Outputs

All outputs of a run are written to the folder set in results.directory (default results/).

Simulation_Results.xlsx

One row per scenario and formulation; repeated runs append rows (the reproduction tools keep the last row of each scenario and formulation).

Column	Meaning
Scenario, Model	Scenario label and formulation directory (Model_Cluster = Novo)
Number of constraints, Number of variables, Non-zero elements in the matrix	LP size from the GLPK solution file
Total cost	Objective value, model cost units
Matrix and Running Time	Wall time of model generation and solve, s (older files: Matriz and Running Time)
Running time	Solver time reported by GLPK, s
Memory used	Peak memory reported by GLPK, MB
Power Capacity	New VRE capacity
Storage Capacity	New storage energy capacity, kW·yr (total = new + 0.05 residual)
Solver status	optimal or FAILED; capacities are empty for failed solves

State-of-charge files

File	Content
<scenario>_Storage_Level_Model_Niet.csv	SoC at the start of each representative timeslice
<scenario>_Storage_Level_Model_Cluster.csv	Novo: SoC at the start of each chronological timeslice (365 × 24/h values)
<scenario>_Storage_Level_Model_Kotzur.csv	SoC at the start of each chronological day
<scenario>_Storage_Level_Model_Kotzur_intraday.csv	Kotzur trajectory reconstructed within each day (VALUEACCUMULATED)
<scenario>_Storage_Level_Model_Welsch.csv	SoC at the start of each day type
<scenario>_Storage_Level_Model_Welsch_intraday.csv	Welsch trajectory reconstructed in grouped day-type order

Other files

File	Content
<scenario>_clustering.json	k, h, seed, storage cap, representative days and the day-to-representative mapping
<scenario>_Conversionlts_sparse.csv	Chronological → representative timeslice mapping used for the Kotzur reconstruction
<scenario>_Figure*.png	Daily, hourly and two-week SoC comparison (only when all four formulations ran)
logs/<scenario>.log	Full console output of the run (batch and notebook runs)
models/Model_<name>/results/*.csv	All OSeMOSYS result tables of the last run of that formulation (not tracked)

Loading results in Python

import src.reproduce as rp
res = rp.load_results("results/Simulation_Results.xlsx")   # tidy table: resource, k, h, formulation, ...
rp.table4(res)                                              # Table 4 of the paper (needs the k365h1 runs)
rp.add_errors(res)                                          # cost and capacity error vs k365h1 Novo