GitLab

# Conflicts:
#	inc/1a_definitions.gms
#	inc/2b_equations.gms
#	inc/3b_inputsLoop.gms
#	inc/3d_setVariableLimits.gms
#	inc/4b_outputInvariant.gms

Replaced q_fixedGenCap2U with q_constrainedCapMultiUnit, which enables different decommissioning and replacement decisions in addition to fixed investment ratios. Removed unnecessary parameters and divided p_groupPolicy into p_groupPolicy and p_groupPolicy3D. Renamed also the linking sets related to groups, e.g. gnu_group --> gnuGroup.

Added fixed operation and maintenance costs for investment options
Adding a new parameter unitOutputCapacityTotal to avoid division by zero
Added unit investment result parameter
Removed equation q_fixedGenCap1U
Added default value calculation for reserves update frequency
Moved tCounter update outside ms loop in 3c_periodicLoop.gms
Updated the limits of v_invest_LP and v_invest_MIP (no more maxGenCap, minGenCap)
Updated investInit_temp.gms
And some smaller updates

Fixing a major bug where the interval counter is updated after each sample loop, messing up the sft-structure.

Results vectors now record dummy variables in all the forecasts, to make sure that no forecasts cause the model to behave in undesired ways.

"reserveContribution" parameter now accounted for in the state variable upwards/downwards equations.

Hopefully finally fixing the erratic behaviour of the state variable reserves in the RealValue results runs.

Continuing to fix the reserves in the deterministic mode. Now they should also work after the first solve.

# Conflicts:
#	README.md

Slight performance improvement to the "d_nodeState" calculation using the new "gn_stateTimeseries" set.

Removing the possibility of using tertiary reserves as backup, permits too large reserve provisions by single units. Now reserve commitments only lock the lower bound of the reserve provision variable.

Fixing the node state forecasts to work as intended, and added a new set for time series bound gn states for minor performance improvements.

Time aggregated reserve demand time series now uses the maximum value within the interval, instead of the averaged reserve demand.

Fixing a major bug where none of the reserve demand was forced for the realized time steps, instead of only the tertiary reserve demand vanishing.

The reserves have to be revised later on to make them more customizable, along with forecasts i suppose.

Release of tertiary reserves for the realization released also primary and secondary reserves. Now corrected, but the selection of reserves should be moved to the input file along with the new set members for reserveTypeReleasedForRealization.

Moved the "holdfixed" commands from the "3e_solve.gms" to the end of the model definition files, since as model properties that is a more suitable place for them.

Added a possibility for input dependent O&M costs, required by electric heating (heat pump!) taxation. Also added a new "fuel_penalty" multiplier as a possible command line input, required for making building stock baseline results reliable for electrical-fuel heating system mixes.

Deterministic model (mSettings('...', 'forecasts') = 0) is corrected to have a model horizon up-to t_horizon (used to be only up-to t_jump).