[TRNSYS-users] Discrepancies Between Results from TRNSYS Simulation and GenOpt TRNSYS Simulation

Subject: [TRNSYS-users] Discrepancies Between Results from TRNSYS Simulation and GenOpt TRNSYS Simulation

From: Jason Ng Cheng Hin <jason.ngchenghin@gmail.com>

Date: Mon, 29 Aug 2011 14:57:42 -0400

Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:date:message-id:subject:from:to:content-type; bh=F0RU2NTqswLkSc/4TdERo6eA8TcbY8OOmhYyh/o/Po4=; b=kcwi+GhKpTuO0BondUdny2EBB9AR2ThWxjlAj1hn/3f65ENKELwdPlIhaoAyssZzKp jnNFfw755UiwbcQf3gsf9D4k5T1d3EBkW3la+ZapFGIma4lI3lG4Av8HTi3+K3drIid2 T6nNMDWWqiFjMPlvgMHZSKFiS/fWFD6ObfVKg=

List-archive: <http://mailman.cae.wisc.edu/pipermail/trnsys-users>

List-help: <mailto:trnsys-users-request@cae.wisc.edu?subject=help>

List-id: "TRNSYS users mailing list at the Solar Energy Lab, UW-Madison" <trnsys-users.cae.wisc.edu>

List-post: <mailto:trnsys-users@cae.wisc.edu>

List-subscribe: <https://mailman.cae.wisc.edu/listinfo/trnsys-users>, <mailto:trnsys-users-request@cae.wisc.edu?subject=subscribe>

List-unsubscribe: <https://mailman.cae.wisc.edu/options/trnsys-users>, <mailto:trnsys-users-request@cae.wisc.edu?subject=unsubscribe>

I am running some optimizations using TRNSYS 16 and GenOpt 3.0.3. My optimizations have completed successfully and I had no troubles performing them.

However, when I try going back to the original dck file from which I made the copies for the GenOpt template and run files and input all the values obtained from the optimizations, the results I obtain are noticeably different.

I have run the original dck file and the run file produced by GenOpt with identical values for each variable through a comparing tool and the files are identical except for the ASSIGN statements (which simply point to different folders that contain the same files). I have also run all of the external files that the simulations call (building files, storage tank files, soil.in for Type 701, and schedules) and they are all identical.

I have run out of ideas to determine where the discrepancies between the two results come from. From what I understand, since the dck files are essentially identical they should produce identical results. I am beginning to think there are other factors which I am not aware of that can affect the results of simulations run through GenOpt.

Any ideas?

Jason