Hello Eddy3d community,
I am new to Eddy3d and am currently working on an outdoor UTCI simulation. I’ve encountered two persistent issues while running Eddy3d for testing, and I’m hoping someone here might be able to assist me. Thanks for anyone who takes the time to look into my problem. Here are the details of the issues I’m facing:
1. Multi-core parallel computation error: I have followed the instructions on the forum ([Other] blueCFD-Core-2016 user compiled solvers not running in parallel – CFD Online Discussion Forums), copied the msmpi.dll, and performed computations using multiple cores (set CPUs to -1) which seemed to run normally (from the “Simulation” and “Residuals” components, it appears to be normal). The “Probing” component also reads the wind speed simulation results correctly. However, when using the “Wprobe” component to read the results, I encounter an error. The error message is:
- Solution exception: An element of type EddyLib.Geometry.EddyVector was null; this might be due to contents in a list/array
Is this an issue with the “Wprobe” component itself? Is there any way to read the simulation results from multi-core computations and input them into the “MRT system” component for further calculations?
#2. Radiance computation error: I have three working computers, and I’ve installed the same versions of Eddy3D, Radiance, and Blue-CFD on all of them, ensuring they were installed in the same manner and path. All computers can compute wind speeds normally with 1 CPU core, and the “Probing” and “Wprobe” component can read the results as expected. However, when I proceed with MRT calculations, the problem arises. Radiance works correctly on one computer, but crashes on the other two. The crash occurs immediately when Radiance reaches the step “Perform an annual direct-only daylight coefficients simulation…”, with no UTCI.eddy or MRT.eddy files generated. I’ve noticed that on the two computers where the crash occurs, the “MRT system” component seems to be unable to read the “wind.eddy*” files correctly. Even when I disconnect all inputs at the CFD input end of the “MRT system” component, the same crash occurs. I’ve found a RadiationErrorLog.log in the calculation results folder, which records the error as:
“dircalc3 command failed with exit code 1: system - cannot open file ‘Rad\output\dcd_r1.mtx’: No such file or directory
: cannot load matrix”
What could be causing this issue?
I have attached the GH file I am using.
EDDY-Outdoor UTCI-template.gh (67.0 KB)
Current configurations:
- Windows 10 & 11
- Eddy3d version 0.4.8.0
- Rhino 8.8 SR8 (8.8.24163.12481)
- Radiance_938fd65c
- blueCFD-Core-2020-1-win64-setup