Dear all,
I am facing a problem while running the MHK RM1 glue code. I’m trying to obtain the deflections along the tower and have used multiple parameters to achieve this. However, most of the parameters in the OutListParameters sheet yielded zero deflection. The parameters that did provide deflection data were related to Tower-Top / Yaw Bearing Motions, specifically “YawBrTDxt, YawBrTDyt, YawBrTDzt” for deflection at the top of the tower. The issue I encountered is that the deflection in the x-direction, “YawBrTDxt,” is very high. Running the default input in the MHK RM1 code without any modifications results in an approximate 5 m deflection, as shown in the picture below.
What steps should I take to resolve this problem? Additionally, are these the correct parameters to use for obtaining deflection data? If I want to get the deflection at each node, what parameters should be used?
Thanks for your assistance.
1 Like
Hello Tareq,
Thank you for asking. I will do my best to answer your question.
A tower property of the RM1 turbine in the example was just made up and the large tower-top/yaw bearing deflection issue is related to a tower stiffness. In the current r-test file, we used an adjustment factor of 1E-3 for fore-aft and side-to-side stiffness. That makes tower structure too soft. This was to speed up simulation with enabling structure’s DOFs, specifically for a regression test purpose. I tested with the adjustment factor of 1 and smaller timestep size of 0.001s, the result looks reasonable now (attached screenshot).
.
In case you want to have a realistic tower design, you could refer to an original RM1 tower structure in the previous study (Page 77).
V. Neary, M. Previsic, R. Jepsen, M. Lawson, Y. Yu, A. Copping, A. Fontaine, K. Hallett, D. Murray (2014). Methodology for Design and Economic Analysis of Marine Energy Conversion (MEC) Technologies. SAND2014-9040
Cheers,
Toan Tran
1 Like
Thank you very much for your answer. Before asking in this forum, the first thing I tried was changing the adjustment factor to 1, even with a small time step of 0.001s. However, I encountered an issue where OpenFAST aborted. Should I change anything else when altering this parameter?
Hi Tareq,
Thank you for letting us know. I tested in old r-test file which had a slight different setting in ElastoDyn input. Specifically, DrTrDOF variable was disabled. I pulled the latest r-test branch and got the same instability error. However, I was able to run the example file with adjusted factor of 1 and a smaller timestep size of 0.0005s. As this relates to a tower stiffness, you could play around with timestep size to get a stability solution.
It is also noted that when DrTrDOF variable is enabled, mechanical properties of drivetrain like torsional stiffness (DTTorSpr) and damper (DTTorDmp) should be defined properly. Currently, they are made-up values.
BRs,
Toan Tran
1 Like
Thank you so much for answering my questions. I truly appreciate it.
