Loading
Phase 1: Warm Up: Completed Phase 2: Completed Weight: 1.0
8678
665
105
785

πŸ“• Make your first submission with ease by following this forecasting-track-starter-kit.

πŸ“² Join the conversation with other participants on CityLearn Discord Server.

πŸ‘₯ Challenge is more fun with friends, find your teammates.

πŸ’¬ Got a query or suggest? Share your feedback.

πŸ•΅οΈ Introduction

The CityLearn Challenge 2023 addresses this multi-faceted nature of advanced control in buildings by blending the challenges of control algorithm design, forecast quality and grid-resilience. The CityLearn Challenge 2023 presents a control track as done in previous challenges as well as introduces an independent forecast track where, both tracks are run in parallel and utilize the same dataset.

In this forecast track, participants will design regression models to predict the 48-hour-ahead end-use load profiles for each building in a synthetic single-family neighborhood as well as the neighborhood-level 48-hour-ahead solar generation and carbon intensity profiles.

πŸ“‘ Problem Statement

In the forecast track, participants are to develop regression prediction models for forecasting building loads, grid carbon intensity and solar generation. The forecasts should be provided at each time step for the following 48 time steps. The variables to be predicted are: 

Building-Level

  1. Cooling Load (kWh)
  2. DHW Load (kWh)
  3. Equipment Electric Power (kWh) (a.k.a. non-shiftable load)

Where each variable is forecasted for each building.

Neighborhood-Level

  1. Carbon Intensity (kgCO2e/kWh)
  2. Solar Generation (W/kW)

Where each variable is forecasted for the entire neighborhood.

πŸ’Ύ Dataset

Please find a detailed breakdown of the dataset over here
Refer to the dataset and schema files.

πŸš€ Submission

Make your first submission using the starter kitπŸš€!

πŸ–Š Evaluation Criteria

The forecast track score, ScoreForecast, is the average over all of the variables being forecast, of the normalised mean root mean square error (RMSE) of the forecasts made.

Where:

  • t: Environment time step index;
  • n: Total number of time steps, t, in 1 episode;
  • Ο„: Forecasting window time step index;
  • w: Length of forecasting window (48hrs);
  • b: Total number of buildings;
  • v: Forecasting variable;
  • V: Total number of variables to forecast (3b + 2),
  • ft,Ο„v: Forecast of variable v for time step t+Ο„, made at time t;

πŸ“… Timeline

See the detailed breakdown of all challenge phases over here.

πŸ† Prizes

This challenge has Leaderboard Prizes and Co-authorship Prizes

Leaderboard Prizes

Top three teams or participants on the private Phase II control track leaderboard will receive the following prizes:

  • πŸ₯‡ 1st Prize: 1000 USD
  • πŸ₯ˆ 2nd Prize: 800 USD
  • πŸ₯‰ 3rd Prize: 700 USD

Co-Authorship

In addition to the cash prizes, we will invite the top three teams or participants from both tracks to co-author a summary manuscript at the end of the competition. At our discretion, we may also include honorable mentions for academically interesting approaches, such as those using exceptionally little computing or minimal domain knowledge. Honorable mentions will be invited to contribute a shorter section to the paper and have their names included inline.

πŸ“•Further Reading

The Energy Efficient Cities initiative [EECi] is a cross-disciplinary research project at the University of Cambridge. The EECi aims to strengthen the UK’s capacity to address energy demand reduction and environmental impact in cities, by research in building and transport technologies, district power systems, and urban planning.

Read their work on Smart Design and Control of Energy Storage Systems. It investigates the present situation of smart design and control strategy of energy storage systems for both demand side and supply side. Specifically, artificial intelligence that has developed significantly in recent years can be expected to make a significant contribution to the smart design and control systems. 

πŸ“± Contact

πŸ“² Join the conversation with other participants on CityLearn Discord Server.
πŸ‘₯ Challenge is more fun with friends, find your teammates.
πŸ’¬ Got a query or suggest? Share your feedback.