cleanup class_ems.py

[NormannK]

• type hints added
• translated to English
• reduced not needed function calls

improves optimization time by 50%

[NormannK]

["Last_Wh_pro_Stunde"][0], ["Netzeinspeisung_Wh_pro_Stunde"][0], ["Netzbezug_Wh_pro_Stunde"][0] were None . But those should be numbers afaik. The lists were initialized with NaN values which later got converted into None.
I changed it by initializing the lists with zeros avoiding the whole conversion.
please confirm.

[NormannK]

old Last_Wh_pro_Stunde result:
[None, 23759.13, 996.88, 531.38, 517.95, 483.15, 472.28, 1011.68, 995.0, 1053.07, 1063.91, 1320.56, 1132.03, 1163.67, 1176.82, 1216.22, 1103.78, 1129.12, 1178.71, 1050.98, 988.56, 912.38, 704.61, 516.37, 868.05, 694.34, 608.79, 556.31, 488.89, 506.91, 804.89, 1141.98, 1056.97, 992.46, 1155.99, 827.01, 1257.98, 1232.67, 871.26, 860.88, 1158.03, 1222.72, 1221.04, 949.99, 987.01, 733.99, 592.97]
new Last_Wh_pro_Stunde result:
[0.0, 23759.13, 996.88, 531.38, 517.95, 483.15, 472.28, 1011.68, 995.0, 1053.07, 1063.91, 1320.56, 1132.03, 1163.67, 1176.82, 1216.22, 1103.78, 1129.12, 1178.71, 1050.98, 988.56, 912.38, 704.61, 516.37, 868.05, 694.34, 608.79, 556.31, 488.89, 506.91, 804.89, 1141.98, 1056.97, 992.46, 1155.99, 827.01, 1257.98, 1232.67, 871.26, 860.88, 1158.03, 1222.72, 1221.04, 949.99, 987.01, 733.99, 592.97]

[michaelosthege]

I haven't familiarized myself with the simulation and optimization yet, but my gut feeling is that NaN should be prefered over None.
For example, a np.array([1,None,3]) is of dtype=object whereas np.array([1, np.nan, 2]) is dtype=float.

[NormannK]

yeah i was also not that sure anymore. I read this article to remember things. NaN is a float by definition and None is a "None Type" so dtype must change.
But for the optimization in the end that would not matter since the last step in the class was/is to translate all NaN into None anyways.

I would understand if, if the first value is NaN or None if we dont want to have effect that value the current running operation. So basically saying "hey don't touch anything for the current hour we only want to have values for the next hour"
Be we only have no value there bc the line 103 in the old for stunde in range(start_stunde + 1, ende): starts one hour later and there is no documentation why it does that. (or i cant find it)

[drbacke]

Home Appliances not working anymore. They are ignored

[drbacke]

Home Appliances not working anymore

[NormannK]

the test_class_ems.py sets haushaltsgeraet = None so it assumes there are none.

also i did check the result vs single_test_optimization.py. Its the same.
The code does not touch any logic, could you provide the input data you used?

[drbacke]

I fixed it, the call (set_startzeitpunkt returns no object, that was the problem).
I also added the home appl. to the unit tests, so that we will see that problem in the future.