Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “reinforcement Learning”

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 19 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Reinforcement learning building control approach harnessing imitation learning

Reinforcement learning (RL) has shown significant success in sequential decision making in fields like autonomous vehicles, robotics, marketing and gaming industries. This success has attracted the attention to the RL control approach for building energy systems which are becoming complicated due to the need to optimize for multiple, potentially conflicting, goals like occupant comfort, energy use and grid interactivity. However, for real world applications, RL has several drawbacks like requiring large training data and time, and unstable control behavior during the early exploration process making it infeasible for an application directly to building control tasks. To address these issues, an imitation learning approach is utilized herein where the RL agents starts with a policy transferred from accepted rule based policies and heuristic policies. This approach is successful in reducing the training time, preventing the unstable early exploration behavior and improving upon an accepted rule-based policy - all of these make RL a more practical control approach for real world applications in the domain of building controls.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
NASA NTRS2004

Time-Extended Policies in Mult-Agent Reinforcement Learning

Reinforcement learning methods perform well in many domains where a single agent needs to take a sequence of actions to perform a task. These methods use sequences of single-time-step rewards to create a policy that tries to maximize a time-extended utility, which is a (possibly discounted) sum of these rewards. In this paper we build on our previous work showing how these methods can be extended to a multi-agent environment where each agent creates its own policy that works towards maximizing a time-extended global utility over all agents actions. We show improved methods for creating time-extended utilities for the agents that are both "aligned" with the global utility and "learnable." We then show how to crate single-time-step rewards while avoiding the pi fall of having rewards aligned with the global reward leading to utilities not aligned with the global utility. Finally, we apply these reward functions to the multi-agent Gridworld problem. We explicitly quantify a utility's learnability and alignment, and show that reinforcement learning agents using the prescribed reward functions successfully tradeoff learnability and alignment. As a result they outperform both global (e.g., team games ) and local (e.g., "perfectly learnable" ) reinforcement learning solutions by as much as an order of magnitude.

Tumer, Kagan↗
DOE OSTI2024

Optimal Management of Grid-Interactive Efficient Buildings via Safe Reinforcement Learning

Reinforcement learning (RL)-based methods have achieved significant success in managing grid-interactive efficient buildings (GEBs). However, RL does not carry intrinsic guarantees of constraint satisfaction, which may lead to severe safety consequences. Besides, in GEB control applications, most existing safe RL approaches rely only on the regularisation parameters in neural networks or penalty of rewards, which often encounter challenges with parameter tuning and lead to catastrophic constraint violations. To provide enforced safety guarantees in controlling GEBs, this paper designs a physics-inspired safe RL method whose decision-making is enhanced through safe interaction with the environment. Different energy resources in GEBs are optimally managed to minimize energy costs and maximize customer comfort. The proposed approach can achieve strict constraint guarantees based on prior knowledge of a set of developed hard steady-state rules. Simulations on the optimal management of GEBs, including heating, ventilation, and air conditioning (HVAC), solar photovoltaics, and energy storage systems, demonstrate the effectiveness of the proposed approach.

Huo, Xiang↗
DOE OSTI2024

Elucidating Abnormal Grain Growth in Thermomagnetic Processed Materials with Transfer Learning and Reinforcement Learning

The goal of this research program is to establish the mechanism governing local grain boundary motion, which is needed to design and process desirable microstructures for better performance, by identifying the relative contributions of grain boundary (GB) energy and mobility to grain growth. Classical models for grain growth assume that the primary mechanism for reducing the total interfacial energy is area reduction and that GB restructuring is not significant. This assumption implies that grain growth is locally driven by curvature. However, recent experimental observations using new non-destructive 3D x-ray diffraction microscopy techniques (3D-XRM) reveal that classic descriptors (i.e., curvature, number of neighbors, grain size) do not predict real grain growth. Instead, local GB motion appears to be governed by its energy relative to its neighbors such that low-energy boundaries replace those of higher energy. However, simulations that incorporate GB energy anisotropy still fail to reproduce these observations. These discrepancies suggest that the common assumption for grain growth theory must be re-examined to predict and, thus, control microstructure evolution in real polycrystals. A significant challenge to testing this assumption is due to anisotropic GB mobility. Mobility may cause abnormal grain growth or affect the final grain shapes or growth rate but its true contributions are unknown because it is difficult to measure. For example, observations in Fe have found that grains associated with high energy and high mobility boundaries tend to experience abnormal grain growth, whereas abnormal grain growth is associated with low energy and high mobility boundaries in alumina. As mobility and energy both control GB motion, it is challenging to isolate the local driving forces necessary to test the common assumption that the primary mechanism is area reduction. The novelty of this work is the use of machine learning tools to capture GB mobility and energy from 3D-XRM measurements in polycrystals to test the common assumption used in grain growth models. Machine learning can capture high-order correlations in dynamic systems like those found in the evolving GB topology. The PIs have developed a physics-regularized interpretable machine learning microstructure evolution (PRIMME) model that accurately replicates the grain growth behavior of its trained data set.

