The code in this repertory uses the mesa python library for the formulation of an agent-based model of community solar adoption.
TO-DO: (see file "updates_before_submission")
Terms: - experiment = simulation and scenarios inputs in JSON file. - scenario = unique combination of simulation, calibration, and economic parameters. One experiment can contain one or several scenarios. - batch = set of simulation runs computed for one scenario.. - run = deterministic simulation of the model for one scenario. - timestep = one year in the simulation model.
Important: results are reported at experiment level, after computing one batch for each scenario in the experiment.
main.py -> principal program: initializes the data required for running the COSA-ABM, reads the inputs for the simulations to be carried, and creates an instantiation of the model that then simulates and collects data from.
COSA_Model -> contains the code for the SolarAdoptionModel object class
COSA_Agent -> contains the code for the BuildingAgent object class
COSA_Tools -> contains auxiliary code for economic evaluation of installation of solar PV, data collection, a modified scheduler that allows a coherent randomisation control, and a creator of social networks
COSA_Data -> contains the input data required for the simulations **Note: large input files are ignored and not uploaded to the repo
COSA_Outputs -> stores the simulation outputs (all content ignored because of the large size of the files)