Hi, I have been looking into the possibilities for RFEM6 to be able to simulate a time-series of wind pressure around a cylinder at a specific geospatial location (with surrounding buildings).
For each time step the data has a polar coordinate (radii, radian), pressure and z-coordinate.
Due to the simulation the wind pressure has a highly complex shape along the perimeter and height.
Is it possible to, in any combination of time series/imperfection/pushover settings to produce this wind time series loading without too much error margin?
The goal is to try and simulate oscilation due to shifts of relative wind pressure.
Can I instead import wind data series/replicate it in RWIND and then implement it through the RWIND-RFEM interface?
Edit: I was wondering if it could be created by having each time-step as a quasi-static load representation with “free rectangular load” then using that time-step with “Consider initial state from” as the start of the new time-step. and then afterwards create a simulation of the output?
You can use our add-on Time History Analysis to calculate the structure's reaction to a time-dependent loading. In your case this could be done following these steps:
Create a load case for every data point
Create a load object in every load case at the according location in the model with a basic load value (e.g. 1 kN)
Create a time-load-factor-diagram for every data point and assign it to the corresponding load case.
Imperfections cannot be applied directly to a time-history analysis but may be considered by pre-deforming the structure.
Can you please further explain what you mean by pushover? A pushover analysis of a single time-step?
Currently, it is not possible to automatically couple a transient wind simulation in RWIND 3 and a time history analysis in RFEM 6. This can be done manually to a certain extent, by running a transient simulation in RWIND 3 and transferring the results manually to RFEM. But as RWIND 3 currently has no API this can only partly automated.
Theoretically, this would also be possible. But I think using the time history add-on is more convenient.
I hope this helps. If you have any further questions, feel free to ask
So #load cases are dependent on the discretization of the free rectangular load to simulate an approximated continuous circular load; the same is applicable for the height intervals: #Load cases = 360/6 = 60 (6 degree intervals) #Height Intervals = between 5 to 10
In essence 300 to 600 (or more) load cases of a unit load; and then create the corresponding time diagram for each load. It sounds doable
I agree; I am slowly starting with implementation of the API as our company just recently acquired it. Before starting I wanted to know the interconnection to the time-history analysis to gain insight to a probable solution for better planning the software implementation
The reason above, I have not had the need to use the Pushover analysis yet, so I hoped to gain insight into planning a probable implementation for the workflow.
With your above proposed solution I also think that is the way to go.
Thank you very much Clemens, that was very helpful!
This procedure is shown in the following webinars:
As this can a very elaborate work depending on the number of data points, I recommend you to take a look at our Dlubal API that enables you to automate this process.
You can check the current capabilities in the API documentation: 