Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “agent based model”

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 289 records · Page 16
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Siting and sizing of public–private charging stations impacts on household and electric vehicle fleets

To facilitate the provision of electric vehicle charging stations (EVCS) in urban areas, this study investigates the benefits of co-locating fleet-owned chargers with public charging stations to enable construction incentives and cord-sharing cost savings. Shared EVCS can serve charging demand from both user types: private (household) EV owners and those managing fleet vehicles – like shared and fully automated EV (SAEV) fleets. Using POLARIS to simulate all person-travel across the 6-county Austin, Texas region, new EVCS were sited and sized with DC fast-charging (DCFC) plugs to lower operating and construction costs while providing public + private (PP) service across an 81-square-mile core geofence (where 200 SAEVs were active) over 24-hour days. When co-location is permitted, 115 DCFC cords were added to the 23 existing (publicly available) stations to enable SAEVs and household EVs (HHEVs) charging access, within the geofence. Each 250-mile-range SAEV was simulated to travel an average of 330 miles per day, serve over 92 person-trips, and recharge 2.7 times a day (for 2.4 h per session). The new DCFC plugs were primarily added to public EVCS at shopping centers and schools, and in residential settings along freeways. The average plug served 4.8 EVs per day. Most co-located PP EVCS permitted immediate (no-wait) charging, except for 2 stations along freeways that averaged 8 min of wait time to begin charging. In conclusion, the co-location strategy lowered fleet owners’ initial EVCS construction costs by 12 % (thanks to cord-sharing to avoid cord duplication), while reducing SAEV wait times to just 3.1 min (versus 10.7 min if SAEV managers had to build and operate their own EVCS).

EV charging modeling↗
DOE OSTI2025

Datum: A Scientific Metadata Catalog

The data catalog market is currently flooded with a myriad of different products, but none serve the scientific community well. There are cloud-native tools like Databricks, Snowflake,to on-premise solutions like Collibra and Datahub. The common failing of all these tools however, is their inability to serve the scientific data community directly. Most catalogs are targeted towards financial, health, or user data - not sensor or scientific domain data. They also prioritize integrations that often don’t exist or are just starting to be used in the scientific realm - all while ignoring common scientific tools and file types. Datum is a catalog which targets the scientific data directly, including the tools and networks in which those tools are used. We work with the producers and consumers of the data where they are, targeting cloud and on-premise with a focus on classified networks. Datum is an Erlang/Elixir application. Technical Features Note: The features listed below are still under development and may change, slightly, upon final delivery of the product. File Formats - Datum has the ability to read additional metadata and provides processing pipelines for the following file formats: Plain Text, PDF, LaTeX, HTML, Open Document Format (.odt), XML, CSV/TSV (and other standard delimiters), OpenDocument Database and Spreadsheets, Geo-Referenced TIFF, Common Data Format, HDF/HDF5, LabView TDMS, Excel, DeltaTables, Parquet, Apache Iceberg, Apache Hudi and many others. Metadata Collection - Scanners for the local and networked file systems and cloud storage providers. Network integration with common databases such as MSSQL and MySQL. User Plugin System - Users are able to provide either file processing, metadata extraction, or sampling plugins in the programming language of their choice. Authentication/Authorization -: OIDC integration, SCIM provisioning and EntraID integration out of the box. Full user and group management system with a “least privilege” operating mode. Governance - Customizable data governance platform; dictate and enforce required metadata, enforce data embargos, and enforce user agreements and NDAs before data access. Ability to create health checks on data, rejecting abandoned or poorly curated data and automatically removing it from the search index. Ability for users to submit corrections. Search - Semantic search is a first class citizen. No licenses to expensive, external software required. Integrated use of vectors and vector-based search allows for AI agent integration at all levels of operation. Metadata Model - Display and control data’s lineage and connections to other data and data directories. Data is modeled after a filesystem - an organization instantly recognizable and navigable by most any user. CLI and SDK - Ships with a Command Line Interface (CLI) tool and with a fully-featured Python SDK. This allows for rapid and programmatic use of Datum by every level of user. Minimal Infrastructure - Datum ships as a single executable file and can be run on any operating system and most CPU architectures. Datum has no reliance on external databases, search indexing tools, or other outside services - and it runs equally well on edge computing devices, cloud services, or in a clustered HPC environment.

