Optimisation

The optimisation module allows for an optimisation of the energy system using a cost function with can consist of monetary or carbon cost or a combination of the two quantities.

The optimisation relies on using a LP or MILP optimizer accessed via the energy system optimization framework Calliope (Pfenninger & Pickering, 2018). Via Calliope, a solver, such as Gurobi, is accessed.

The optimisation happens using the methods built into the optimizer. The more technologies are present in the energy system, the longer the optimisation takes. Activating integer variables (i.e. the asynchronous storage) leads to significantly longer optimisation times.

In addition to the conventional optimisation mode, a pareto front mode is also available. In this mode, a monetary-co2 pareto front is constructed.

The parameters of the optimisation are

Attribute	Description	Standard value	Unit	Data type
enabled	MILP optimisation is enabled if set to “True”.	False	—	bool
clustering	*Deprecated: This option is not available.*	False	—	bool
pareto_monetary_co2	Set to True to calculate a pareto front	False	—	bool
N_pareto	Number of pareto points	6	—	int
objective_monetary	Weight of monetary objective for objective function in optimisation.	1	—	float
objective_co2	Weight of emissions objective for objective function in optimisation.	0	—	float
objective_ess	Not yet implemented.	0	—	float
objective_tss	Not yet implemented.	0	—	float
bigM_value	Default: 1e9; cost of unmet demand; large value makes model convergence slow; https://calliope.readthedocs.io/en /stable/user/building.html#allowing-for- unmet-demand	100000	CHF/kWh	float
solver	Solver for optimisation, e.g. ‘cbc’, ‘gurobi’, etc. Must be compatible with Calliope.	‘gurobi’	—	string
solver_option_NumericFocus	Default: 0; https://docs.gurobi.com/proj ects/optimizer/en/current/reference/para meters.html#parameternumericfocus	1	—	int
solver_option_TimeLimit	Default: ‘Infinity’; https://docs.gurobi .com/projects/optimizer/en/current/refer ence/parameters.html#timelimit	36000	s	int
solver_option_Presolve	Default: -1; https://docs.gurobi.com/pro jects/optimizer/en/current/reference/par ameters.html#presolve	-1	—	int
solver_option_Aggregate	Default: 1; https://docs.gurobi.com/proj ects/optimizer/en/current/reference/para meters.html#aggregate	1	—	int
solver_option_FeasibilityTol	Default: 1e-6; https://docs.gurobi.com/p rojects/optimizer/en/current/reference/p arameters.html#feasibilitytol	0.01	var	float
solver_option_MIPGap	Default: 1e-4; https://docs.gurobi.com/p rojects/optimizer/en/current/reference/p arameters.html#mipgap	0.0001	—	float
MIPGap_increase	If set to True, MIPGap will be increased to 0.01 if a storage technology is activated in order to avoid numerical problems.	False	—	bool
save_math_model	If set to ‘true’, the math. model formulations are written to an .lp file; can take long to produce and result in large file;	False	—	bool
save_calliope_files	If set to ‘True’, the calliope files (input and results) will be saved.	False	—	bool

References

Pfenninger, S., & Pickering, B. (2018). Calliope: a multi-scale energy systems modelling framework. Journal of Open Source Software, 3(29), 825. https://doi.org/10.21105/joss.00825