More precision needed in boundary I/O time.dat file

[lawrenceccheung]

If we're using the native format for the boundary I/O files, then it writes out an text file called time.dat to index the iteration numbers against the simulation time. However, if the simulation proceeds long enough, there isn't enough precision in the file to accommodate the exact times. This is what it looks like with dt=0.25 and t>10000:

60000 15000
60001 15000.2
60002 15000.5
60003 15000.8
60004 15001
60005 15001.2
60006 15001.5
60007 15001.8
60008 15002
60009 15002.2

This is not a huge problem, but there will be slight errors accumulating as it interpolates the boundary data against the wrong times.

The fix looks pretty straightforward, looks like we need to add a precision modifier to

amr-wind/amr-wind/wind_energy/ABLBoundaryPlane.cpp

Lines 537 to 539 in ef0a198

	std::ofstream oftime(m_time_file, std::ios::out | std::ios::app);
	oftime << t_step << ' ' << time << '\n';
	oftime.close();

Let me look and see if this type of problem appears anywhere else.

Lawrence

[github-actions]

This issue is stale because it has been open 30 days with no activity.

[github-actions]

This issue is stale because it has been open 30 days with no activity.

[marchdf]

Closed in #947