darrington, john↗
DOE OSTI2025

CHEMREASONER: Heuristic Search over a Large Language Model’s Knowledge Space using Quantum-Chemical Feedback

The discovery of new catalysts is essential for the design of new and more efficient chemical processes in order to transition to a sustainable future. We introduce an AI-guided computational screening framework unifying linguistic reasoning with quantum-chemistry based feedback from 3D atomistic representations. Our approach formulates catalyst discovery as an uncertain environment where an agent actively searches for highly effective catalysts via the iterative combination of large language model (LLM)-derived hypotheses and atomistic graph neural network (GNN)-derived feedback. Identified catalysts in intermediate search steps undergo structural evaluation based on spatial orientation, reaction pathways, and stability. Scoring functions based on adsorption energies and barriers steer the exploration in the LLM's knowledge space toward energetically favorable, high-efficiency catalysts. We introduce planning methods that automatically guide the exploration without human input, providing competitive performance against expert-enumerated chemical descriptor-based implementations. By integrating language-guided reasoning with computational chemistry feedback, our work pioneers AI-accelerated, trustworthy catalyst discovery.

artificial intelligence↗
DOE OSTI2021

Graph Convolutional Network-Based Topology Embedded Deep Reinforcement Learning for Voltage Stability Control

Topological variations in power system is a common phenomenon and can impose significant challenges to traditional controllers of power system. Recent study revealed the strength of deep reinforcement learning (DRL) based approaches in power system preventive and corrective control. But topological variations are difficult to capture using classical fully connected neural network (FCN) model and has not been explicitly modeled in previous work. Hence, we develop a Graph Convolutional Network (GCN) based DRL framework to tackle topology changes in control design of power system. The GCN model exploits the graph structure of the power network and helps the DRL agent to embed the topology information during learning process. Our GCN based approach is evaluated using the IEEE-39 bus system and it outperforms the FCN-based DRL scheme in terms of training convergence and control performance considering grid topology changes.

Hossain, Ramij Raja↗
DOE OSTI2021

Distributed deep reinforcement learning for simulation control

Abstract Several applications in the scientific simulation of physical systems can be formulated as control/optimization problems. The computational models for such systems generally contain hyperparameters, which control solution fidelity and computational expense. The tuning of these parameters is non-trivial and the general approach is to manually ‘spot-check’ for good combinations. This is because optimal hyperparameter configuration search becomes intractable when the parameter space is large and when they may vary dynamically. To address this issue, we present a framework based on deep reinforcement learning (RL) to train a deep neural network agent that controls a model solve by varying parameters dynamically. First, we validate our RL framework for the problem of controlling chaos in chaotic systems by dynamically changing the parameters of the system. Subsequently, we illustrate the capabilities of our framework for accelerating the convergence of a steady-state computational fluid dynamics solver by automatically adjusting the relaxation factors of the discretized Navier–Stokes equations during run-time. The results indicate that the run-time control of the relaxation factors by the learned policy leads to a significant reduction in the number of iterations for convergence compared to the random selection of the relaxation factors. Our results point to potential benefits for learning adaptive hyperparameter learning strategies across different geometries and boundary conditions with implications for reduced computational campaign expenses 4 4 Data and codes available at https://github.com/Romit-Maulik/PAR-RL . .

42 ENGINEERING↗
DOE OSTI2026

Augmenting LLM-Based Agents for Improved Performance in Pentesting and Commissioning Operational Technology in Critical Infrastructure

Artificial intelligence (AI), and more specifically large language models (LLMs) have the potential for use in penetration testing (“pentesting”) against devices, networks, and computer systems in information technology (IT). We explore the possibility of extending pentesting from IT systems to operational technology (OT) systems, which are more obscure than IT systems in their protocols and design. A challenge therefore exists when applying pretrained LLMs to OT systems as corpora are likely to underrepresent OT systems in comparison to other more prevalent systems. We evaluate augmentations of LLMs with various methods, especially retrieval augmented generation (RAG), to improve performance of the LLMs in the OT domain. In addition to pentesting, some of the testing of these OT devices may include commissioning to ensure that the newly installed devices work correctly. Our framework may also be applied in such cases.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Beyond the List: Bioagent-Agnostic Signatures Could Enable a More Flexible and Resilient Biodefense Posture Than an Approach Based on Priority Agent Lists Alone

