Skip to content

Discharge overview#

Overview#

pysimdeum allows for the simulation of water discharge at the end-use level. The discharge simulation relies on the simulation of water consumption at the end-use level. On a basic level, discharge simulations work on the principle of conservation of volume, with the exception for consumption events that result in use of water, such as drinking water from a kitchen tap.

By default, discharge simulations are not simulated, but can be enabled by the simulate_discharge=True argument when simulating a house: 

house.simulate(duration='1 day', num_patterns=1, simulate_discharge=True)

Discharge of water typically occurs after the consumption event begins. For simple enduses such as a BathroomTap this discharge begins when consumption is initiated. Other appliances such as a Bathtub have a usage time resulting in a delay before a discharge event begins following a consumption event. More complex enduses such as the WashingMachine and Dishwasher have cycles of discharge events that follow their respective consumption cycles. The Wc is an exception where the discharge event must occur immediately before a consumption event.

Methodology#

Discharge calculations in pysimdeum follow a simple method of calculating the volume of water consumed over a consumption event of an enduse, and discharging this at a sampled discharge flow rate. To provide stochastic behaviour, low and high ranges of discharge flow rates inform the bounds of a uniform distribution from which a discharge flow rate for a specific enduse discharge event is sampled. Details for each enduse are described in the sections Common Methods and Enduse specifics.

Object structure#

The discharge object follows an almost identical structure to the consumption object. Results are stored as a xarray.DataArray so that simulation information can be accessed and aggregared afterwards (e.g. specific enduses, sum over water discharge of users, rolling means over time etc.).

The discharge array has four dimensions:

  • Time
  • User
  • Enduse
  • Patterns