sunpeek.core_methods.common.main.CoreAlgorithm

class sunpeek.core_methods.common.main.CoreAlgorithm(component, strategies=None, *args, **kwargs)

Bases: ABC

Superclass for all SunPeek core algorithms, mainly virtual sensors, Power Check and D-CAT energy yield methods.

This class handles various strategies for an algorithm (e.g. various implementations to calculate thermal power, or various Performance Check methods, equations etc.

*args and **kwargs passed to object creation are forwarded to define_strategies().

Methods

__init__(component[, strategies])
allowed_components()
check_interval([eval_start, eval_end])	Make sure we have data in the specified interval, and check that plant's virtual sensors are up-to-date.
create_variants(arg, allowed_type, default)	Create list with sanitized inputs for algo strategies.
define_strategies(*args, **kwargs)
get_config_feedback()	Cycle through all algo strategies, return CoreMethodFeedback of all strategy problems.
run([on_strategy_error])	Calculates algorithm using its defined strategies, stopping at the first successful strategy.

check_interval(eval_start=None, eval_end=None)

Make sure we have data in the specified interval, and check that plant’s virtual sensors are up-to-date.

Return type:

None

static create_variants(arg, allowed_type, default)

Create list with sanitized inputs for algo strategies. Set default if needed.

:raises TypeError : if `arg` or default does not match allowed_type.: :raises AlgorithmError : if no valid variants are found:

Return type:

List[Any]

get_config_feedback()

Cycle through all algo strategies, return CoreMethodFeedback of all strategy problems. Stops at first successful strategy, copies problem slots from strategy.

Return type:

CoreMethodFeedback

run(on_strategy_error='skip')

Calculates algorithm using its defined strategies, stopping at the first successful strategy.

Parameters:

on_strategy_error (str, optional) – If ‘raise’, exceptions that occur during a strategy.execute() are raised. If not, they are saved as own_feedback in self.problems. In any case, errors are logged.

Return type:

AlgoResult

:raises AlgorithmError : if algorithm has no strategies defined, or if getting some strategy problems fails.: