of a 5-year program, which would have constituted the initial phase for the establishment of the Center. Almost immediately (i.e. before receiving even the first year of funding) we were informed that we would not be receiving any funding beyond the initial phase; one year. This seriously impacted our ability to deliver on our objectives and, with that, a re-scoping of the Center priorities was designed to fit the 1-year constraints on funding. It was decided that, given the Center’s emphasis on building IAMs, the best way to proceed was to first focus on one particularly important component of the IAM – a natural sciences model that would be useful for research and forecasting of the circualation/ecology/biogeochemistry of RI’s coastal waters. We have succeeded on that necessarily more limited objective, as we will describe below.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

Defender Policy Evaluation and Resource Allocation against MITRE ATT&CK Data and Evaluations

Protecting against multi-step attacks of uncertain duration and timing forces defenders into an indefinite, always ongoing, resource-intensive response. To effectively allocate resources, a defender must be able to analyze multi-step attacks under assumption of constantly allocating resources against an uncertain stream of potentially undetected attacks. To achieve this goal, we present a novel methodology that applies a game-theoretic approach to the attack, attacker, and defender data derived from MITRE´s ATT&CK ® Framework. Time to complete attack steps is drawn from a probability distribution determined by attacker and defender strategies and capabilities. This constraints attack success parameters and enables comparing different defender resource allocation strategies. By approximating attacker-defender games as Markov processes, we represent the attacker-defender interaction, estimate the attack success parameters, determine the effects of attacker and defender strategies, and maximize opportunities for defender strategy improvements against an uncertain stream of attacks. This novel representation and analysis of multi-step attacks enables defender policy optimization and resource allocation, which we illustrate using the data from MITRE´ s APT3 ATT&CK ® Framework.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Reinforcement Learning Approach to Cybersecurity in Space (RELACSS)

Securing satellite groundstations against cyber-attacks is vital to national security missions. However, these cyber threats are constantly evolving. As vulnerabilities are discovered and patched, new vulnerabilities are discovered and exploited. In order to automate the process of discovering existing vulnerabilities and the means to exploit them, a reinforcement learning framework is presented in this report. We demonstrate that this framework can learn to successfully navigate an unknown network and detect nodes of interest despite the presence of a moving target defense. The agent then exfiltrates a file of interest from the node as quickly as possible. This framework also incorporates a defensive software agent that learns to impede the attacking agents progress. This setup allows for the agents to work against each other and improve their abilities. We anticipate that this capability will help uncover unforeseen vulnerabilities and the means to mitigate them. The modular nature of the framework enables users to swap out learning algorithms and modify the reward functions in order to adapt the learning tasks to various use cases and environments. Several algorithms, viz., tabular Q learning, deep Q networks, proximal policy optimization, advantage actor-critic, generative adversarial imitation learning, are explored for the agents and the results highlighted. The agent learns to solve the tasks in a light-weight abstract environment. Once the agent learns to perform sufficiently well, it can be deployed in a minimega virtual machine environment (or a real network) with wrappers that map abstract actions to software commands. The agent also uses a local representation of the actions called a ‘slot-mechanism’. This allows the agent to learn in a certain network and generalize it to different networks. The defensive agent learns to predict the actions taken by an offensive agent and uses that information to anticipate the threat. This information can then either be used to raise an alarm or to take actions to thwart the attack. We believe that with the appropriate reward design, a representative environment, and action set, this framework can be generalized to tackle other cybersecurity tasks. By sufficiently training these agents, we can anticipate vulnerabilities leading to robust future designs. We can also deploy automated defensive agents that can help secure satellite groundstation and their vital national security missions.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗
NASA NTRS1978

Life cycle costing with a discount rate

This article studies life cycle costing for a capability needed for the indefinite future, and specifically investigates the dependence of optimal policies on the discount rate chosen. The two costs considered are reprocurement cost and maintenance and operations (M and O) cost. The procurement price is assumed known, and the M and O costs are assumed to be a known function, in fact, a non-decreasing function, of the time since last reprocurement. The problem is to choose the optimum reprocurement time so as to minimize the quotient of the total cost over a reprocurement period divided by the period. Or one could assume a discount rate and try to minimize the total discounted costs into the indefinite future. It is shown that the optimum policy in the presence of a small discount rate hardly depends on the discount rate at all, and leads to essentially the same policy as in the case in which discounting is not considered.

