Non-uniform lambda in using CounterflowTwinPremixedFlame

[yeanment]

Problem description
When calculating the CounterflowPremixed flame, the lambda should be uniform acorss the compuattional domain. However, the lambda is set to zero in the right symmetry boundary. This might have no significant influence on the overall results. But I believe modify it back should be a good option for consistence.

Steps to reproduce

import os
import time
import matplotlib.pyplot as plt
import numpy as np
import cantera as ct

def plot_solution(f, solution_num, path='output'):
    fig = plt.figure(figsize=(10,10))
    axs = []
    axs.append(plt.subplot2grid((2,2), (0,0))) # top left
    axs.append(plt.subplot2grid((2, 2), (1, 0)))  # middle left
    axs.append(plt.subplot2grid((2, 2), (0, 1)))  # top right
    axs.append(plt.subplot2grid((2, 2), (1, 1)))  # middle right
    
    axs[0].plot(f.grid, f.T, 'o-')
    axs[0].set_xlabel('Distance, m')
    axs[0].set_ylabel('Temperature, K')
    axs[0].set_ylim([200, 2900])
    
    axs[1].plot(f.grid, f.velocity, 'o-')
    axs[1].set_xlabel('Distance, m')
    axs[1].set_ylabel('Axial Velocity, m/s')
    
    axs[2].plot(f.grid, f.spread_rate, 'o-')
    axs[2].set_xlabel('Distance, m')
    axs[2].set_ylabel('Spread Rate, 1/s')
        
    axs[3].plot(f.grid, f.L, 'o-')
    axs[3].set_xlabel('Distance, m')
    axs[3].set_ylabel('Lambda, m/s')
    fig.tight_layout()
    plt.savefig(path + '/combined_flamelet_' + str(solution_num+1) + '.png')
    plt.close()


# Input parameters
p = 1* ct.one_atm  # pressure
tin = 300.0  # fuel inlet temperature
width = 0.02 # Distance between inlets

comp_o = 'O2:1, N2:3.76'  # oxidizer composition
comp_f = 'H2:1.0'  # fuel composition

loglevel = 1  # amount of diagnostic output (0 to 5)

gas = ct.Solution('gri30.yaml')
gas.TPX = tin, p, comp_f
gas.TPX = tin, p, comp_o

gas.set_equivalence_ratio(phi=1.0, fuel=comp_f, oxidizer=comp_o)
density_u = gas.density
f = ct.CounterflowTwinPremixedFlame(gas, width=width)

# Set the state of the two inlets
f.reactants.mdot = density_u*8
f.reactants.Y = gas.Y
f.reactants.T = tin

#f.set_refine_criteria(ratio=4, slope=0.2, curve=0.3, prune=0.04)
f.set_refine_criteria(ratio=10, slope=0.1, curve=0.1, prune=0.05)

# Solve the problem
f.solve(loglevel, auto=True)
f.save('previous_solution.yaml',overwrite=True)

plot_solution(f, -10, '.')

1. Run the above code
2. Check the result of lambda

Behavior

A non-uniform lambda is observed at the boundary point. Ideally, this should be uniform across the computational domain.

System information

• Cantera version: commit cde1e79
• OS: wsl-Ubuntu20
• Python/MATLAB/other software versions: Python 3.11

Attachments

Additional context

This problem arises due to the incorrect set of boundary conditions in function evalRightBoundary at Stflow.cpp. At line 1030 (Stflow.cpp), the code
rsd[index(c_offset_L, j)] = lambda(x, j); should be changed to rsd[index(c_offset_L, j)] = lambda(x, j) - lambda(x, j-1);, so that an zero gradient condition is set for the lambda.

[speth]

It looks like this was introduced in #1555. I had commented on exactly this issue in my review (#1555 (comment)), but it looks like the fix got put into the wrong branch of the if statement:

cantera/src/oneD/StFlow.cpp

Lines 1028 to 1034 in cde1e79

	if (m_usesLambda) {
	rsd[index(c_offset_U, j)] = rho_u(x, j);
	rsd[index(c_offset_L, j)] = lambda(x, j);
	} else {
	rsd[index(c_offset_U, j)] = rho_u(x, j) - rho_u(x, j-1);
	rsd[index(c_offset_L, j)] = lambda(x, j) - lambda(x, j-1);
	}

Please feel free to open a PR.

[ischoegl]

mea culpa 😞

[ischoegl]

@yeanment … I’ll reopen this issue until the proposed fix in #1593 is merged.