2

I want to calculate auto-ignition delay time by using cantera in python linux. Here, I only change the t_end = 40 ， dt = 1，and gas.TPX = 801, P, 'H2:0.1667,O2:0.0833,N2:0.75' in custom.py example from Official Website. But, the result only calculate 20s. Anyone know how about this? my code:

import cantera as ct
import numpy as np
import scipy.integrate


class ReactorOde:
    def __init__(self, gas):
        # Parameters of the ODE system and auxiliary data are stored in the
        # ReactorOde object.
        self.gas = gas
        self.P = gas.P

    def __call__(self, t, y):
        """the ODE function, y' = f(t,y) """

        # State vector is [T, Y_1, Y_2, ... Y_K]
        self.gas.set_unnormalized_mass_fractions(y[1:])
        self.gas.TP = y[0], self.P
        rho = self.gas.density

        wdot = self.gas.net_production_rates
        dTdt = - (np.dot(self.gas.partial_molar_enthalpies, wdot) /
                  (rho * self.gas.cp))
        dYdt = wdot * self.gas.molecular_weights / rho

        return np.hstack((dTdt, dYdt))
gas = ct.Solution('h2o2.yaml')
P = ct.one_atm
gas.TPX = 801, P, 'H2:0.1667,O2:0.0833,N2:0.75'
y0 = np.hstack((gas.T, gas.Y))
ode = ReactorOde(gas)
solver = scipy.integrate.ode(ode)
solver.set_integrator('vode', method='bdf', with_jacobian=True)
solver.set_initial_value(y0, 0.0)
t_end = 40
states = ct.SolutionArray(gas, 1, extra={'t': [0.0]})
dt = 1
while solver.successful() and solver.t < t_end:
    solver.integrate(solver.t + dt)
    gas.TPY = solver.y[0], P, solver.y[1:]
    states.append(gas.state, t=solver.t)
print(states.t)

And the output is

[ 0.          1.          2.          3.          4.          5.
  6.          7.          8.          9.         10.         11.
 12.         13.         14.         15.         16.         17.
 18.         19.         19.45388009]

And the Fig about OH and T is here.enter image description here

Ray
  • 4,531
  • 1
  • 23
  • 32
YOU
  • 29
  • 3
  • what's the expected output? can you link the example from the docs? this example runs fine: https://cantera.org/examples/python/reactors/custom.py.html – willwrighteng Dec 08 '22 at 10:00
  • [github source code](https://github.com/Cantera/cantera/blob/main/samples/python/reactors/custom.py) looks like you also changed line 29 from `gas = ct.Solution('h2o2.yaml')` – willwrighteng Dec 08 '22 at 10:06
  • you'll need to provide more information. I get this error when trying to run with the kinet_C.yaml file ```CanteraError thrown by Application::findInputFile: Input file kinet_C.yaml not found in directories``` – willwrighteng Dec 08 '22 at 10:14
  • Thank you very much for your trying and finding the problem. Now, I modify my question! thanks! – YOU Dec 12 '22 at 03:36

1 Answers1

1

The loop terminates because the other condition, solver.successful() is not satisfied. The scipy.integrate.ode object also has a get_return_code() method that provides some hints about the problem. In your example, it returns -1, which, for the vode solver means:

Excess work done on this call. (Perhaps wrong MF.)

In this case, the "excess work" is because the intervals at which you're requesting output (1 second) are much longer than the time scales of the ODE that is being solved, which means that the solver has taken so many internal timesteps without reaching the next time that it assumes something has gone wrong.

Instead of specifying the output intervals explicitly, for stiff problems such as this one, you're better off just taking the solution at the times used by the integrator. This can be done by specifying the step=True argument to the integrate function, so it will just take one step toward the specified time. In this case, your integration loop would just look like:

while solver.successful() and solver.t < t_end:
    solver.integrate(solver.t + dt, step=True)
    gas.TPY = solver.y[0], P, solver.y[1:]
    states.append(gas.state, t=solver.t)

With this change, I see that the end time exceeds the specified t_end, as expected. You can also see that integrating this system requires time steps much smaller than the 1 second output interval you've specified by calculating the median step size:

>>> np.median(np.diff(states.t))
3.229414249972251e-05
Ray
  • 4,531
  • 1
  • 23
  • 32