CityLearn
This output is listed in the Digital Public Goods Alliance Registry and contributes to the following UN Sustainable Development Goals (SDGs):
CityLearn is an open source Farama Foundation Gymnasium environment for the implementation of Multi-Agent Reinforcement Learning (RL) for building energy coordination and demand response in cities [24, 26]. A major challenge for RL in demand response is the ability to compare algorithm performance [28]. Thus, CityLearn facilitates and standardizes the evaluation of RL agents such that different algorithms can be easily compared with each other.
Districts and cities have periods of high demand for electricity, which raise electricity prices and the overall cost of the power distribution networks. Flattening, smoothening, and reducing the overall curve of electrical demand helps reduce operational and capital costs of electricity generation, transmission, and distribution networks. CityLearn allows the easy implementation of reinforcement learning agents in a single or multi-agent setting to reshape their aggregated curve of electrical demand by controlling active energy storage for load shifting and heat pump or electric heater power for load shedding.
Applications
CityLearn has been utilized in the following projects and publications:
Application |
Examples |
---|---|
Benchmarking reinforcement learning algorithms |
Dhamankar et al. [4], Khayatian et al. [7], Nweye et al. [11], Pinto et al. [18], Qin et al. [20] |
Coordinated energy management |
Deltetto [2], Glatt et al. [5], Kathirgamanathan et al. [6], Pinto et al. [17], Pinto et al. [19], Qin et al. [21], Vazquez-Canteli et al. [27], Yang et al. [29] |
Incentive based demand-response |
|
Independent energy management |
|
Meta-learning |
Zhang et al. [30] |
Model predictive control |
Zhan, Sicheng et al. [31] |
Transfer learning |
Nweye et al. [13] |
Voltage regulation |
Pigott et al. [16] |
Cite CityLearn
@inproceedings{10.1145/3360322.3360998,
author = {V\'{a}zquez-Canteli, Jos\'{e} R. and K\"{a}mpf, J\'{e}r\^{o}me and Henze, Gregor and Nagy, Zoltan},
title = {CityLearn v1.0: An OpenAI Gym Environment for Demand Response with Deep Reinforcement Learning},
year = {2019},
isbn = {9781450370059},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
url = {https://doi.org/10.1145/3360322.3360998},
doi = {10.1145/3360322.3360998},
booktitle = {Proceedings of the 6th ACM International Conference on Systems for Energy-Efficient Buildings, Cities, and Transportation},
pages = {356–357},
numpages = {2},
keywords = {Smart Grid, Building Energy Control, Smart Buildings},
location = {New York, NY, USA},
series = {BuildSys '19}
}