Project variables

project_variables.csv: This file is the main option file. The file has three columns, feature, which holds the variable name, value the value assigned, and comment, which gives explanations. In the following, we briefly present all variables and their configuration options.

scenarios_iteration: yes/no

If yes, DIETERpy uses iteration_table.csv to run multiple scenario runs, which have to be configured properly prior to running the model. If no, a single run is performed.

skip_input: yes/no

Allows for skipping the generation of the necessary input gdx files from the input excel files. If yes, the generation is skipped. Be aware, if skipped, the gdx_input folder must already contain the correct gdx files. This can be useful to reduce computation time if you want to run the model again using the same the input data. If no, input excel files are transformed to input gdx files.

skip_iteration_data_file: yes/no

If the iteration_data_file.xlsx sheet has not been changed between runs, you can use yes to reduce computation time and skip the creation of gdx files from excel files. If data have been changed, make sure to use no so that changes in the data are imported to the model.

base_year: e.g. 2030

Choose the year of the time series data. Only relevant if several years are provided in the input data. Verify this in timeseries_input.xlsx.

end_hour: e.g. h336

For testing purposes, the model can run from h1 to end_hour. Default value is h8760 which represents an entire year.

dispatch_only: yes/no

If you select yes, the model will run in dispatch only mode, which means that power plant, storage and transmission capacities are fixed. Verify in the static_input.xlsx sheet that fixed values are provided. To run DIETERpy as an investment and dispatch model, select no.

network_transfer: yes/no

Select yes to allow for electricity flows between nodes. Select no to set cross-nodal electricity flows to zero. If the model is run in the dispatch only mode, net tranfer capacities are fixed according to the column fixed_capacities_ntc on sheet spatial in static_input.xlsx. If investment and dispatch mode is active, net transfer capacities have a static upper and lower bound defined by the column max_installable and min_installable on sheet spatial in static_input.xlsx, respectively.

no_crossover: yes/no

Select yes to switch off crossover of the solver CPLEX. This settings is relevant when CPLEX uses the Barrier Optimizer to solve an LP.

infeasibility: yes/no

Select yes to activate a slack variable in the energy balance of the model representing an unspecified generator. If demand cannot be met in a certain timestep, this generator will catch the missing generation, thus avoiding infeasibility of the optimization problem. Check the variable when debugging. Select no to deactivate this variable.

GUSS: yes/no

Select yes to activate the “Gather-Update-Solve-Scatter” (GUSS) tool. The GUSS tool uses a GAMSCheckpoint as a basis for solving several scenario runs. If several scenario runs are sufficiently similar (only single parameter or variable values varied), using the GUSS tool decrease the overall computation time to solve all scenarios because the model has to compiled only once.

GUSS_parallel: yes/no

If the GUSS is activated (see above), selection yes for this option will solve several scenario runs in parallel. Every scenario run is solved on single CPU thread, yet several scenarios at the same time in parallel. However, be aware that this option demands a high amount of RAM. If insufficient RAM is provided, the optimization can abort.

GUSS_parallel_threads: choose an integer, e.g. 4

This option defines the number of threads used to solve scenario runs in parallel. If 0 is chosen and GUSS parallel is yes, then all available CPU threads are used. To avoid running out of RAM, you can either choose to reduce the number of threads used (hence smaller number) or deactivate the options GUSS parallel altogether.

static_input_file: filename e.g. static_input.xlsx

Name of the file (in the folder data_input) that contains the time-invariant data. If empty, the import of data is skipped.

timeseries_input_file: filename e.g. timeseries_input.xlsx

Name of the file (in the folder data_input) that contains the time-varying data. If empty, the import of data is skipped.

iteration_data_file: filename e.g. iteration_data.xlsx

Defines the file name of the file that contains the data for scenario iteration (if data will be varied in different scenario runs). If empty, the import of data is skipped.

gdx_convert_parallel_threads: choose an integer, e.g. 4

Defines the number of CPU threads used to convert the output GDX files to other files. If 0 is chosen, the maximum number of CPU threads will be used.

gdx_convert_to_csv: yes/no

Select yes to convert the GDX output files to CSV files. For every symbol (variables, parameter, equation), a separate CSV file will be created. The files are saved in folder named CSV within the output folder of each scenario run.

gdx_convert_to_pickle: yes/no

Select yes to convert the GDX output files to PICKLE files. For every scenario run, a separate PICKLE file is created that stores all symbols (variables, parameter, equation) and their values. Important: these PICKLE files are required to created the reporting files.

gdx_convert_to_vaex: yes/no

Select yes to convert the GDX output files to HDF5 files. For every scenario run, a separate HDF5 file is created that stores all symbols (variables, parameter, equation) and their values. HDF5 files are large but can be used out of RAM. IMPORTANT: the vaex package has to be installed before and separately for a successful conversion of files.

report_data: yes/no

Select yes to create report files that contain the same symbols of all scenario runs. These files are saved in report folder and are used for the web interface to plot the results. To choose the symbols to be reported, you have to edit the file reporting_symbols.csv.