36 MATERIALS SCIENCE↗
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 OSTI2022

On-policy learning-based deep reinforcement learning assessment for building control efficiency and stability

Artificial intelligence technologies have emerged as a game changer not only in specific applications such as image recognition and machine translation but also in many scientific domains. In particular, as deep reinforcement learning (DRL) has shown great success in complex control problems, DRL-based control has been considered as a potential solution to efficiently control and manage building systems. However, broad assessment of DRL-based building control is still required to characterize their pros and cons in comparison with conventional building control methods (e.g., rule-based feedback controls). In this paper, we assessed DRL-based controls with on-policy learning-based algorithms and continuous control actions for cooling control of large office buildings in the summer season to minimize whole-building energy use and occupant discomfort. We compared DRL-based control methods with two baseline control methods: (1) a pre-determined schedule with supply temperature and static pressure setpoints, and (2) advanced reset method that adjusts setpoints based on heuristic rules, i.e., ASHRAE Guideline 36. We also tested the DRL algorithms to evaluate their performances in multiple climate locations. We found that DRL-based control methods outperformed the baseline control methods in terms of energy savings while maintaining a thermal comfort. DRL reduced energy use between ~4%–22% on average compared to the baseline methods, depending on climate location. We also evaluated DRL-based control in terms of control stability and showed that DRL-based methods should address the span of hardware lifetimes in practical operations.

control stability↗
NASA NTRS1992

Using Fuzzy Logic for Performance Evaluation in Reinforcement Learning

Current reinforcement learning algorithms require long training periods which generally limit their applicability to small size problems. A new architecture is described which uses fuzzy rules to initialize its two neural networks: a neural network for performance evaluation and another for action selection. This architecture is applied to control of dynamic systems and it is demonstrated that it is possible to start with an approximate prior knowledge and learn to refine it through experiments using reinforcement learning.

Berenji, Hamid R.↗
DOE OSTI2026

Learning to Trigger: Reinforcement Learning at the Large Hadron Collider

High-throughput scientific facilities such as the Large Hadron Collider depend on real-time event filtering (\textit{triggering}) under tight constraints on bandwidth, latency, and storage. In practice, trigger menus are largely static and hand-tuned and can become suboptimal as detector conditions, pileup, and background composition drift over time. We cast online threshold tuning as a sequential decision-making problem: a reinforcement learning agent ingests streaming summaries of recent rates and signal-sensitive features and updates trigger thresholds to maximize signal efficiency while tracking a target background rate within a tolerance band. We adapt Group-Filtered Policy Optimization (GFPO) to streaming control and introduce two variants (GFPO-F, GFPO-FR) that enforce background rate feasibility during training. On a benchmark that emulates realistic collider operation, we study two representative triggers: a total transverse energy ($H_{T}$) trigger sensitive to pileup variation, and an anomaly-detection (AD) trigger based on reconstruction loss for rare or non-standard signatures. On Monte Carlo streams, our agent increases the fraction of in-tolerance time intervals by 48% ($H_T$) and 28% (AD), with a cumulative gain of up to 2% in signal efficiency on those in-tolerance intervals. Transferring from simulation to \emph{real} collision data (CMS Run 283408), the same agent, without fine-tuning, achieves a 56% ($H_T$) and 28% (AD) in-tolerance improvement over baselines, with further signal-efficiency gain on both triggers. To our knowledge, this is the \emph{first} demonstration of RL-based trigger control on real Large Hadron Collider collision data. Code is available at https://github.com/Zixind/GFPO_LHC (see repo for details).

Ding, Zixin [Chicago U.]↗
DOE OSTI2023

Reinforcement-Learning-Based Smart Water Heater Control: An Actual Deployment

Utilizing smart control algorithms for electric water heaters (EWHs) is essential for fully harnessing the demand response (DR) potential of EWHs. For this reason, the use of reinforcement learning (RL) algorithms for EWHs has received increasing attention in recent years. However, existing RL approaches are either simulation-based or use pretrained RL agents. To this end, this paper presents the real-world deployment of a set of model-free RL approaches that aim to minimize the electricity cost of a EWH under a time-of-use electricity pricing policy using standard DR commands (e.g., shed, load up). The experiment results showed that the RL agents can help save electricity cost in the range of 11% to 14% compared to the baseline operation. This study demonstrated that RL-based EWH controllers can be deployed in real world without any prior training and can still save electricity cost.

deep learning↗
DOE OSTI2022

PowerGridworld: A Framework for Multi-Agent Reinforcement Learning in Power Systems: Preprint