Posner, E. C.↗
NASA NTRS1987

Optimal service allocation among two heterogeneous traffic types with no queueing

Two communication traffic streams with Poisson statistics arrive at a network node. These are to be transmitted across a channel with a total bandwidth capacity of C slots. Messages not accepted at the node are assumed to be lost. Under the assumptions of exponential service time distributions, the problem of dynamic allocation of available channel bandwidth among the two traffic types is studied in order to minimize a weighted sum of blocking probabilities. Modeling the system as a two-dimensional Markov chain is studied to minimize a weighted sum of blocking probabilities. Modeling the system as a two-dimensional Markov chain, it is shown by an application of dynamic programming principles that the optimal policy has the form of a 'switching curve'.

Lambadaris, I.↗
NASA NTRSDate not supplied

Adaptive Stress Testing of Collision Avoidance Systems for Small UASs with Deep Reinforcement Learning

The next-generation Airborne Collision Avoidance System for smaller UASs (ACAS sXu) is currently being developed and tested by the Federal Aviation Administration (FAA) to provide detect-and-avoid capability for small unmanned aircraft operating beyond line-of-sight. Due to the complexity and safety-critical nature of the system, safety validation is important not only for the certification of the final system, but also for informing changes during the iterative development process. In this paper, we analyze a prototype of ACAS sXu in simulated aircraft encounters to discover scenarios of small near mid-air collisions (sNMACs), an important safety event in which two aircraft come closer than 50 feet horizontally and 15 feet vertically. Due to the size and complexity of the system as well as rarity of sNMAC events, traditional methods such as Monte Carlo testing often require informed setup and targeting to elicit failures. However, such a dependence on domain knowledge can be incompatible with the independent verification and validation (IV&V) process, the aim of which is to discover unforeseen issues. To address these challenges, we apply an accelerated validation method called adaptive stress testing (AST) to find the most likely sNMAC scenarios without reliance on system introspection. AST uses reinforcement learning to adapt the search towards the most promising areas of the search space as it progresses. We use a state-of-the-art deep reinforcement learning algorithm, proximate policy optimization, to more efficiently search the large and continuous state space. We find that this approach significantly improves the performance of AST compared to a prior approach based on Monte Carlo tree search. We perform experiments using AST to find sNMAC events under various encounter configurations, varying parameters pertaining to dynamics and coordination. Our experiments show AST to be very effective at finding sNMAC scenarios. We summarize our findings, presenting high-level categories of discovered sNMACs and specific examples of encounters in each category.

aircraft collision avoidance↗
NASA NTRS2021

Scheduling the NASA Deep Space Network with Deep Reinforcement Learning

With three complexes spread evenly across the Earth, NASA’s Deep Space Network (DSN) is the primary means of communications as well as a significant scientific instrument for dozens of active missions around the world. A rapidly rising number of spacecraft and increasingly complex scientific instruments with higher bandwidth requirements have resulted in demand that exceeds the network’s capacity across its 12 antennae. The existing DSN scheduling process operates on a rolling weekly basis and is time-consuming; for a given week, generation of the final baseline schedule of spacecraft tracking passes takes roughly 5 months from the initial requirements submission deadline, with several weeks of peer-to-peer negotiations in between. This paper proposes a deep reinforcement learning (RL) approach to generate candidate DSN schedules from mission requests and spacecraft ephemeris data with demonstrated capability to address real-world operational constraints. A deep RL agent is developed that takes mission requests for a given week as input, and interacts with a DSN scheduling environment to allocate tracks such that its reward signal is maximized. A comparison is made between an agent trained using Proximal Policy Optimization and its random, untrained counterpart. The results represent a proof-of-concept that, given a well-shaped reward signal, a deep RL agent can learn the complex heuristics used by experts to schedule the DSN. A trained agent can potentially be used to generate candidate schedules to bootstrap the scheduling process and thus reduce the turnaround cycle for DSN scheduling.

Wilson, Brian↗