We should test time integration in simple test cases

[menon-karthik]

Description

The current tests which have pulsatile boundary conditions only compare the first time step with an expected/analytical solution. We should be testing them against analytical/semi-analytical solution at multiple time steps to make sure the integration is being tested.

Reproduction

Example for RCR BC test case:

svZeroDSolver/tests/test_solver.py

Lines 294 to 298 in 06d2f24

	def test_pulsatile_flow_r_rcr():
	results = run_test_case_by_name("pulsatileFlow_R_RCR")
	assert np.isclose(
	np.mean(np.array(results["pressure_in"][0])), 4630.973018148328, rtol=RTOL_PRES
	) # inlet pressure

Example for coronary BC test case:

svZeroDSolver/tests/test_solver.py

Lines 310 to 314 in 06d2f24

	def test_pulsatile_flow_r_coronary():
	results = run_test_case_by_name("pulsatileFlow_R_coronary")
	assert np.isclose(
	np.mean(np.array(results["pressure_in"][0])), 880.7264415091931, rtol=RTOL_PRES
	) # inlet pressure

In both the tests above, only timestep 0 is compared with the reference solution.

Expected behavior

For simple test cases, such as the RCR and coronary BC blocks, we should compare with semi-analytical solutions. We can write out the governing equations for specific quantities (inlet pressure, for example) and integrate them in time using simple Euler/RK schemes to compute a reference solution to compare against.

Additional context

This is related to #95

Code of Conduct

• I agree to follow this project's Code of Conduct and Contributing Guidelines