We present the PowerGridworld software package to provide users with a light-weight, modular, and customizable framework for creating power systems-focused, multi-agent gym environments that readily integrate with existing training frameworks for reinforcement learning (RL). While many frameworks exist for training multi-agent (MA) RL policies, none exist to rapidly prototype and develop the environments themselves, especially in the context of heterogeneous (composite, multi-device) power systems where power flow solutions are required to define grid-level variables and costs. PowerGridworld is an open-source software package that helps to fill this gap. To highlight PowerGridworld's key features, we present two case studies and demonstrate learning multi-agent RL policies using both OpenAI's MADDPG and RLLib's PPO algorithms where, in both cases, at least some subset of agents incorporate elements of the power flow solution at each time step as part of their reward (negative cost) structures.

MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Deep Reinforcement Learning Based HVAC Control for Reducing Carbon Footprint of Buildings

In this paper, we present our work on deep reinforcement learning (DRL) based intelligent control of Heating, Ventilation, and Air Conditioning (HVAC) with the goal of reducing carbon emission. We performed this task using 1) Marginal Operating Emission Rates (MOER), where the objective was to shift the demand to the low emission period of the day and 2) Time-Of-Use (TOU) demand-response price where the objective was to shift the demand to low price period of the day. This was achieved by learning an optimal pre-cooing strategy. We found the carbon emission reduction in the range of 6%-16% depending on the opportunity presented by the MOER signal. Similarly, we observed the carbon emission reduction in the range of 23%-29% during the peak price period when TOU price was used. The results clearly demonstrated the applicability of our approach in reducing the carbon footprint of the building.

carbon emission↗
DOE OSTI2023

Quantifying Uncertainty of Deep Reinforcement Learning Based Decision Making for Operations and Maintenance of Nuclear Power Plant

This paper summarizes research that integrates condition monitoring and prognostics with decision making for nuclear power plant operations and maintenance. As part of this research, we have developed an online asset management tool to help reduce life-cycle maintenance and repair costs. Using the latest advancements in condition monitoring, supply chain analytics, and deep reinforcement learning, we have created a predictive maintenance tool that can optimize the maintenance and spare-part management of a repairable nuclear system. To demonstrate these methods, preliminary studies were conducted on a simple, representative maintenance system undergoing a stochastic degradation process that requires repairs or replacement to continue operation. Through Monte Carlo simulations, we were able to reduce maintenance spending by approximately 50% compared to optimized, time-based maintenance strategies. Not only does the decision maker reduce the average life-cycle costs, it also minimizes the chance of high cost scenarios, lowering the variance of the expected cost distributions, and reducing overall financial risk. Furthermore, this work also studies the ability of the decision maker to handle various levels of noise from observation uncertainty. By introducing uncertainty into the decision-making process, we have quantified the robustness and resiliency of the decision maker, as well as identified necessary levels of observability to demonstrate cost effectiveness.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2023

Operation Optimization using Reinforcement Learning with Integrated Artificial Reasoning Framework

In large and complex systems, operational decision-making requires a systematic analysis with a vast amount of data from both process parameters and component status monitoring. In this paper, we present an integrated artificial reasoning approach for system state transition models that can help operational decision-making with explainable and traceable reasoning. The integrated artificial reasoning framework is a physics-based approach of defining the system structure in a Bayesian network, so we leveraged it in a Markov decision process (MDP) for finding optimal operational solutions. In our proposed framework, the MDP is implemented on a dynamic Bayesian network (DBN), which represents causalities in a system. The multilevel flow modeling was utilized in order to extract these causalities in a more efficient and objective manner. Since multilevel flow modeling is based on the fundamental energy and mass conservation laws, the target system is decomposed into several mass, energy, and information structures, which serve as the basis for a DBN. The MDP consists of the processes of finding a solution for the Bellman equation, which can be derived from the conditional probability equations of the constructed DBN. System operators can capture stochastic system dynamics as multiple subsystem state transitions based on their physical relations and uncertainties coming from the component degradation process or random failures. We analyzed a simplified example system to illustrate finding an optimal operational policy with this approach.

99 GENERAL AND MISCELLANEOUS↗
NASA NTRS1996

Fuzzy Q-Learning for Generalization of Reinforcement Learning

Fuzzy Q-Learning, introduced earlier by the author, is an extension of Q-Learning into fuzzy environments. GARIC is a methodology for fuzzy reinforcement learning. In this paper, we introduce GARIC-Q, a new method for doing incremental Dynamic Programming using a society of intelligent agents which are controlled at the top level by Fuzzy Q-Learning and at the local level, each agent learns and operates based on GARIC. GARIC-Q improves the speed and applicability of Fuzzy Q-Learning through generalization of input space by using fuzzy rules and bridges the gap between Q-Learning and rule based intelligent systems.

