Amazon sponsors contest on energy management in buildings


Buildings are responsible for 30% of greenhouse gas emissions. At the same time, buildings are beginning to take a more active role in the power system by providing benefits to the electrical grid.

As such, buildings are an unexplored opportunity to help address climate change. Energy storage devices such as home batteries can reduce the energy grid’s peak loads by shifting buildings’ energy use to different times. Solar photovoltaic generation can reduce the overall demand on the grid while also reducing emissions. However, all these resources must be carefully managed simultaneously in many buildings to unlock their full energy potential and reduce homeowners’ costs.

Related content

How data-driven methods can help to identify fault detection and drive energy efficiencies for facilities of all sizes.

The CityLearn Challenge 2022, organized by the Intelligent Environments Lab and AICrowd, with computational resources provided by Amazon, focuses on the opportunity presented by home battery storage and photovoltaics (PVs). Participants will develop reinforcement-learning agents that control battery charge and discharge in buildings and reward functions that minimize both the monetary cost of electricity drawn from the grid and CO2 emissions. Winners will be announced at NeurIPS 2022.

In the challenge’s multi-agent scenario, agents may share data in order to jointly optimize energy management.

The challenge leverages CityLearn, a development environment for distributed energy resource management and demand response. The challenge uses one year of operational electricity demand and PV generation data from 17 single-family buildings in the Sierra Crest home development in Fontana, California, which the nonprofit research organization EPRI studied in its paper for the American Council for an Energy-Efficient Economy “Grid integration of zero net energy communities”.

Challenge participants may develop either single-agent or multi-agent policies and reward functions. A single-agent setup will mean that one policy is used to control all building batteries, while a multi-agent setup will mean that each building’s battery is controlled using a unique policy. However, in the multi-agent scenario, agents are allowed to share information about their observations.

To ensure that occupant comfort is guaranteed at all times, the electric load of the building will not change. However, the agents must learn when to use electricity directly from on-site solar panels, when to charge/discharge batteries, and when to rely on the grid. If they rely on the grid, they should learn to use it when it’s cheap and/or provides low-carbon power.

The competition started on July 18 and will conclude on October 31. There are cash prizes for the top three teams, student travel grants, and an opportunity to co-author a research paper. NeurIPS will be held in New Orleans from November 28 to December 9; the competition track will be virtual.

Source link


Please enter your comment!
Please enter your name here

Share post:


More like this

Apple marks World AIDS Day with (RED)

December 1, 2023 UPDATE Apple marks World AIDS...

AI’s Impact On Venture Capital And 5 Tips For Success

AI stands as the bedrock for innovative products...

WhatsApp now lets you use secret codes to lock your private chats

WhatsApp is making its Chat Lock tool even...

TikTok: US judge blocks Montana’s ban citing free speech

Chinese-owned app welcomes the ruling against "unconstitutional law"...