src.blocks.ErrorIntegration module

This module is responsible for the integration / aggregation of simulation errors and uncertainties.

It includes one class for the error integration. See details in its own documentation.

Contact person: Stefan Riedmaier Creation date: 20.04.2020 Python version: 3.8

class src.blocks.ErrorIntegration.ErrorIntegration(config, domain)

Bases: object

This class is responsible for the integration / aggregation of simulation errors and uncertainties.

It includes a main method called “error_integration”. See more details in its own documentation.

error_integration(kpi_model_da, error_validation_da, numerical_uncertainty_da)

This function aggregates errors from validation and verification to the model predictions in the application.

It offers two techniques: - bias correction: use the errors to correct the nominal model predictions, - uncertainty expansion: add conservatism by expanding uncertainties. In the deterministic case, we offer both options; in the non-deterministic case, just the uncertainty expansion.

Parameters:

• kpi_model_da – array of nominal model responses

• error_validation_da – model-form errors or uncertainties from the validation domain

• numerical_uncertainty_da – numerical uncertainties from the verification domain

Returns:

array of aggregated model predictions as estimator of the actual system values

Return type:

xr.DataArray