def test_invalid_density():
"Check if density with shape[0] != 3 raises an error"
model = np.array([[-100, 100, -100, 100, 100, 200]])
density = np.array([1000])
coordinates = np.array([0, 0, 0])
with pytest.raises(ValueError):
prism.grav(coordinates, model, density, field="g_z")
def test_invalid_grav_field():
"Check if passing an invalid field raises an error"
model = np.array([[-100, 100, -100, 100, 100, 200]])
density = np.array([1000])
coordinates = np.array([[0], [0], [0]])
with pytest.raises(ValueError):
prism.grav(coordinates, model, density, field="invalid field")
def test_invalid_coordinates():
"Check if passing an invalid coordinates raises an error"
model = np.array([[-100, 100, -100, 100, 100, 200]])
density = np.array([1000])
# shape (1,)
coordinates = np.array([0, 0, 0])
with pytest.raises(ValueError):
prism.grav(coordinates, model, density, field="g_z")
# shape (4,3)
coordinates = np.zeros((4, 3))
with pytest.raises(ValueError):
prism.grav(coordinates, model, density, field="g_z")
def test_Poisson_equation():
"Sum of derivatives xx, yy and zz must -4*pi*rho*G inside the prism"
model = np.array([[-100, 100, -130, 100, 100, 213]])
coords = np.array([[0, -10, 80], [0, 30, -62.1], [150, 110, 200]])
rho = np.array([1300])
gxx = prism.grav(coordinates=coords,
prisms=model,
density=rho,
field="g_xx")
gyy = prism.grav(coordinates=coords,
prisms=model,
density=rho,
field="g_yy")
gzz = prism.grav(coordinates=coords,
prisms=model,
density=rho,
field="g_zz")
reference_value = -4 * np.pi * 1300 * prism.GRAVITATIONAL_CONST * 1e9
npt.assert_almost_equal(gxx + gyy + gzz,
np.zeros(coords.shape[1]) + reference_value,
decimal=12)
def test_Laplace_equation():
"Sum of derivatives xx, yy and zz must be zero outside the prism"
model = np.array([[-100, 100, -130, 100, 100, 213]])
coords = np.array([[0, -100, 100, -100, 400], [0, -130, 100, 50, 400],
[0, -100, -100, -100, -100]])
rho = np.array([1300])
gxx = prism.grav(coordinates=coords,
prisms=model,
density=rho,
field="g_xx")
gyy = prism.grav(coordinates=coords,
prisms=model,
density=rho,
field="g_yy")
gzz = prism.grav(coordinates=coords,
prisms=model,
density=rho,
field="g_zz")
npt.assert_almost_equal(gxx + gyy + gzz,
np.zeros(coords.shape[1]),
decimal=12)
def test_invalid_prism():
"Check if passing an invalid prism raises an error"
density = np.array([1000])
coordinates = np.array([[0], [0], [0]])
field = "g_potential"
# shape (1,)
model = np.array([100, -100, -100, 100, 100, 200])
with pytest.raises(ValueError):
prism.grav(coordinates, model, density, field="g_potential")
# shape (2,4)
model = np.empty((2, 4))
with pytest.raises(ValueError):
prism.grav(coordinates, model, density, field="g_x")
# shape (1,4)
model = np.empty((1, 5))
with pytest.raises(ValueError):
prism.grav(coordinates, model, density, field="g_z")
def test_invalid_prism_boundaries():
"Check if passing an invalid prism boundaries raises an error"
density = np.array([1000])
magnetization = np.array([[1], [1], [1]])
coordinates = np.array([[0], [0], [0]])
# wrong x boundaries
model = np.array([[100, -100, -100, 100, 100, 200]])
with pytest.raises(ValueError):
prism.grav(coordinates, model, density, field="g_x")
prism.mag(coordinates, model, magnetization, field="b_y")
# wrong y boundaries
model = np.array([[-100, 100, 100, -100, 100, 200]])
with pytest.raises(ValueError):
prism.grav(coordinates, model, density, field="g_z")
prism.mag(coordinates, model, magnetization, field="b_z")
# wrong z boundaries
model = np.array([[-100, 100, -100, 100, 200, 100]])
with pytest.raises(ValueError):
prism.grav(coordinates, model, density, field="g_potential")
prism.mag(coordinates, model, magnetization, field="b_x")
def test_field_decreases_with_distance():
"Check if field decreases with distance"
model = np.array([[-100, 100, -100, 100, 100, 200]])
density = np.array([1000])
close = np.array([[0], [20], [0]])
far = np.array([[0], [20], [-100]])
# potentia
potential_close = prism.grav(close, model, density, field="g_potential")
potential_far = prism.grav(far, model, density, field="g_potential")
# gz
gz_close = prism.grav(close, model, density, field="g_z")
gz_far = prism.grav(far, model, density, field="g_z")
# gx
gx_close = prism.grav(close, model, density, field="g_x")
gx_far = prism.grav(far, model, density, field="g_x")
diffs = np.array([
np.abs(potential_far) < np.abs(potential_close),
np.abs(gz_far) < np.abs(gz_close),
np.abs(gx_far) < np.abs(gx_close)
])
npt.assert_allclose(diffs, np.ones((3, 1), dtype=bool))