Hi James, You could call that a
communication bottle neck, but the slow speed mainly comes from the fact that
MATLAB has to be called (and executed) during each iteration. This implies a
huge overhead. An alternative is to
compile your MATLAB model into an S-function and create a DLL that can be used
like any TRNSYS type (adding the TRNSYS API to the generated source). I wrote a
MATLAB real time workshop template for the MSVC C++ compiler to automize this
from within MATLAB. For this to work, you need MATLAB real time workshop (expensive)
/ SIMULINK (expensive) and Microsoft Visual C++ 2003 or 2005 (cheap/free). Let me know if you want
to try it out (it’s a ‘beta’ ;-) ). Werner De :
James S. McNeill [mailto:James.Mcneill@Colorado.EDU] I am currently using the Type 155 to find roots of several
functions. There are 18 inputs and 8 outputs connected to the Type
155. The simulation requires almost 10 times longer to complete when I
have this model attached. Is there an communication bottleneck between
Matlab and TRNSYS? I am planning a large parametric study and need to
find ways to speed up the simulation. Thanks. James McNeill Dept. of Civil, Env., and Architectural Engineering Building Systems Program ************************************************************************************ This footnote confirms that this email message has been scanned by PineApp Mail-SeCure for the presence of malicious code, vandals & computer viruses. ************************************************************************************ ************************************************************************************ This footnote confirms that this email message has been scanned by PineApp Mail-SeCure for the presence of malicious code, vandals & computer viruses. ************************************************************************************ |