Rework periodic value deserialisation

[nailend]

Some changes are necessary. Predecessor was #124.

Some parameters have to have a specific length due to the implementation of multi-period approach, e.g for every timestep or just for every period. Periodically changing values are indicated in datapackage as list.

The reading routine is checking whether the number of values in the list are equal to the number of periods. If so, they are converted to a timeseries for all timesteps. In some cases this is not necessary and only one value per period is sufficient.

The following parameter conversions have to be checked or maybe implemented:
For sake of clarity, creating a dict to define to which format a parameter is converted might be useful.

sth like this pseudo code

conversion_parameters = {"fixed_costs" : create_periodic_values, ...}
if key in conversion_parameters.keys():
  facade_value =  conversion_parameters[key](value)

Parameters to catch:

• yearly:

  • fixed_costs
  • marginal_costs

• periodically:

  • capacity
  • capacity_cost
  • capacity_potential