Class GridModule

src/pymgrid/modules/grid_module.py:8–358

An electrical grid module. By default, ``GridModule`` is a *fixed* module; it can be transformed to a flex module with ``GridModule.as_flex``. ``GridModule`` is the only built-in module that can be both a fixed and flex module. Parameters ---------- max_import : float

class GridModule(BaseTimeSeriesMicrogridModule):
    """
    An electrical grid module.

    By default, ``GridModule`` is a *fixed* module; it can be transformed to a flex module with ``GridModule.as_flex``.
    ``GridModule`` is the only built-in module that can be both a fixed and flex module.

    Parameters
    ----------
    max_import : float
        Maximum import at any time step.

    max_export : float
        Maximum export at any time step.

    time_series : array-like, shape (n_features, n_steps), n_features = {3, 4}
        If n_features=3, time series of ``(import_price, export_price, co2_per_kwH)`` in each column, respectively.
        Grid is assumed to have no outages.
        If n_features=4, time series of ``(import_price, export_price, co2_per_kwH, grid_status)``
        in each column, respectively. ``time_series[:, -1]`` -- the grid status -- must be binary.

    forecaster : callable, float, "oracle", or None, default None.
        Function that gives a forecast n-steps ahead.

        * If ``callable``, must take as arguments ``(val_c: float, val_{c+n}: float, n: int)``, where

          * ``val_c`` is the current value in the time series: ``self.time_series[self.current_step]``

          * ``val_{c+n}`` is the value in the time series n steps in the future

          * n is the number of steps in the future at which we are forecasting.

          The output ``forecast = forecaster(val_c, val_{c+n}, n)`` must have the same sign
          as the inputs ``val_c`` and ``val_{c+n}``.

        * If ``float``, serves as a standard deviation for a mean-zero gaussian noise function
          that is added to the true value.

        * If ``"oracle"``, gives a perfect forecast.

        * If ``None``, no forecast.

    forecast_horizon : int.
        Number of steps in the future to forecast. If forecaster is None, ignored and 0 is returned.

    forecaster_increase_uncertainty : bool, default False
        Whether to increase uncertainty for farther-out dates if using a GaussianNoiseForecaster. Ignored otherwise.

    cost_per_unit_co2 : float, default 0.0
        Marginal cost of grid co2 production.

    normalized_action_bounds : tuple of int or float, default (0, 1).
        Bounds of normalized actions.
        Change to (-1, 1) for e.g. an RL policy with a Tanh output activation.

    raise_errors : bool, default False
        Whether to raise errors if bounds are exceeded in an action.
        If False, actions are clipped to the limit possible.