Initial conditions

Two ways to set the initial state of a block, both in numpy (n, n). The layout convention is row-major (ny, nx): the first index (rows) is the slow y axis, the second (columns) is the fast x axis; a field is indexed ne[j, i] (j = row / y, i = column / x).

set_density(name, arr) sets the density (component 0) and leaves the rest at rest (for a fluid: zero velocity, consistent energy). It is the usual shortcut for a scalar transport.

coord = (np.arange(n) + 0.5) / n * L
xx, yy = np.meshgrid(coord, coord, indexing="xy")          # xx, yy of shape (n, n) = (ny, nx)
r = np.hypot(xx - 0.5 * L, yy - 0.5 * L)
ne = np.full((n, n), 1e-3)
ne[(r > 0.15) & (r < 0.20)] = 1.0
sim.set_density("ne", ne)

set_primitive_state(name, **prims) initializes a fluid block from its primitive variables, named (rho, u, v, p…). Each primitive is an (n, n) array; the block model converts them into conservative variables (compressible: E = p/(g-1) + 1/2 rho|v|^2). The names expected are those of the model; an unknown or missing name raises a clear error.

sim.set_primitive_state("electrons", rho=rho0, u=u0, v=v0, p=p0)

For an explicit conservative state (diagnostic / advanced case), set_state(name, u) takes the flattened (ncomp, n, n) array.