NCR9026
Aerospace
- Dec 4, 2007
- 12
Hi, I'm relatively new to ANSYS and I'm looking for advice on calculating the forced, periodic response of a nonlinear structure. I'm using ANSYS 9.0.
Specifically, I'm dealing with the inertial loading of a flapping wing subject to periodic imposed flapping motion (via OMEGA, DOMEGA, and ACEL loads). I want to find the steady-state periodic deformations of the structure over one period. The structure will be undergoing large strains, so I can't use a linear or harmonic analysis. What I'd planned to do is run a transient analysis from steady state until the initial transients damp out. I expect this to take a minimum of 30 periods to be sure the transients are gone.
I guess the naive way to do this is to define 30+ periods with of load steps and let the program run. But this would produce an unwieldy number of load step files, most of which would be redundant (and since the time-steps are small for convergence, I'm afraid of running up against load step limits). Furthermore, I don't care about the time histories prior to the last period: I just need the state at the end of one period as the initial state for the next period.
What I'd like is to define N load steps over a single period and reuse them. Is there a command in ANSYS that will let me loop the load steps like that? Could I use macros or tables to automate this process? If so, do I have to continually reset the time back to t=0 at the beginning of every new period? Are there any (dis)advantages to a particular method?
Thank you.
Specifically, I'm dealing with the inertial loading of a flapping wing subject to periodic imposed flapping motion (via OMEGA, DOMEGA, and ACEL loads). I want to find the steady-state periodic deformations of the structure over one period. The structure will be undergoing large strains, so I can't use a linear or harmonic analysis. What I'd planned to do is run a transient analysis from steady state until the initial transients damp out. I expect this to take a minimum of 30 periods to be sure the transients are gone.
I guess the naive way to do this is to define 30+ periods with of load steps and let the program run. But this would produce an unwieldy number of load step files, most of which would be redundant (and since the time-steps are small for convergence, I'm afraid of running up against load step limits). Furthermore, I don't care about the time histories prior to the last period: I just need the state at the end of one period as the initial state for the next period.
What I'd like is to define N load steps over a single period and reuse them. Is there a command in ANSYS that will let me loop the load steps like that? Could I use macros or tables to automate this process? If so, do I have to continually reset the time back to t=0 at the beginning of every new period? Are there any (dis)advantages to a particular method?
Thank you.