Setting up a multi-process laminar flame speed parameter study#

One of the prevailing use case of the freely propagating premixed flame model is to build a flame speed table to be imported by another combustion simulation tools. PyChemkin provides the flexibility to customize the data structure of the flame speed table depending on the simulation goals and the tool. Furthermore, over the years, the chemkin flame speed calculator has derived a set of default solver settings that would greatly improve the convergence performance, especially for those widely adopted hydrocarbon fuel combustion mechanisms. The required input parameters the flame speed calculator are reduced to the composition of the fuel-oxidizer mixture, the initial/inlet pressure and temperature, and the calculation domain.

This tutorial shows the steps of setting up a flame speed parameter study for CH4-air mixtures at the 5 atmosphere pressure. The predicted flame speed values are compared against the experimental data as a function of the mixture equivalence ratio. The parameter study is performed in the multi-process (or multi-core) mode by using ProcessPoolExecutor from the concurrent.futures package. If the computer has enough number of CPU cores, the parameter study can be run in parallel with each case running on a designated CPU core.

Since the transport processes are critical for flame calculations, the transport data must be included in the mechanism data and preprocessed.


Import PyChemkin packages and start the logger#

from concurrent.futures import ProcessPoolExecutor
import os
from pathlib import Path
import time

import ansys.chemkin.core as ck  # Chemkin
from ansys.chemkin.core.inlet import Stream  # external gaseous inlet
from ansys.chemkin.core.logger import logger

# Chemkin 1-D premixed freely propagating flame model (steady-state)
from ansys.chemkin.core.premixedflames.premixedflame import (
    FreelyPropagating as FlameSpeed,
)
from ansys.chemkin.core.utilities import WorkingFolders
import matplotlib.pyplot as plt  # plotting
import numpy as np  # number crunching

# check working directory
current_dir = str(Path.cwd())
logger.debug("working directory: " + current_dir)
# set verbose mode
ck.set_verbose(True)
# set interactive mode for plotting the results
# interactive = True: display plot
# interactive = False: save plot as a PNG file
global interactive
interactive = True

Create a flame speed calculator class#

Create a local class that wraps around the actual FlameSpeed class to make the setup of the multi-process flame speed calculation parameter study more convenient.

class FlameSpeedCalculator:
    """Laminar flame speed calculator with fixed set up parameters."""

    def __init__(self, fresh_mixture: Stream, index: int):
        """Create a Laminar flame speed calculator."""
        """
        Create a Laminar flame speed calculator that instantiates a FlameSpeed object
        with the given fresh (unburnt) mixture condition.

        Parameters
        ----------
            fresh_mixture: Mixture object
                the initial/fresh/unburnt condition
            index: integer
                run index of this flame speed calculator
        """
        # instantiate a flame speed object
        # set up the run and working directory name
        self.name = "Flame_Speed_" + str(index)
        # instantiate the FlameSpeed object for this run
        self.fs_calculator = FlameSpeed(fresh_mixture, label=self.name)
        # set the required premixed flame model parameters
        #
        # set the maximum total number of grid points allowed
        # in the calculation (optional)
        # self.fs_calculator.set_max_grid_points(150)
        # define the calculation domain [cm]
        self.fs_calculator.end_position = 1.0
        # set the root directory
        self.root_dir = str(Path.cwd())
        # set the working directory
        self.work_dir = str(Path(self.root_dir) / self.name)
        # run status
        self.runstatus = -100
        # calculated laminar flame speed [cm/sec]
        self.flame_speed = 0.0

    def run(self):
        """Run the flame speed calculation in a separate working directory."""
        # run the flame speed calculation
        self.runstatus = self.fs_calculator.run()
        # extract the laminar flame speed from the solution
        if self.runstatus == 0:
            # postprocess the solutions
            self.fs_calculator.process_solution()
            # get the flame speed value [cm/sec]
            # because the memory is shared, it must be done as soon as
            # the run is finished
            self.flame_speed = self.fs_calculator.get_flame_speed()

    def get_flame_speed(self) -> float:
        """Get the predicted laminar flame speed."""
        """
        Get the predicted laminar flame speed.

