Feature/scenarios dims order

[FBumann]

No description provided.

[FBumann]

Change the order of dimensions

• time is always the first dimension.
  This seems more intuitive and brings more consistency

Furthermore, converting a solution to a Dataframe with array.DataArray.to_pandas()results in:

• a pd.Series if no Scenario is Present
• a pd.DataFrame with the scenario as the columns.

In[4]: 
calculation.results.solution['CHP(Q_th)|flow_rate'].to_pandas()
Out[4]: 
scenario             Base Case  High Demand
time                                       
2020-01-01 00:00:00  25.000000    25.000000
2020-01-01 01:00:00  21.666667    21.666667
2020-01-01 02:00:00  75.000000    75.000000
2020-01-01 03:00:00  75.000000    75.000000
2020-01-01 04:00:00  75.000000    75.000000
2020-01-01 05:00:00  15.000000    15.000000
2020-01-01 06:00:00  15.000000    15.000000
2020-01-01 07:00:00  15.000000    15.000000
2020-01-01 08:00:00  15.000000    15.000000
2020-01-01 09:00:00        NaN          NaN

When initiating data from numpy, this needs to be done as follows:

For Data for each timestep and scenario, a 2D array is needed. But :

heat_demand_per_h = array([
       [ 30,  30],
       [  0,   0],
       [ 90, 100],
       [110, 118],
       [110, 125],
       [ 20,  20],
       [ 20,  20],
       [ 20,  20],
       [ 20,  20]
])

# Or for convenience:
heat_demand_per_h = np.array([[30, 0, 90, 110, 110, 20, 20, 20, 20], 
                                                         [30, 0, 100, 118, 125, 20, 20, 20, 20]]).T

For Data for each timestep, a 1D array is enough:

  heat_demand_per_h = np.array([30, 0, 90, 110, 110, 20, 20, 20, 20])

As for data per scenario

 heat_demand_per_h = np.array([20, 30])