smartgrid.agents.profile.need.NeedProfile

class smartgrid.agents.profile.need.NeedProfile(need_per_hour)

Bases: object

Determines the energy needed by an Agent for each step of the simulation.

We recall that, at each time step, all Agents consume energy to satisfy their need. In other words, the need acts as some sort of target, and Agents aim to get as close as possible to this target, while taking into account their other ethical considerations (e.g., equity).

The need may differ between Agents and time steps; to offer variety, and yet a common structure, we introduce the NeedProfile. Intuitively, it represents some sort of distribution from which needs are drawn.

This class uses a data array as source of needs (see need_per_hour for details), but can be extended, e.g., for introducing stochasticity. However, it is important to specify the max_energy_needed attribute, as it is used in other components of the simulator, e.g., to determine the quantity of energy that the grid should generate at each time step (see smartgrid.util.available_energy).

The NeedProfile is usually constructed in the DataConversion class, and is part of the AgentProfile.

__init__(need_per_hour)

Create a NeedProfile.

Parameters:

need_per_hour – The list (array) of needs, i.e., floats, for each hour, such that the 1st element represents the 1st hour, and so on.

Methods

__init__(need_per_hour)	Create a NeedProfile.
compute([step])	Determine the need at a given time step.

Attributes

need_per_hour	Array of needs (one for each time step).
max_energy_needed	Maximum amount of energy that can be needed through this NeedProfile.

compute(step=0) → float

Determine the need at a given time step.

Parameters:

step – The desired time step.

Returns:

The need for this time step.

max_energy_needed: float

Maximum amount of energy that can be needed through this NeedProfile.

This value is used by other components of the simulator, such as the EnergyGenerator, which is based on the total need from all agents.

need_per_hour: List[float]

Array of needs (one for each time step).

We assume that a time step represents an hour; this array should contain the expected need of an Agent for each hour. For example, if the array is [800, 600, 1200], it means that the Agent’s need will be, during the first hour (t=0), 800, then 600 during the second hour (t=1), and 1200 during the third hour (t=2).

This class automatically handles out-of-bounds time steps, simply by cycling over the array. Thus, in the previous example, the need at time step t=3 will be the same as t=(3 % 3)=0, i.e., 800. This allows for any number of time steps in the simulation, whereas the NeedProfile itself has a fixed length.

Typically, this array should have a coherent number of elements, such as 24 (a daily profile, with a need for each hour of a single day), or 365 * 24 = 8760 (an annual profile, with a need for each hour of each day). However, this is not a hard requirement, any length would work.