As of 2021, the biothreat policy and research communities organize their efforts around lists of priority agents, which elides consideration of novel pathogens and biotoxins. For example, the Select Agents and Toxins list is composed of agents that historic biological warfare programs had weaponized or that have previously caused great harm during natural outbreaks. Similarly, lists of priority agents promulgated by the World Health Organization and the National Institute of Allergy and Infectious Diseases are composed of previously known pathogens and biotoxins. To fill this gap, we argue that the research/scientific and biodefense/biosecurity communities should categorize agents based on how they impact their hosts to augment current list-based paradigms. Specifically, we propose integrating the results of multi-omics studies to identify bioagent-agnostic signatures (BASs) of disease—namely, patterns of biomarkers that accurately and reproducibly predict the impacts of infection or intoxication without prior knowledge of the causative agent. Here, we highlight three pathways that investigators might exploit as sources of signals to construct BASs and their applicability to this framework. The research community will need to forge robust interdisciplinary teams to surmount substantial experimental, technical, and data analytic challenges that stand in the way of our long-term vision. However, if successful, our functionality-based BAS model could present a means to more effectively surveil for and treat known and novel agents alike.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2022

Hybrid-RL-MPC4CLR (Hybird-Reinforcement-Learning-Model-Predictive-Control-for-Reserve-Policy-Assisted-Critical-Load-Restoration-in-Distribution-Grids)

Hybrid-RL-MPC4CLR was developed as a hybrid controller for active distribution grid critical load restoration, combining deep reinforcement learning (RL) and model predictive control (MPC) aiming at maximizing total restored load following an extreme event. The RL determines a policy for quantifying operating reserve requirements, thereby hedging against uncertainty, while the MPC models grid operations incorporating the RL policy actions (i.e., reserve requirements), renewable (wind and solar) power predictions, and load demand forecasts. The developers formulated the reserve requirement determination problem as a sequential decision-making problem based on the Markov Decision Process (MDP) and design an RL learning environment based on the OpenAI Gym framework and MPC simulation. The RL agent reward and MPC objective function aim to maximize and monotonically increase total restored load and minimize load shedding and renewable power curtailment. The software is developed using various software packages in Python. The MPC's optimal power flow (OPF) model is implemented using the Pyomo package, the RL simulation environment is implemented using the MPC simulation with various scenarios of renewable energy and load demand profiles and power outage beginning times, based on the OpenAI Gym framework. The RL agent training is performed using the RLlib Ray package. The RL algorithm is trained offline using historical forecasts of renewable generation and load demand profiles. Simulation analysis and performance tests are conducted using a modified IEEE 13-bus distribution test feeder containing wind turbine, photovoltaic, microturbine, and battery.

Eseye, Abinet Tesfaye↗
DOE OSTI2023

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

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

99 GENERAL AND MISCELLANEOUS↗
DOE OSTI2021

The Response of Precipitation Extremes to the Twentieth- and Twenty-First-Century Global Temperature Change in a Comprehensive Suite of CESM1 Large Ensemble Simulation: Revisiting the Role of Forcing Agents Vs. the Role of Forcing Magnitudes

The response of precipitation extremes (PEs) to global warming is found to be nonlinear in Community Earth System Model version 1 (CESM1) and other global climate models (Pendergrass et al., 2019), which led to the concern that it is not accurate to approximate the response of PE to a single forcing as the difference between simulations with all forcing agents and those that exclude one specific forcing. This calls into question previous model-based results that the sensitivity of PE with warming due to aerosol forcing is significantly larger than that due to greenhouse gases (GHGs). We reevaluate the PE sensitivity to GHGs and aerosols using available CESM1 ensemble simulations. We show that although the PE response to warming is nonlinear in CESM1, especially for the high warming projected in the twenty-first-century, PE sensitivity to aerosols is still significantly stronger than that due to GHGs when the comparison is made within similar warming regimes to avoid the bias induced by the nonlinear behavior. But the difference is smaller than previously estimated. We also conclude that the additivity assumption is largely valid to isolate the PE response due to aerosol forcing from the paired simulations including the “all forcing” experiment when the warming regime is small (e.g., 1°C–2°C in the next few decades when aerosol forcing is projected to decline and becomes a major source of uncertainty for model projection).

