NYSERDA - Tenant Energy Data

Category Description

 

The Tenant Energy Data category is designed to improve understanding of the drivers of electricity consumption within sub-metered office space in NYC.  The purpose of this effort is to identify new modeling assumptions that account for COVID-19’s impact on occupancy levels.  This category provides real world consumption data from a single large office tenant in Midtown Manhattan along with some building-wide data.

 

Problem Statements 
  1. What is your forecasted consumption across all 18 tenant usage meters for the 24 hours of 8/31/20 in 15 minute intervals (1728 predictions)?

  2. How correlated are building-wide occupancy and tenant consumption?

  3. What is the mean absolute error for your model?

  4. What feature(s)/predictor(s) were most important in determining energy efficiency?

  5. What is the most energy-efficient occupancy level as a percentage of max occupancy provided (i.e., occupancy on 2/10/20)?

  6. What else, if anything, can be concluded from your model?

  7. What other information, if any, would you need to better your model?

 

All participants should answer all seven questions in their submissions.  The forecasts should be in .csv or .xlsx format.  A one to two minute, publicly accessible video presentation overviewing submissions is required from all participants. 

 

Winning submissions will be promoted by NYSERDA, REBNY, and Connected RE Magazine to facilitate participants’ access to the world’s leading energy consumers.  

Datasets

The test data for this challenge includes the following:

  • Occupancy.xlsx

  • Tenant_Usage.xlsx

  • ConEd_Electric.xlsx

  • ConEd_Steam.xlsx

Test data can be downloaded here: https://app.box.com/s/9z5xf9t7tkrt2hhl4qk3a6fvimieq973

 

Tenant Usage

Time series data of demand and consumption of electricity from one office tenant spanning from 1/1/18 12:00 AM - 8/30/20 12:00 AM where readings are taken at the end of 15-minute intervals.

  • meter - unique meter ID

  • date_time - date and time of reading

  • consumption - electricity used since the last measurement (in kWh)

  • max_demand - highest electrical power reading in the last 15 minutes (in kW)

  • min_demand - lowest electrical power reading in the last 15 minutes (in kW)

  • avg_demand - average electrical power reading over the last 15 minutes (in kW)

 

Occupancy

The number of unique entries to the building for a given day.

 

ConEd Electric:

1 - Time series data of building-wide (totalizer) demand and consumption of electricity spanning from 1/1/18 12:00 AM - 9/1/20 12:00 AM where readings are taken at the end of 15-minute intervals.

  • meter - unique meter ID

  • date_time - date and time of reading

  • consumption - electricity used since the last measurement (in kWh)

  • max_demand - highest electrical power reading in the last 15 minutes (in kW)

  • min_demand - lowest electrical power reading in the last 15 minutes (in kW)

  • avg_demand - average electrical power reading over the last 15 minutes (in kW)

2 - Time series data of external temperature and humidity readings taken on the hour from the building’s rooftop weather station.

  • temp - hourly readings of external temperature (in Fahrenheit)

  • humidity - hourly readings of external humidity (in %)

 

ConEd Steam:

1 - Time series data of building-wide demand and consumption of steam spanning from 1/1/18 12:00 AM - 9/1/20 12:00 AM where readings are taken at the end of 15-minute intervals.

  • date_time - date and time of reading

  • consumption - steam used since the last measurement (in lbs)

  • max_demand - highest steam reading in the last 15 minutes (in lbs/hour)

  • min_demand - lowest steam reading in the last 15 minutes (in lbs/hour)

  • avg_demand - average steam reading over the last 15 minutes (in lbs/hour)

 

2 - Time series data of external temperature and humidity readings taken on the hour from the building’s rooftop weather station

  • temp - hourly readings of external temperature (in Fahrenheit)

  • humidity - hourly readings of external humidity (in %)


 

  • Twitter Social Icon
  • LinkedIn
  • YouTube

© 2020 PropTech Challenge