2. Consumption object¶

The consumption object is one of the core objects in pySIMDEUM. By default, a consumption profile is simulated when generating a House. Effectively, it is a representation of the water consumption for a household.

Consumption data structure¶

The xarray.DataArray structure of consumption is a multidimensional array to handle the labelled flow data. The core of the DataArray is the numerical data of consumption flow rates in litres per second.

Dimensions define the axes of the data, and are:

• time: temporal resolution of the data
• enduse: appliances that are sampled to populated this household
• user: individual household members
• patterns: represents different consumption patterns / scenarios

The coordinates are effectively the axes labels. For example, time will be populated with timestamps. enduse will be specific appliances such as shower, wc etc.

In [1]:

import pysimdeum
import matplotlib.pyplot as plt

import pysimdeum import matplotlib.pyplot as plt

Initialise a two-person household

In [2]:

# Build a house (two-person household)
house = pysimdeum.built_house(house_type='two_person')

# Build a house (two-person household) house = pysimdeum.built_house(house_type='two_person')

The consumption profile is simulated when generating a House and can be accessed directly from the House instance.

In [3]:

house.consumption

house.consumption

Out[3]:

<xarray.DataArray (time: 86401, user: 3, enduse: 8, patterns: 1, flowtypes: 2)> Size: 33MB
array([[[[[0., 0.]],

         [[0., 0.]],

         [[0., 0.]],

         ...,

         [[0., 0.]],

         [[0., 0.]],

         [[0., 0.]]],


        [[[0., 0.]],

         [[0., 0.]],

         [[0., 0.]],
...
         [[0., 0.]],

         [[0., 0.]],

         [[0., 0.]]],


        [[[0., 0.]],

         [[0., 0.]],

         [[0., 0.]],

         ...,

         [[0., 0.]],

         [[0., 0.]],

         [[0., 0.]]]]])
Coordinates:
  * time       (time) datetime64[ns] 691kB 2025-04-25 ... 2025-04-26
  * user       (user) <U9 108B 'user_1' 'user_2' 'household'
  * enduse     (enduse) <U14 448B 'Wc' 'Bathtub' ... 'Shower' 'WashingMachine'
  * patterns   (patterns) int64 8B 0
  * flowtypes  (flowtypes) <U9 72B 'totalflow' 'hotflow'

xarray.DataArray

• time: 86401
• user: 3
• enduse: 8
• patterns: 1
• flowtypes: 2

• 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 ... 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

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

• Coordinates: (5)
  • time
    (time)
    datetime64[ns]
    2025-04-25 ... 2025-04-26

    array(['2025-04-25T00:00:00.000000000', '2025-04-25T00:00:01.000000000',
           '2025-04-25T00:00:02.000000000', ..., '2025-04-25T23:59:58.000000000',
           '2025-04-25T23:59:59.000000000', '2025-04-26T00:00:00.000000000'],
          dtype='datetime64[ns]')

  • user
    (user)
    <U9
    'user_1' 'user_2' 'household'

    array(['user_1', 'user_2', 'household'], dtype='<U9')

  • enduse
    (enduse)
    <U14
    'Wc' 'Bathtub' ... 'WashingMachine'

    array(['Wc', 'Bathtub', 'BathroomTap', 'Dishwasher', 'KitchenTap',
           'OutsideTap', 'Shower', 'WashingMachine'], dtype='<U14')

  • patterns
    (patterns)
    int64
    0

    array([0])

  • flowtypes
    (flowtypes)
    <U9
    'totalflow' 'hotflow'

    array(['totalflow', 'hotflow'], dtype='<U9')

• Indexes: (5)
  • time
    PandasIndex

    PandasIndex(DatetimeIndex(['2025-04-25 00:00:00', '2025-04-25 00:00:01',
                   '2025-04-25 00:00:02', '2025-04-25 00:00:03',
                   '2025-04-25 00:00:04', '2025-04-25 00:00:05',
                   '2025-04-25 00:00:06', '2025-04-25 00:00:07',
                   '2025-04-25 00:00:08', '2025-04-25 00:00:09',
                   ...
                   '2025-04-25 23:59:51', '2025-04-25 23:59:52',
                   '2025-04-25 23:59:53', '2025-04-25 23:59:54',
                   '2025-04-25 23:59:55', '2025-04-25 23:59:56',
                   '2025-04-25 23:59:57', '2025-04-25 23:59:58',
                   '2025-04-25 23:59:59', '2025-04-26 00:00:00'],
                  dtype='datetime64[ns]', name='time', length=86401, freq='s'))

  • user
    PandasIndex

    PandasIndex(Index(['user_1', 'user_2', 'household'], dtype='object', name='user'))

  • enduse
    PandasIndex

    PandasIndex(Index(['Wc', 'Bathtub', 'BathroomTap', 'Dishwasher', 'KitchenTap',
           'OutsideTap', 'Shower', 'WashingMachine'],
          dtype='object', name='enduse'))

  • patterns
    PandasIndex

    PandasIndex(Index([0], dtype='int64', name='patterns'))

  • flowtypes
    PandasIndex

    PandasIndex(Index(['totalflow', 'hotflow'], dtype='object', name='flowtypes'))

• Attributes: (0)

You can access the dimensions and coordinates of the consumption DataArray.

In [4]:

print(house.consumption.dims)

print(house.consumption.dims)

('time', 'user', 'enduse', 'patterns', 'flowtypes')

In [5]:

print(house.consumption.coords)

print(house.consumption.coords)

Coordinates:
  * time       (time) datetime64[ns] 691kB 2025-04-25 ... 2025-04-26
  * user       (user) <U9 108B 'user_1' 'user_2' 'household'
  * enduse     (enduse) <U14 448B 'Wc' 'Bathtub' ... 'Shower' 'WashingMachine'
  * patterns   (patterns) int64 8B 0
  * flowtypes  (flowtypes) <U9 72B 'totalflow' 'hotflow'

The consumption contains two flowtypes labels of totalflow and hotflow. hotflow is a fractional representation of the totalflow based on the ratio of the enduse temperature to the globally defined cold and hot water temperatures. See documentation for more information.

In [6]:

house.consumption.sum(["enduse","user"]).sel(flowtypes="totalflow").plot()
house.consumption.sum(["enduse","user"]).sel(flowtypes="hotflow").plot()
plt.show()

house.consumption.sum(["enduse","user"]).sel(flowtypes="totalflow").plot() house.consumption.sum(["enduse","user"]).sel(flowtypes="hotflow").plot() plt.show()

You can access the consumption water profile for specific enduses. Enduse consumption patterns are generated in slightly differing way to represent the actual function of the appliance. See documentation for detailed explanations.

In [7]:

house.consumption.sum(["user"]).sel(enduse="WashingMachine").sel(flowtypes="totalflow").plot()

house.consumption.sum(["user"]).sel(enduse="WashingMachine").sel(flowtypes="totalflow").plot()

Out[7]:

[<matplotlib.lines.Line2D at 0x13cb259d0>]