Berenji, Hamid R.↗
DOE OSTI2023

Interpreting Primal-Dual Algorithms for Constrained Multiagent Reinforcement Learning

Constrained multiagent reinforcement learning (C-MARL) is gaining importance as MARL algorithms find new applications in real-world systems ranging from energy systems to drone swarms. Most C-MARL algorithms use a primal-dual approach to enforce constraints through a penalty function added to the reward. In this paper, we study the structural effects of this penalty term on the MARL problem. First, we show that the standard practice of using the constraint function as the penalty leads to a weak notion of safety. However, by making simple modifications to the penalty term, we can enforce meaningful probabilistic (chance and conditional value at risk) constraints. Second, we quantify the effect of the penalty term on the value function, uncovering an improved value estimation procedure. We use these insights to propose a constrained multiagent advantage actor critic (C-MAA2C) algorithm. Simulations in a simple constrained multiagent environment affirm that our reinterpretation of the primal-dual method in terms of probabilistic constraints is effective, and that our proposed value estimate accelerates convergence to a safe joint policy.

chance constraints↗
DOE OSTI2024

An innovative heterogeneous transfer learning framework to enhance the scalability of deep reinforcement learning controllers in buildings with integrated energy systems

Deep Reinforcement Learning (DRL)-based control shows enhanced performance in the management of integrated energy systems when compared with Rule-Based Controllers (RBCs), but it still lacks scalability and generalisation due to the necessity of using tailored models for the training process. Transfer Learning (TL) is a potential solution to address this limitation. However, existing TL applications in building control have been mostly tested among buildings with similar features, not addressing the need to scale up advanced control in real-world scenarios with diverse energy systems. This paper assesses the performance of an online heterogeneous TL strategy, comparing it with RBC and offline and online DRL controllers in a simulation setup using EnergyPlus and Python. The study tests the transfer in both transductive and inductive settings of a DRL policy designed to manage a chiller coupled with a Thermal Energy Storage (TES). The control policy is pre-trained on a source building and transferred to various target buildings characterised by an integrated energy system including photovoltaic and battery energy storage systems, different building envelope features, occupancy schedule and boundary conditions (e.g., weather and price signal). The TL approach incorporates model slicing, imitation learning and fine-tuning to handle diverse state spaces and reward functions between source and target buildings. Results show that the proposed methodology leads to a reduction of 10% in electricity cost and between 10% and 40% in the mean value of the daily average temperature violation rate compared to RBC and online DRL controllers. Moreover, online TL maximises self-sufficiency and self-consumption by 9% and 11% with respect to RBC. Conversely, online TL achieves worse performance compared to offline DRL in either transductive or inductive settings. However, offline Deep Reinforcement Learning (DRL) agents should be trained at least for 15 episodes to reach the same level of performance as the online TL. Therefore, the proposed online TL methodology is effective, completely model-free and it can be directly implemented in real buildings with satisfying performance.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2021

A novel implicit hybrid machine learning model and its application for reinforcement learning

A novel methodology to develop implicit hybrid models is presented. PyTorch is used to integrate physics-based equations with machine learning models. Automatic differentiation of the hybrid model is leveraged to solve the implicit equations. Iterative solving enables gradient based updates to the machine learning model. The novel methodology is compared to an explicit hybrid approach on a continuously stirred tank reactor (CSTR). The novel method results in a lower modelling error. Both hybrid models effectively train with noisy data. To test the implicit hybrid model, it is employed as a reinforcement learning (RL) training model. The RL algorithm trained on the hybrid model outperforms real time optimization of the CSTR and performs nearly as well as RL trained directly on the CSTR and a traditional gradient based approach. Training RL directly on the CSTR requires over 60,000 system interactions compared to 6000 historical data points for hybrid model development.

42 ENGINEERING↗
DOE OSTI2021

Learning Sequential Distribution System Restoration via Graph-Reinforcement Learning

We report a distribution service restoration algorithm as a fundamental resilient paradigm for system operators provides an optimally coordinated, resilient solution to enhance the restoration performance. The restoration problem is formulated to coordinate distribution generators and controllable switches optimally. A model-based control scheme is usually designed to solve this problem, relying on a precise model and resulting in low scalability. To tackle these limitations, this work proposes a graph-reinforcement learning framework for the restoration problem. We link the power system topology with a graph convolutional network, which captures the complex mechanism of network restoration in power networks and understands the mutual interactions among controllable devices. Latent features over graphical power networks produced by graph convolutional layers are exploited to learn the control policy for network restoration using deep reinforcement learning. The solution scalability is guaranteed by modeling distributed generators as agents in a multi-agent environment and a proper pre-training paradigm. Comparative studies on IEEE 123-node and 8500-node test systems demonstrate the performance of the proposed solution.

24 POWER TRANSMISSION AND DISTRIBUTION↗