- if it is a continous time state system the xcos solver is able to automatically adapt its step size so, it is not necessary to handle two copies of your superblock - if it is a discret time system you can use two SampleCLK block one with 1/10 period and the other one with 1/1000 period as explained in the attached xcos diagram
Serge Steer Le 21/07/2014 08:45, Quentin Mazué a écrit :
Hello, I need to model a special phenomenon which has two different states: ->1) A transient state: from 0 to 1 sec. For this phase I need calculations with a very small time step (around 0.001s) ->2) Then a steady state: from 1 to end of simulation (around 10-15sec) For this phase I need calculations with a large time step (0.1s) Enclosed, a picture of an equivalent Simulink diagram which matches my needs.Illustration_EnabledSuperblocks_MultifreqDiagram.png<http://mailinglists.scilab.org/file/n4030940/Illustration_EnabledSuperblocks_MultifreqDiagram.png> Both phases have the same inputs calculated with the same block. I would like to: ->a) Specify that I have 2 superblocks within the same Xcos diagram that make the computations with different time step and a third one with a time step which change during the simulation (the Input superblock) ->b) be able to deactivate the transient superblock when I don’t need transient calculations and activate the steady block only when I need it. Why I would like to do that? ->I’m working with a new way of computing a phenomenon and I have to show that this new method is faster than the previous method and enough accurate. If I don’t deactivate the transient block, Xcos is still running this block and I lose all the interest of my new model. ->at the same time, I have to show that Scilab/Xocs is able to match the need of my company. They are actually using a lot Matlab/Simulink, especially the blocks: enabled subsystem, triggered subsystem, trigger, zero order hold, rate transition,… Thanks for your help Quentin Mazué -- View this message in context: http://mailinglists.scilab.org/Xocs-enabled-superblocks-and-multi-frequence-diagram-tp4030940.html Sent from the Scilab users - Mailing Lists Archives mailing list archive at Nabble.com. _______________________________________________ users mailing list users@lists.scilab.org http://lists.scilab.org/mailman/listinfo/users
Mazue.zcos
Description: Binary data
_______________________________________________ users mailing list users@lists.scilab.org http://lists.scilab.org/mailman/listinfo/users