DR Potential prediction task:
DR_Potential_Prediction_CNN.py : Demand response potential prediction CNN model
handle_weather_data_season.py : Download and pre-process weather data based on each respective county data and stores it in weather_data_total_seasons.xlsx.
upgrade01_metadata_and_annual_results.parquet : Metadata and annual energy consumption data from Comstock simulation output (all states) (Download from https://data.openei.org/s3\_viewer?bucket=oedi-data-lake\&prefix=nrel-pds-building-stock%2Fend-use-load-profiles-for-us-building-stock%2F2024%2Fcomstock\_amy2018\_release\_1%2Fmetadata\_and\_annual\_results%2Fnational%2Fparquet%2F)
Weather_data_merger.py : Merging the excel weather data file with upgrade01_metadata_and_annual_results.parquet based on county code and preprocess the data to eliminate pollution
selected_features.xlsx : Selected features name
selected_output.xlsx : Selected output name
HVAC load Dissagregation & prediction task:
Weather_data_merger.py : Merging the excel weather data file with upgrade01_metadata_and_annual_results.parquet based on county code and preprocess the data to eliminate pollution
HVAC_Dissagregation_CNN_model.py: 2D CNN model for HVAC Dissagregation (using buildings total energy consumption and temperature as features)
HVAC_Dissagregation.py: Preparing data for 2D CNN model for HVAC Dissagregation
HVAC_Prediction_Download.py : Download data for HVAC_Prediction_Data_Preparing.py (Comstock New York State individual buildings data)
HVAC_Prediction_Data_Preparing.py : Preprocess data for data for HVAC_Prediction_Model.py
HVAC_Prediction_Model.py : multi dimensional buildings HVAC prediction model (using metadata and climate/weather data as features)
selected_features.xlsx : Selected features name
selected_output.xlsx : Selected output name
Note:
Change all the file data directories to suit your saved folder
Metadata: Constant buildings characteristic data (area, built year, number of stories, etc.)