58 GEOSCIENCES↗
DOE OSTI2024

chatHPC: Empowering HPC users with large language models

The ever-growing number of pre-trained large language models (LLMs) across scientific domains presents a challenge for application developers. While these models offer vast potential, fine-tuning them with custom data, aligning them for specific tasks, and evaluating their performance remain crucial steps for effective utilization. However, applying these techniques to models with tens of billions of parameters can take days or even weeks on modern workstations, making the cumulative cost of model comparison and evaluation a significant barrier to LLM-based application development. To address this challenge, we introduce an end-to-end pipeline specifically designed for building conversational and programmable AI agents on high performance computing (HPC) platforms. Our comprehensive pipeline encompasses: model pre-training, fine-tuning, web and API service deployment, along with crucial evaluations for lexical coherence, semantic accuracy, hallucination detection, and privacy considerations. Here, we demonstrate our pipeline through the development of chatHPC, a chatbot for HPC question answering and script generation. Leveraging our scalable pipeline, we achieve end-to-end LLM alignment in under an hour on the Frontier supercomputer. We propose a novel self-improved, self-instruction method for instruction set generation, investigate scaling and fine-tuning strategies, and conduct a systematic evaluation of model performance. The established practices within chatHPC will serve as a valuable guidance for future LLM-based application development on HPC platforms.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Development and Implementation of a New AI-Based Tool to Support Fast Reactor Software Model Generation and Validation

This report summarizes FY26 work to develop Maggie, an artificial intelligence-based assistant designed to support software model generation and validation activities for fast reactor analysis codes. The project established a modular, code-agnostic software architecture that separates reusable agent capabilities from code-specific knowledge and tools, with initial implementation focused on the FRP-supported fast reactor safety analysis code SAS4A/SASSYS1 (SAS). A curated SAS-specific knowledge base was assembled from the code manual, training materials, historical analysis reports, and representative input files, and was integrated through retrieval-augmented generation to ground Maggie’s responses in authoritative sources. Maggie was deployed on the internal Argonne network, where it demonstrated practical user-facing capability as a chatbot for answering natural language questions about SAS and retrieving relevant technical information. Demonstration cases also showed that Maggie can generate useful snippets of SAS input for selected modeling tasks, while highlighting current limitations in reliability and consistency for more complex input generation tasks. Overall, the FY26 effort established the technical foundation for an AI-assisted capability intended to improve the efficiency, consistency, and accessibility of fast reactor software model development at Argonne and, with further improvements, to support eventual use by the broader fast reactor community, including industry users of FRP-supported analysis tools.