        Returns
        -------
            flame_speed: double
                predicted laminar flame speed [cm/sec]
        """
        return self.flame_speed

Set up the flame speed parameter study for multi-processing#

Create a list of FlameSpeedCalculator objects with different initial methane-air equivalence ratios. Each object represents one parameter study case and will be run on an designated cpu core when the parameter study is executed.

def flame_speed_run(case: tuple[int, float]) -> tuple[float, float]:
    """Set up the flame speed parameter study."""
    """
    Set up the parameter study runs for multi-processing.

    Parameters
    ----------
        case: tuple (integer, double)
            flame speed calculation case condition: case index and
            the equivalence ratio

    Returns
    -------
        phi: double
            the parameter: fresh mixture equivalence ratio
        flame_speed: double
            predicted laminar flame speed of the fresh mixture [cm/sec]
    """
    ##########################################
    # Create an instance of the Chemistry Set
    # ========================================
    # The mechanism loaded is the GRI 3.0 mechanism for methane combustion.
    # The mechanism and its associated data files come with the standard Ansys Chemkin
    # installation under the subdirectory *"/reaction/data"*.
    #
    # case index
    index = case[0]
    # fresh mixture equivalence ratio
    phi = case[1]
    # create and change to the working directory for this run
    name = "Flame_Speed_" + str(index)
    work_folder = WorkingFolders(name, current_dir)
    # set mechanism directory (the default Chemkin mechanism data directory)
    data_dir = Path(ck.ansys_dir) / "reaction" / "data"
    mechanism_dir = data_dir
    # including the full file path is recommended
    chemfile = str(mechanism_dir / "grimech30_chem.inp")
    thermfile = str(mechanism_dir / "grimech30_thermo.dat")
    tranfile = str(mechanism_dir / "grimech30_transport.dat")
    # create a chemistry set based on GRI 3.0
    MyGasMech = ck.Chemistry(
        chem=chemfile, therm=thermfile, tran=tranfile, label="GRI 3.0"
    )

    ##############################
    # Preprocess the Chemistry Set
    # ============================

    # preprocess the mechanism files
    ierror = MyGasMech.preprocess()
    if ierror != 0:
        print("Error: failed to preprocess the mechanism!")
        print(f"       error code = {ierror}")
        exit()

    ########################################################################
    # Set up the CH\ :sub:`4`\ -air mixture for the flame speed calculation
    # ======================================================================
    # Instantiate a stream named ``premixed`` for the inlet gas mixture.
    # This stream  is a mixture with the addition of the
    # inlet flow rate. You can specify the inlet gas properties the same way you
    # set up a ``Mixture``. Here the ``x_by_equivalence_ratio`` method is used.
    # You create the ``fuel`` and the ``air`` mixtures first. Then define the
    # *complete combustion product species* and provide the *additives* composition
    # if applicable. And finally, during the parameter iteration runs, you can
    # simply set different values to ``equivalenceratio`` to create different
    # methane-air mixtures.
    #

    # create the fuel mixture
    fuel = ck.Mixture(MyGasMech)
    # set fuel composition: methane
    fuel.x = [("CH4", 1.0)]
    # setting pressure and temperature condition for the flame speed calculations
    fuel.pressure = 5.0 * ck.P_ATM
    fuel.temperature = 300.0  # inlet temperature

    # create the oxidizer mixture: air
    air = ck.Mixture(MyGasMech)
    air.x = ck.Air.x()
    # setting pressure and temperature is not required in this case
    air.pressure = fuel.pressure
    air.temperature = fuel.temperature

    # create the fuel-air Stream for the premixed flame speed calculation
    premixed = Stream(MyGasMech, label="premixed")
    # products from the complete combustion of the fuel mixture and air
    products = ["CO2", "H2O", "N2"]
    # species mole fractions of added/inert mixture.
    # can also create an additives mixture here
    add_frac = np.zeros(MyGasMech.kk, dtype=np.double)  # no additives: all zeros

