GridLearn: Multiagent Reinforcement Learning for Grid-Aware Building Energy Management

  title={GridLearn: Multiagent Reinforcement Learning for Grid-Aware Building Energy Management},
  author={Aisling Pigott and Constance Crozier and Kyri Baker and Zolt{\'a}n Nagy},

Figures and Tables from this paper

MERLIN: Multi-agent offline and transfer learning for occupant-centric energy flexible operation of grid-interactive communities using smart meter data and CityLearn

The decarbonization of buildings presents new challenges for the reliability of the electrical grid as a result of the intermittency of renewable energy sources and increase in grid load brought

Systematic Review on Deep Reinforcement Learning-Based Energy Management for Different Building Types

Owing to the high energy demand of buildings, which accounted for 36% of the global share in 2020, they are one of the core targets for energy-efficiency research and regulations. Hence, coupled with

PowerGridworld: a framework for multi-agent reinforcement learning in power systems

We present the PowerGridworld open source software package to provide users with a lightweight, modular, and customizable framework for creating power-systems-focused, multi-agent Gym environments

Multi-Agent Reinforcement Learning for Fast-Timescale Demand Response of Residential Loads

To integrate high amounts of renewable energy resources, electrical power grids must be able to cope with high amplitude, fast timescale variations in power generation. Frequency regulation through

Beobench: a toolkit for unified access to building simulations for reinforcement learning

Beobench is a Python toolkit that provides unified access to building simulations from multiple frameworks using a container-based approach and demonstrates the power of the approach with an example showing how Beobench can launch RL experiments in any supported framework with a single command.

Ad Hoc Teamwork in the Presence of Adversaries

The purpose of extending ad hoc teamwork to include the presence of adversaries and why this problem is important are explained and some directions are proposed for new research opportunities that leads to more robust multi-agent cyber-physical infrastructure systems.

PowerGym: A Reinforcement Learning Environment for Volt-Var Control in Power Distribution Systems

This work examines state-of-the-art reinforcement learning algorithms in PowerGym, an open-source reinforcement learning environment for Volt-Var control in power distribution systems, and validate the environment by studying controller behaviors.



CityLearn: Standardizing Research in Multi-Agent Reinforcement Learning for Demand Response and Urban Energy Management

CityLearn, an OpenAI Gym Environment which allows researchers to implement, share, replicate, and compare their implementations of RL for demand response, and The CityLearn Challenge, a RL competition to propell further progress in this field are discussed.

MARLISA: Multi-Agent Reinforcement Learning with Iterative Sequential Action Selection for Load Shaping of Grid-Interactive Connected Buildings

MarlISA is introduced, a controller that combines multi-agent RL with the proposed iterative sequential action selection algorithm for load shaping in urban energy systems and achieves superior results over multiple independent/uncooperative RL agents using the same reward function.

Deep Reinforcement Learning Based Volt-VAR Optimization in Smart Distribution Systems

This paper develops a model-free volt-VAR optimization (VVO) algorithm via multi-agent deep reinforcement learning (DRL) in unbalanced distribution systems and realizes the dual goals for VVO.

MARLA-SG: Multi-Agent Reinforcement Learning Algorithm for Efficient Demand Response in Smart Grid

This article presents the development and evaluation of a Multi-Agent Reinforcement Learning Algorithm for efficient demand response in Smart Grid (MARLA-SG) and shows a simple and flexible way of choosing state elements to reduce the possible number of states, regardless of the device type, range of operation, and maximum allowable delay.

A Multi-Agent Deep Reinforcement Learning Based Voltage Regulation Using Coordinated PV Inverters

Comparative results with other methods show the enhanced control capability of the proposed method under various conditions.

Integration of smart grid technologies for voltage regulation in low voltage distribution grids

Integration of distributed generation (DG) units, energy storage devices and new loads into the distribution grids introduces new challenges such as voltage rise problems and overloading of grid

Network-Cognizant Voltage Droop Control for Distribution Grids

Examination of distribution systems that have a high integration level of distributed energy resources (DERs) and the design of local control methods for real-time voltage regulation focuses on proportional control strategies wherein the active and reactive power output of DERs are adjusted in response to (and proportionally to) local changes in voltage levels.

CityLearn v1.0: An OpenAI Gym Environment for Demand Response with Deep Reinforcement Learning

A new framework, CityLearn, based on the OpenAI Gym Environment, is introduced, which will allow researchers to implement, share, replicate, and compare their implementations of reinforcement learning for demand response applications more easily.

Decentralised‐distributed hybrid voltage regulation of power distribution networks based on power inverters

The simulation results on IEEE 33-bus distribution network with real-world data have validated the effectiveness of the proposed voltage regulation method and a decentralised voltage control is designed to regulate voltage ramp-rate for mitigating voltage fluctuations.

Centralized control of step voltage regulators and energy storage system under high photovoltaic penetration

The penetration level of distributed energy resources (DERs) especially photovoltaic (PV) system is envisioned to keep on increasing in future distribution grid. Current step voltage regulators