A Centralised Soft Actor Critic Deep Reinforcement Learning Approach to
District Demand Side Management through CityLearn
- URL: http://arxiv.org/abs/2009.10562v1
- Date: Tue, 22 Sep 2020 14:03:11 GMT
- Title: A Centralised Soft Actor Critic Deep Reinforcement Learning Approach to
District Demand Side Management through CityLearn
- Authors: Anjukan Kathirgamanathan, Kacper Twardowski, Eleni Mangina, Donal Finn
- Abstract summary: Reinforcement learning is a promising model-free and adaptive controller for demand side management.
This paper presents the results of the algorithm that was submitted for the CityLearn Challenge.
The proposed solution secured second place in the challenge using a centralised 'Soft Actor Critic' deep reinforcement learning agent.
- Score: 0.0
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Reinforcement learning is a promising model-free and adaptive controller for
demand side management, as part of the future smart grid, at the district
level. This paper presents the results of the algorithm that was submitted for
the CityLearn Challenge, which was hosted in early 2020 with the aim of
designing and tuning a reinforcement learning agent to flatten and smooth the
aggregated curve of electrical demand of a district of diverse buildings. The
proposed solution secured second place in the challenge using a centralised
'Soft Actor Critic' deep reinforcement learning agent that was able to handle
continuous action spaces. The controller was able to achieve an averaged score
of 0.967 on the challenge dataset comprising of different buildings and
climates. This highlights the potential application of deep reinforcement
learning as a plug-and-play style controller, that is capable of handling
different climates and a heterogenous building stock, for district demand side
management of buildings.
Related papers
- A Distributed ADMM-based Deep Learning Approach for Thermal Control in Multi-Zone Buildings under Demand Response Events [1.1126342180866646]
This research combines distributed optimization using ADMM with deep learning models to plan indoor temperature setpoints effectively.
A two-layer hierarchical structure is used, with a central building coordinator at the upper layer and local controllers at the thermal zone layer.
The proposed algorithm, called Distributed Planning Networks, is designed to be both adaptable and scalable to many types of buildings.
arXiv Detail & Related papers (2023-12-08T14:46:50Z) - Specialized Deep Residual Policy Safe Reinforcement Learning-Based
Controller for Complex and Continuous State-Action Spaces [1.3678669691302048]
It is impractical to explore randomly, and replacing conventional controllers with black-box models is also undesirable.
We propose a specialized deep residual policy safe reinforcement learning with a cycle of learning approach adapted for complex and continuous state-action spaces.
arXiv Detail & Related papers (2023-10-15T21:53:23Z) - MERLIN: Multi-agent offline and transfer learning for occupant-centric
energy flexible operation of grid-interactive communities using smart meter
data and CityLearn [0.0]
Decarbonization of buildings presents new challenges for the reliability of the electrical grid.
We propose the MERLIN framework and use a digital twin of a real-world grid-interactive residential community in CityLearn.
We show that independent RL-controllers for batteries improve building and district level compared to a reference by tailoring their policies to individual buildings.
arXiv Detail & Related papers (2022-12-31T21:37:14Z) - Human-instructed Deep Hierarchical Generative Learning for Automated
Urban Planning [57.91323079939641]
We develop a novel human-instructed deep hierarchical generative model to generate optimal urban plans.
The first stage is to label the grids of a target area with latent functionalities to discover functional zones.
The second stage is to perceive the planning requirements to form urban functionality projections.
The third stage is to leverage multi-attentions to model the zone-zone peer dependencies of the functionality projections to generate grid-level land-use configurations.
arXiv Detail & Related papers (2022-12-01T23:06:41Z) - Learning to Walk Autonomously via Reset-Free Quality-Diversity [73.08073762433376]
Quality-Diversity algorithms can discover large and complex behavioural repertoires consisting of both diverse and high-performing skills.
Existing QD algorithms need large numbers of evaluations as well as episodic resets, which require manual human supervision and interventions.
This paper proposes Reset-Free Quality-Diversity optimization (RF-QD) as a step towards autonomous learning for robotics in open-ended environments.
arXiv Detail & Related papers (2022-04-07T14:07:51Z) - Curriculum Based Reinforcement Learning of Grid Topology Controllers to
Prevent Thermal Cascading [0.19116784879310028]
This paper describes how domain knowledge of power system operators can be integrated into reinforcement learning frameworks.
A curriculum-based approach with reward tuning is incorporated into the training procedure by modifying the environment.
A parallel training approach on multiple scenarios is employed to avoid biasing the agent to a few scenarios and make it robust to the natural variability in grid operations.
arXiv Detail & Related papers (2021-12-18T20:32:05Z) - Real-world challenges for reinforcement learning in building control [0.0]
We propose a non-exhaustive nine real world challenges for reinforcement learning building controller.
We argue that building control research should be expressed in this framework.
arXiv Detail & Related papers (2021-11-25T02:59:53Z) - Ada-Segment: Automated Multi-loss Adaptation for Panoptic Segmentation [95.31590177308482]
We propose an automated multi-loss adaptation (named Ada-Segment) to flexibly adjust multiple training losses over the course of training.
With an end-to-end architecture, Ada-Segment generalizes to different datasets without the need of re-tuning hyper parameters.
Ada-Segment brings 2.7% panoptic quality (PQ) improvement on COCO val split from the vanilla baseline, achieving the state-of-the-art 48.5% PQ on COCO test-dev split and 32.9% PQ on ADE20K dataset.
arXiv Detail & Related papers (2020-12-07T11:43:10Z) - Demand-Side Scheduling Based on Multi-Agent Deep Actor-Critic Learning
for Smart Grids [56.35173057183362]
We consider the problem of demand-side energy management, where each household is equipped with a smart meter that is able to schedule home appliances online.
The goal is to minimize the overall cost under a real-time pricing scheme.
We propose the formulation of a smart grid environment as a Markov game.
arXiv Detail & Related papers (2020-05-05T07:32:40Z) - Data-driven control of micro-climate in buildings: an event-triggered
reinforcement learning approach [56.22460188003505]
We formulate the micro-climate control problem based on semi-Markov decision processes.
We propose two learning algorithms for event-triggered control of micro-climate in buildings.
We show the efficacy of our proposed approach via designing a smart learning thermostat.
arXiv Detail & Related papers (2020-01-28T18:20:43Z) - NeurOpt: Neural network based optimization for building energy
management and climate control [58.06411999767069]
We propose a data-driven control algorithm based on neural networks to reduce this cost of model identification.
We validate our learning and control algorithms on a two-story building with ten independently controlled zones, located in Italy.
arXiv Detail & Related papers (2020-01-22T00:51:03Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.