    # setting pressure and temperature is not required in this case
    premixed.pressure = fuel.pressure
    premixed.temperature = fuel.temperature

    # set estimated value of the flame mass flux [g/cm2-sec]
    premixed.mass_flowrate = 0.4
    # create mixture by using the equivalence ratio
    ierror = premixed.x_by_equivalence_ratio(
        MyGasMech, fuel.x, air.x, add_frac, products, equivalenceratio=phi
    )
    # check fuel-oxidizer mixture creation status
    if ierror != 0:
        print(
            "Error: failed to create the methane-air mixture "
            + "for equivalence ratio = "
            + str(phi)
        )
        exit()
    # create a flame speed calculation instance
    this_run = FlameSpeedCalculator(premixed, index=index)
    # run the case
    this_run.run()
    # change back to the original top folder
    work_folder.done()
    #
    return phi, this_run.flame_speed

Set up and start the multi-process runs#

Use the ProcessPoolExecutor() method to assign the flame speed runs. In this project, each flame speed run/case runs on an exclusive cpu core. The first parameter of map() method should be flame_speed_run(). Make the flame_speed_run() method return the required parameter and results to make post-processing the results from the parameter study easier.

Note

if __name__ == '__main__' is required when multiprocessing package is used.

if __name__ == "__main__":
    # number of available cpu cores
    numb_cores = max(os.cpu_count(), 1)
    # set the number of worker cpu cores to be used by the parameter study
    numb_workers = 14
    numb_workers = max(numb_workers, 1)
    numb_workers = min(numb_workers, numb_cores - 2)
    # Set up the flame speed parameter study for multi-processing
    # equivalence ratio for the first case
    phi = 0.6
    # total number of parameter cases
    numb_cases = 21
    # equivalence ratio increment
    delta_phi = 0.05
    # set up flame speed calculation runs with different equivalence ratios
    flamespeed_cases: list[tuple[int, float]] = []
    for i in range(numb_cases):
        # create mixture by using the equivalence ratio
        flamespeed_cases.append((i, phi))
        # update parameter
        phi += delta_phi
    # start the multi-process
    case_id = 0
    # set the start wall time
    start_time = time.time()
    #
    numb_workers = min(numb_workers, numb_cases)
    equiv: list[float] = []
    flame_speed: list[float] = []
    with ProcessPoolExecutor(max_workers=numb_workers) as e:
        for ret_value in e.map(flame_speed_run, flamespeed_cases):
            # results returned by the flame speed calculator
            # equivalence ratio
            equiv.append(ret_value[0])
            # predicted laminar flame speed [cm/sec]
            flame_speed.append(ret_value[1])

    # compute the total runtime
    runtime = time.time() - start_time
    print()
    print(f"total simulation duration: {runtime} [sec]")
    print()

    # experimental data by Kochar
    # equivalence ratios
    data_equiv = [
        0.7005,
        0.8007,
        0.9009,
        1.001,
        1.1032,
        1.2014,
        1.3014,
    ]
    # methane flame speeds at 5 atm
    data_speed = [
        6.906,
        12.0094,
        15.9072,
        19.2376,
        19.6601,
        15.8274,
        10.2925,
    ]

    ###########################################
    # Plot the premixed flame solution profiles
    # =========================================
    # Plot the predicted flame speeds against the experimental data.

    plt.plot(
        data_equiv, data_speed, label="data", linestyle="", marker="^", color="blue"
    )
    plt.plot(equiv, flame_speed, label="GRI 3.0", linestyle="-", color="blue")
    plt.legend()
    plt.ylabel("Flame Speed [cm/sec]")
    plt.xlabel("Equivalence Ratio")

    # clean up
    ck.done()

    # plot results
    if interactive:
        plt.show()
    else:
        plt.savefig("plot_flame_speed_pooling.png", bbox_inches="tight")

Gallery generated by Sphinx-Gallery