Thomas, Rachel [Argonne National Laboratory (ANL),↗
DOE OSTI2024

The Transactive Energy Network Template Metamodel

While transactive energy, which is defined as an allocation of electricity based on dynamically discovered values or prices, has been extensively studied, its uptake and use has been slow. This report describes a tool, the transactive network template, which should hasten the creation and uptake of transactive energy networks. Some basic principles of transactive energy are familiar from existing wholesale electricity markets. Locational prices are calculated today for zones within bulk electric transmission systems. Locational prices differ while accounting for the locational costs of electricity generation and the losses and constraints incurred when electricity is transmitted from generators and distributed to consumers. A transactive energy network might include these transmission zones. However, current research strives to apply transactive energy also in electricity distribution circuits, buildings, and even for individual generating and consuming devices. At the same time, researchers explore how to apply transactive energy in real time during increasingly shorter time intervals. Automated computational agents become necessary as transactive energy becomes applied to smaller circuit zones and at faster dynamic timescales. A transactive energy network is an example of a multi-agent system. Each zone in the network is represented by its transactive agent, which makes decisions for and acts on behalf of a business entity that is responsible for and manages one of the circuit regions. A transactive energy network is also an example of a decentralized, distributed control system. Control decisions and responsibilities are distributed among the network’s transactive agents. The transactive agents are independent; that is, there typically is no centralized authority or oversight function. Instead, transactive agents exchange transactive signals and thereby negotiate the prices and quantities of electricity that they will exchange. Initially, the circuit regions and responsibilities of transactive agents appear to be very dissimilar. Each circuit region may comprise transmission, distribution, or building-level circuits. Each has a unique position and electrical connectivity within the transactive energy network. Each possesses unique assets that either generate or consume electricity, and these (e.g., renewable energy generator, diesel generator, aggregate utility load, building load, space conditioning, refrigerator, etc.) may further differ in their price flexibility and in their strategies for responding to dynamic electricity prices. Given such diversity, an implementer’s first inclination might be to start from scratch to define all these devices and to engineer their seemingly unique interactions. Given that each implementer’s perspective may be narrow within a transactive energy network, it is unlikely that uniquely engineered systems would interact well. This is where the transactive network template is applicable. The transactive network template is a metamodel that has been developed to guide implementers as they configure their own transactive agent within a network of such agents. The object-oriented design of the transactive network template provides basic code object types that may be used and extended by implementers to represent each of the assets in their circuit region. These objects further facilitate the transactive agent’s necessary computations, which are divided among responsibilities to schedule power usage, balance electric supply and demand, and coordinate the exchange of electricity with the other transactive agents. This report addresses the conceptual transactive network template design. Implementers are directed to more formal design documents and reference implementations. A Python™-based1 reference implementation of the transactive network template has been coded, and three implementations have been configured to represent a national laboratory and two university campuses. Version 2 of the transactive node template generalizes the market class and its methods to facilitate multiple, and more diverse market coordination mechanisms than were facilitated by and demonstrated using Version 1. Version 3 includes new Appendix B, which addresses the designs of methods that would make dynamic prices track approved electricity rates. In the future, the author wishes to make the transactive network template more generally applicable to networks that require more accurate power flow. Development of the transactive network template is jointly funded by the U.S. Department of Energy (DOE) Energy Efficiency and Renewable Energy and the DOE Office of Electricity. In late 2015, one of the first projects to be funded by the DOE Grid Laboratory Modernization Laboratory Consortium was the Clean Energy and Transactive Campus project, led by Pacific Northwest National Laboratory. DOE funds were matched by an investment by the Washington Department of Commerce through its Clean Energy Fund. The transactive network template was developed to guide the implementation of transactive energy networks within this project’s scope.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2026

BrickQA: Bridging the Semantic Gap in Building Operations with Dynamic Graph Exploration

While standardized ontologies like the Brick schema address data heterogeneity in Building Automation Systems (BAS), accessing this semantic data remains a challenge as domain experts often lack the expertise to formulate complex SPARQL queries. To bridge this gap, we present BrickQA, a Large Language Model (LLM)-based framework that translates natural language into executable SPARQL queries through structured query decomposition, dynamic schema exploration, and inline validation. BrickQA utilizes an iterative reasoning agent to actively navigate graph topology through dynamic exploration actions without requiring exhaustive context injection or model fine-tuning. This approach effectively mitigates hallucinations, particularly in large-scale building knowledge graphs. Empirical evaluation on BuildingQA, a standardized benchmark, demonstrates that BrickQA significantly outperforms ReAct baselines, delivering a 0.291–0.355 absolute F1 improvement while achieving 3 × –12.7 × higher token cost-efficiency. Beyond these metrics, the framework maintains structural fidelity across heterogeneous buildings and remains resilient to ambiguous queries without requiring site-specific fine-tuning. Furthermore, a case study on operational analytics validates the framework’s capability to handle temporal and aggregation constraints, effectively transforming abstract semantic models into actionable facility management insights.1

