notebook.community

Edit and run



In [1]:

    
from nilmtk import DataSet, MeterGroup
import warnings
warnings.filterwarnings("ignore")









    



Vendor:  Continuum Analytics, Inc.
Package: mkl
Message: trial mode expires in 29 days



In [2]:

    
ds = DataSet("/Users/nipunbatra/Downloads/wikienergy-2.h5")



In [3]:

    
from nilmtk.disaggregate import CombinatorialOptimisation



In [4]:

    
co = CombinatorialOptimisation()



In [5]:

    
building_num = 11



In [6]:

    
elec = ds.buildings[building_num].elec

Reducing time window



In [7]:

    
ds.set_window(start='2014-04-01 00:00:00', end='2014-05-01 00:00:00')



In [8]:

    
elec









    Out[8]:





MeterGroup(meters=
  ElecMeter(instance=1, building=11, dataset='WikiEnergy', site_meter, appliances=[])
  ElecMeter(instance=2, building=11, dataset='WikiEnergy', appliances=[Appliance(type='air conditioner', instance=1)])
  ElecMeter(instance=3, building=11, dataset='WikiEnergy', appliances=[Appliance(type='sockets', instance=1)])
  ElecMeter(instance=4, building=11, dataset='WikiEnergy', appliances=[Appliance(type='sockets', instance=2)])
  ElecMeter(instance=5, building=11, dataset='WikiEnergy', appliances=[Appliance(type='dish washer', instance=1)])
  ElecMeter(instance=6, building=11, dataset='WikiEnergy', appliances=[Appliance(type='spin dryer', instance=1)])
  ElecMeter(instance=7, building=11, dataset='WikiEnergy', appliances=[Appliance(type='electric furnace', instance=1)])
  ElecMeter(instance=8, building=11, dataset='WikiEnergy', appliances=[Appliance(type='sockets', instance=3)])
  ElecMeter(instance=9, building=11, dataset='WikiEnergy', appliances=[Appliance(type='sockets', instance=4)])
  ElecMeter(instance=10, building=11, dataset='WikiEnergy', appliances=[Appliance(type='fridge', instance=1)])
  ElecMeter(instance=11, building=11, dataset='WikiEnergy', appliances=[Appliance(type='electric water heating appliance', instance=1)])
)



In [9]:

    
fridge_elecmeter = elec['fridge']



In [10]:

    
fridge_elecmeter









    Out[10]:





ElecMeter(instance=10, building=11, dataset='WikiEnergy', appliances=[Appliance(type='fridge', instance=1)])



In [11]:

    
fridge_mg = MeterGroup([fridge_elecmeter])



In [12]:

    
co.train(fridge_mg)









    



Training model for submeter 'ElecMeter(instance=10, building=11, dataset='WikiEnergy', appliances=[Appliance(type='fridge', instance=1)])'
Done training!



In [13]:

    
co.model









    Out[13]:





[{'states': array([  0, 139, 431], dtype=int32),
  'training_metadata': ElecMeter(instance=10, building=11, dataset='WikiEnergy', appliances=[Appliance(type='fridge', instance=1)])}]

So, fridge is learnt as a 3 state appliance. What if we wanted to specify it to use 2 states? The latest version of nilmtk allows us to specify the number of states for an appliance.



In [14]:

    
num_states_dict = {fridge_elecmeter:2}



In [15]:

    
co = CombinatorialOptimisation()
co.train(fridge_mg, num_states_dict=num_states_dict)









    



Training model for submeter 'ElecMeter(instance=10, building=11, dataset='WikiEnergy', appliances=[Appliance(type='fridge', instance=1)])'
Done training!



In [16]:

    
co.model









    Out[16]:





[{'states': array([  0, 143], dtype=int32),
  'training_metadata': ElecMeter(instance=10, building=11, dataset='WikiEnergy', appliances=[Appliance(type='fridge', instance=1)])}]

Now, fridge is learnt as a 2 state appliance.

Let us try the same thing with FHMM now



In [17]:

    
from nilmtk.disaggregate import FHMM



In [18]:

    
f = FHMM()



In [19]:

    
f.train(fridge_mg)









    



Learnt states:  3
Training model for submeter 'ElecMeter(instance=10, building=11, dataset='WikiEnergy', appliances=[Appliance(type='fridge', instance=1)])'



In [20]:

    
f.model.means_









    Out[20]:





array([[   3.02034344],
       [ 146.2202185 ],
       [ 176.49027576]])



In [21]:

    
f = FHMM()
f.train(fridge_mg, num_states_dict)









    



User specified  2
Training model for submeter 'ElecMeter(instance=10, building=11, dataset='WikiEnergy', appliances=[Appliance(type='fridge', instance=1)])'



In [22]:

    
f.model.means_









    Out[22]:





array([[   0.        ],
       [ 144.87951888]])

So, now we have a 2 state learnt model for the Fridge.



In [ ]: