The velocity of inflow code problem

[dokken]

You could use an UserExpression, where you use numpy.random to create your random variation on each evaluation.
See for instance:

Material property depending on solution

Hi Rem, I’m really not an expert but have you tried a UserExpression? Something like class CoeffClass(UserExpression): def __init__(self, temp, **kwargs): super().__init__(kwargs) self.temp = temp def eval(self, values, x): values[0] = ... 2*self.temp(x) ... if x[0] > 0.5: values[1] = 1 else: ... values[1] = 0 def value_shape(self): return () Indexing needs to match your function space dimensions,…

or

Define random values through expression

Hi, f=Expression(“g”, g=g, degree=1) should be thought of as defining f such that f(x)=g(x). In ran_mu_expr=Expression(“ran_mu”,ran_mu=ran_mu, degree=4) you seem to make an assumption that Expression will do the lookup in ran_mu array to assign the coefficient value appropriately. That is not the case. Consider instead from dolfin import * import numpy as np class RandomScalar(UserExpression): def eval(self, value, x): value[:] = np.random.rand() def value_shape(self, ): …

or

Initial Conditions for mixed Element

Hello, I’m trying to solve a different problem: I have a poromechanic Material and I’m pulling on the right side. I’m not sure, if my implementation of the initial Condition is correct, because I define, that the pressure at p0 = 10000 but the pressure of my next time step is in a range from 0 (because of my boundary Condition) to -100. I dont understand the big difference between t0 and t1. Is there a mistake in my implementation? Or do I have to implement an initial pressure differently? …