Integrator Settings

[jrleeman]

We need to enable passing other settings to the integrator. I've noticed in some cases that the integrator will die when doing something like setting the maximum step size to a much smaller size works.

import numpy as np
import matplotlib.pyplot as plt
import rsf

model = rsf.RateState()

# Set model initial conditions
model.mu0 = 0.6 # Friction initial (at the reference velocity)
model.a = 0.007 # Empirical coefficient for the direct effect
model.b = 0.003 # Empirical coefficient for the evolution effect
model.dc = 100. # Critical slip distance
model.k = 0.00128 # Normalized System stiffness (friction/micron)
model.v = 300. # Initial slider velocity, generally is vlp(t=0)
model.vref = 300. # Reference velocity, generally vlp(t=0)
model.stateLaw = rsf.dieterichState # Which state relation we want to use

model.model_time = np.arange(0,450.01,0.01)

lp_velocity = 300* np.ones_like(model.model_time)
lp_velocity[10*100:12*100] = -300. 
lp_velocity[12*100:110.5*100] = 0.

# Set the model load point velocity, must be same shape as model.model_time
model.loadpoint_velocity = lp_velocity

# Run the model!
results = model.solve()

# Make the phase plot
model.phasePlot()

# Make a plot in displacement
model.dispPlot()

# Make a plot in time
model.timePlot()

[dopplershift]

One solution would be to have solve() take arbitrary keyword arguments with **kwargs and then pass them to odeint().