initial commit for end use load profiles neighborhoods
refactoring and bug fixes
added example workflow to build neighborhood
refactored
updated docs for lst training
added return type
added doe_xstock to requirements
deleted doe_xstock installation from git
added pricing to shared observations
moved datasets to dataset directory
first draft for model generation
added SAM to requirements to use in simulation pv generation
added tracking the sun dataset
added numpy random state property to prevent numpy seed changing in parallel runs
added version to CLI arguments
added real world battery model parameters
added functions to read tracking the sun dataset and battery models
updated device parameters e.g. efficiency to be sampled from a uniform distribution by default where the distribution limits can be user defined. Also updated pv and battery autosize functions to use real world data and to be set based on ZNE target and estimated electricity consumption. These changes are not tested at this point.
updated data file column names to snakecase and to match variable names in Data classes they are loaded into. Also updated floats to float32 and int to int32
fixed pricing column name error
defined directories and filepaths as constants at the module level
set yaml and json indentation level to 2 spaces
moved setting function to data module
Set all autosize safety_factor to 1.0 by default and set random seed prior to setting other variables
changed indentation to 2 spaces
renamed simulator class
Removed distinction between cooling and heating dynamics models to have just one dynamics model. Added comfort band time series to energy_simulation to be used in reward and evaluation. Added a cooling_or_heating_device_action to use for cooling and heating device control.
Set to use comfort band defined in building file by default. Will load files with the assumption that column names match parameter names in the data class init function (help for flexible ordering of columns and adding new columns for a specific dataset that has unique columns). Lastly, building devices are initialized with unique random seed based on building name, type, device name and type to have unique values sampled for their properties when not explicitly defined.
comfort band set to be defined as float or time series
added action mapping for cooling_or_heating_device
added seed key, cooling_or_heating_device action, and single dynamics variable for a building.
refactored comfort band reward. Not yet fully developed for case where there is both cooling and heating during simulation.
updates
updated lstm columns to match observation names
added preliminary schema template
bs2023 dataset initial commit
fixed minor naming issues
deleted config.py
bug fixes in autosize funcs
refactored
refactoring and bug fixes
updated neighborhoods datasets
updated
refactored and set to try 3 times sampling PV data to simulate in PySAM before failing if there are simulation errors.
added hvac_mode and comfort_band obs
modularized load function
fixed battery capacity error
fixed carbon intensity and pricing init bug when simulation time steps are not same as data length
fixed missing parameter bug
refactored
commented out lstm model generation and schema build code
fixed column name error
added citylearn weather and energy simulation queries
updated baseline agent to return empty action list
updated default DoD to 1.0
explicity set box type
added updated environment and control figures from v2 paper
Fixed wrapper variable user warnings
unwrapped env properties that may raise user warning when using env wrappers
added white space
refactored
updated to beta version 2.2b0
Screenshots
NIL
Checklist
[x] I have tested the changes locally and they work as intended.
[x] I have updated the documentation, if applicable.
[ ] I have added new tests, if applicable.
[x] I have added any required dependencies to the requirements.txt file, if applicable.
[x] I have followed the project's code style and conventions.
Description
See title
Issue
NIL
Changes
Screenshots
NIL
Checklist
requirements.txtfile, if applicable.Additional notes
NIL