Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.
Time dependent simulation
Posted 2012年10月1日 GMT-4 00:02 Structural Mechanics 2 Replies
Please login with a confirmed email address before reporting spam
Dear Sir,
I am trying to model an energy harvester in which discrete pressures must be applied one after the other. The pressure applied is not continuous but varying values of pressure is applied one after the other in short succession.
On the whole , a varying values of pressure are applied to the energy harvester and it must produce varying values of voltage continuously , as long as pressure is applied.
I found that even 'Time Dependent' study does not give a provision to model such a situation. It would be of great help if you can guide me through the steps involved.
Thanking you in advance,
Vijay
I am trying to model an energy harvester in which discrete pressures must be applied one after the other. The pressure applied is not continuous but varying values of pressure is applied one after the other in short succession.
On the whole , a varying values of pressure are applied to the energy harvester and it must produce varying values of voltage continuously , as long as pressure is applied.
I found that even 'Time Dependent' study does not give a provision to model such a situation. It would be of great help if you can guide me through the steps involved.
Thanking you in advance,
Vijay
2 Replies Last Post 2012年10月1日 GMT-4 15:47
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hello,
You can implement some analytic but discontinuous functions, as:
deltap*floor(t/deltat), where deltap and deltat are increments in pressure and time, respectively, and t is the time;
1*(x>=0)*(x<=10)+5*(x>10)*(x<=15)+7*(x>15)*(x<=30), where the functions is 1 for x in (0,10], 5 for x in (10, 15] and 7 for x in (15,30].
Bye.
Jesus.
You can implement some analytic but discontinuous functions, as:
deltap*floor(t/deltat), where deltap and deltat are increments in pressure and time, respectively, and t is the time;
1*(x>=0)*(x<=10)+5*(x>10)*(x<=15)+7*(x>15)*(x<=30), where the functions is 1 for x in (0,10], 5 for x in (10, 15] and 7 for x in (15,30].
Bye.
Jesus.
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
yes bolean functions is one way, but they are not derivable so often the solver has problems to converge over the steps. You can also get steps by using the default step functions that are "smoothed such to make them derivable and continuous over the steps.
Steps also make a system ring if you have inertial terms on, in time series analysis you might also have conversion issues when using a too low damping
--
Good luck
Ivar
yes bolean functions is one way, but they are not derivable so often the solver has problems to converge over the steps. You can also get steps by using the default step functions that are "smoothed such to make them derivable and continuous over the steps.
Steps also make a system ring if you have inertial terms on, in time series analysis you might also have conversion issues when using a too low damping
--
Good luck
Ivar