Ko, Yun-Dam↗
DOE OSTI2024

Trajectory Shaper: A Solution for Disrupted Cooperative Adaptive Cruise Control

Cooperative adaptive cruise control (CACC) can effectively reduce energy consumption, alleviate traffic congestion, and enhance safety. However, communication-related constraints and uncooperative vehicle users can disrupt CACC during real-world operations, significantly undermining the putative benefits of CACC. To alleviate the negative impacts of disrupted CACC, this study develops the trajectory shaper (TS) methods as backup solutions for two scenarios: (i) communication between vehicles is infeasible, and vehicles execute adaptive cruise control (ACC) using local sensor measurements; (ii) follower vehicles reject forming a cooperative platoon and execute their local distributed controllers using the information attained via communication. When communication is infeasible, a distributed TS is devised on each vehicle to modify the sensor measurements, enabling safe and efficient ACC operations. When communication is available but uncooperative agents are involved, the lead vehicle of the platoon executes a centralized TS to modify the information shared with uncooperative agents, achieving optimal platoon-level performance. The centralized and distributed TSs are implemented based on the model predictive control algorithms to yield optimal modifications on input information. Robustness is also factored to tackle model uncertainties during TS operations to ensure safety and efficiency. Numerical experiments validate the control performance of the proposed TSs.

Zhou, Anye [ORNL] (ORCID:0000000301455579)↗
DOE OSTI2025

ADEPT: A Pedagogical Framework for Integrating Agentic AI with Deterministic Scientific Workflows

The integration of Large Language Models (LLMs) into scientific research promises to accelerate discovery, yet a significant gap remains between the dynamic reasoning of Artificial Intelligence (AI) agents and the static, deterministic nature of canonical scientific workflows. This paper introduces ADEPT (Agentic Discovery and Exploration Platform for Tools), a reference architecture and pedagogical framework explicitly designed to bridge this gap. ADEPT's primary mission is to provide a transparent, "glass-box" environment where researchers and engineers can learn to effectively wrap established scientific software (e.g., BLAST, Nextflow pipelines) and compose it into reliable, agent-driven workflows. We describe its modular, multi-server architecture, which leverages the Model Context Protocol (MCP) for tool serving, LangGraph for robust agentic orchestration, and a secure nsjail-based sandbox for safe code execution. By prioritizing architectural clarity, safety, and modularity, ADEPT serves as an extensible blueprint for building trustworthy AI-augmented systems and fosters the collaborative development necessary to responsibly employ agentic AI for science. We provide practical examples of how to adapt and extend this framework, highlighting its utility in workforce development and AI-readiness capabilities across research and development projects.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Data-Driven Multi-agent Deep Reinforcement Learning for Distribution System Decentralized Voltage Control with High Penetration of PVs

This paper proposes a novel model-free/data-driven centralized training and decentralized execution multi-agent deep reinforcement learning (MADRL) framework for distribution system voltage control with high penetration of PVs. The proposed MADRL can coordinate both the real and reactive power control of PVs with existing static var compensators and battery storage systems. Unlike the existing DRL-based voltage control methods, our proposed method does not rely on a system model during both the training and execution stages. This is achieved by developing a new interaction scheme between the surrogate modeling of the original system and the multi-agent soft actor critic (MASAC) MADRL algorithm. In particular, the sparse pseudo-Gaussian process with a few-shots of measurements is utilized to construct the surrogate model of the original environment, i.e., power flow model. This is a data-driven process and no model parameters are needed. Furthermore, the MASAC enabled MADRL allows to achieve better scalability by dividing the original system into different voltage control regions with the aid of real and reactive power sensitivities to voltage, where each region is treated as an agent. This also serves as the foundation for the centralized training and decentralized execution, thus significantly reducing the communication requirements as only local measurements are required for control. Comparative results with other alternatives on the IEEE 123-nodes and 342-nodes systems demonstrate the superiority of the proposed method.

14 SOLAR ENERGY↗
DOE OSTI2020

Decentralised demand response market model based on reinforcement learning

A new decentralised demand response (DR) model relying on bi‐directional communications is developed in this study. In this model, each user is considered as an agent that submits its bids according to the consumption urgency and a set of parameters defined by a reinforcement learning algorithm called Q‐learning. The bids are sent to a local DR market, which is responsible for communicating all bids to the wholesale market and the system operator (SO), reporting to the customers after determining the local DR market clearing price. From local markets’ viewpoint, the goal is to maximise social welfare. Four DR levels are considered to evaluate the effect of different DR portions in the cost of the electricity purchase. The outcomes are compared with the ones achieved from a centralised approach (aggregation‐based model) as well as an uncontrolled method. Numerical studies prove that the proposed decentralised model remarkably drops the electricity cost compare to the uncontrolled method, being nearly as optimal as a centralised approach.

Shafie‐Khah, Miadreza↗