RCAIDE.Library.Methods.Emissions.Chemical_Reactor_Network_Method.train_CRN_EI_surrogates

train_CRN_EI_surrogates#

train_CRN_EI_surrogates(emissions)[source]#

Generates training data for emission index surrogate models using Chemical Reactor Network simulations.

Parameters:

emissions (Data) –

Container for emissions data and settings

trainingData
Training data container
pressurendarray
Array of pressure values to evaluate [Pa]

temperaturendarray
Array of temperature values to evaluate [K]

air_mass_flowratendarray
Array of air mass flow rates to evaluate [kg/s]

fuel_to_air_rationdarray
Array of fuel-to-air ratios to evaluate [-]
vehicleData
Vehicle configuration data
networkslist

List of propulsion system networks

propulsorslist
Propulsion units containing combustors

Returns:

EI_CO2ndarray: CO2 emission index training data [kg_CO2/kg_fuel]
EI_COndarray: CO emission index training data [kg_CO/kg_fuel]
EI_H2Ondarray: H2O emission index training data [kg_H2O/kg_fuel]
EI_NOndarray: NO emission index training data [kg_NO/kg_fuel]
EI_NO2ndarray: NO2 emission index training data [kg_NO2/kg_fuel]

Return type:

Updates emissions.training with

Notes

This function evaluates emission indices across a grid of operating conditions to generate training data for surrogate models. The grid is formed by the combinations of the input arrays (pressure, temperature, mass flow, and fuel-to-air ratio).

Extra modules required:

numpy
Cantera

Major Assumptions

Operating points form a regular grid
All combinations of input parameters are valid

Theory Training data is generated by:

Creating a grid of operating points

Evaluating detailed chemical kinetics at each point

Storing resulting emission indices in 4D arrays

The grid dimensions are: \([P] \times [T] \times [\dot{m}] \times [FAR]\)

Where:

\(P\) = Pressure points
\(T\) = Temperature points
\(\dot{m}\) = Mass flow rate points
\(FAR\) = Fuel-to-air ratio points

train_CRN_EI_surrogates#

This Page