Open LuisPantoja opened 1 year ago
Hello Luis,
a) here is a list of tools that support DCP https://dcp-standard.org/tools/. I'm sure others support it in an experimental fashion. Also I know of a proof-of-concept implementation in Simulink: https://doi.org/10.1109/ACCESS.2023.3237075. b) DCP was designed to be as close to FMI for this exact purpose. One can use the DCPLib to implement such a proxy to make an FMU available via DCP (I have done this using the FMILIB, simply calling the FMI-functions whenever the corresponding DCP-functions are executed) c) Correct. If everything is implemented in as a DCP there is no need for FMI. DCP is and independent standard that takes inspiration from FMI. However FMI is very broadly adopted so it's not going anywhere :)
Kind regards, Clemens
Thank you Clemens for the explanation, I agree with your comments ( and as a user of this technology, I should prefer an evolution of the current comercial tools using FMI to DCP standard. After reading DCP standar , in my opinion it is really a good standard :), good job! )
This is just some questions, because we need to take a decision about the implementation of distributed FMI using a comercial simulation tool like Labview / Simulik / Vector / Siemens ... .
a) Any of this tools or comercial ones supports this new standard? b) Is there any proxy (for running in our embedded device) from DCP to FMU, so we can have our current FMU running in a remote computer? c) If every FMU is modelled using DCP (so it can be executed in the same computer or in a remote one), the FMU interface is not neccesary, isn´t it? (so the DCP can be seem as an evolution of FMI and overwritte the previous FMI definition)
Many, many thanks in advance,
Luis