TAGGED: ansysaqwa, ansysaqwa, aqwa, aqwaerrors, aqwasuite, aqwareader, jointsansysaqwa


October 25, 2024 at 11:59 amS.S.S.ShankaVasukiSubscriber
Hi,
I am working on a floating problem with 15 connected bodies as shown in the figure. The dimensions of the structure is 1.88x1.88x0.14m and the draft is 0.096m. The mesh size was set to 0.08m with 20,000 panels.Â
Kxx = 0.48m, Kyy = 0.48m, Kxx = 0.62m, CoG = 0.03m
Hydrodynamic diffraction analysis: To run the diffraction analysis, I specified the connection stiffness matrix between each body (a total of 21 matrices) as shown in the figure below. I have not included the mooring stiffness matrix in this calculation and the simulation runs without errors BUT the results dont match the reference results as shown below for the pitch RAO for all the middle blocks. I dont know where I am going wrong here. As even if I include the mooring line stiffness, the RAOs will reduce even further which still does not match the reference values. Also, I have not included any additional damping matrices as well. So can you please let me know where the mistake is?
Hydrodynamic Response: For the time domain analysis, I replaced all double hinges (as shown in the first image) with a ball and socket joint as shown in the figure below to avoid the closed loop error along with a soft mooring line with the properties as shown below. I used an irregular wave spectrum with a JONSWAP spectrum with properties as shown in the last image.Â With all these definitions, the time domain solution tends to diverge about RZYaw (as shown in the last image), I also tried to reduce the time step to 0.001s, but no luck. Hence could you please let me know on how to resolve this issue as I have been asking this question from a month, but have not received any response from the Ansys team.
Â

October 30, 2024 at 4:38 pm

October 30, 2024 at 5:05 pmS.S.S.ShankaVasukiSubscriber
Thank you for the reply.
I read the thread you attached. But could not quite understand it well, hence could you please explain it and how I can incorporate that in my simulation?
Also, I dont know why I am unable to run the hydrodynamic response analysis. because It says solution diverging as shown in the image above for the given JONSWAP spectrum.I checked the mass and inertial properties and they seem to be okay.Â

October 30, 2024 at 6:33 pmShuangxing DuAnsys Employee
Will you need to define more connection stiffness 6x6 submatrices?Â For example, from the linkages of the first row of 5 structures, you may need
(1A,1A), (1A,2A), (2A,2A), (2A,3A), (3A,3A), (3A,4A), (4A,4A), (4A,5A), (5A,5A) (9 6X6 submatrices for the first row). 27 6x6 submatrices for all 3 rows.
For the first column's additional submatrices due to the linkages, you should have (1A, 1B), (1B,1C) 6X6 submatrices. 10 additional submatrices for all the columns.
In total, you may have 37 6x6 submatrices.
Will the fixed point to structure mooring lines represent the joint condition well?Â Â It may be nice to use the connection stiffness matrices directly.Â You could use a presolve script to remove NASF option (Not use the additional structure stiffness matrix) from the hydrodynamic response analysis input data file. Alternative, switch on the beta option,
Â Then in Hydrodyanmci response analsyis, set Include Linear Connection Stiffness Matrices as Yes. It will include the connection stiffness if it is defined.

October 30, 2024 at 6:42 pmShuangxing DuAnsys Employee
Note that the diagonal 6x6 submatrices may be the assebles of submatrices from all the linkages.Â For example the 6x6 submatrix for structure 3B to Structure 3B, the total assembled 6x6 matrix should be the summation of the contributions from the linkages between 3A and 3B, 3B and 3C, 2B and 3B, 3B and 4B.

October 30, 2024 at 7:14 pmS.S.S.ShankaVasukiSubscriber
So if I understand correctly, I need to have a total 37 connection stiffness matrices for the joints and two fixed "equilibrium" connection stiffness matrices for the mooring lines right? I have made an image which shows all the connection matrices I need to define in my simulation as highlighted in yellow. Could you please let me know if this is correct?
Also I have some questions on your next comment:
 You mentioned  "Note that the diagonal 6x6 submatrices may be the assembles of submatrices from all the linkages" ... I could not quite understand this. Because the reference article I am referring to gives the matrices as shown in the figure below. So should I give all the 37 connection stiffness matrix (excluding the mooring lines) as a diagonal matrix with C11,C22,...C66?
2. If I use the beta options and include the connection stiffness matrices for the hydrodynamic response analysis, do I still need to define joints such as hinge joints or ball or socket? And do I have enable a linear cable connection for the mooring lines?Â
Please help me out with this as I have been trying a lot of things which have not been working.

October 31, 2024 at 9:25 amShuangxing DuAnsys Employee
(1) Only 37 6x6 submatrices are added. The matrices due to moorling lines need to be added in the diagonal 6x6 matrices of M3a3a and M3c3c respectively.Â Please read the finite element method book for the global stiffness assemble from the element stiffness submatrices. Each connection and mooring stiffness matrix could be considered as a stiffness matrix of a structural element, and needs to be assembled into the global stiffness matrix, which is the connection stiffness required by the Hydrodynamic Diffraction and Hydrodynamic Response analyses.
(2) If the connection stiffness matrices have been used, there is no need to define any joint/mooring, otherwise they will be duplicated.Â

October 31, 2024 at 12:03 pmS.S.S.ShankaVasukiSubscriber
I tried including the 37 matrices as you suggested and here are the results:
Hydrodynamic diffraction: The figure below shows the pitch RAO (deg/m) as a function of frequncy (rad/s) for the middle section of the bodies (1b,2b,3b,4b,5b). I tried running the diffraction analysis with the 37 connection matrices and without the connection matrices. The response of the system with 37 connection stiffness matrices is lower than when there are no connection matrices mentioned which is suprising. Because just with the mass and inertial properties of all the 15 bodies, the results are still lower than the reference values. Also when i add the structure motion amplitude, the motion looks weird as shown in the second image below. So I dont know what the error here. Could it be the zlocation of CoG? The reference mention the CoG from keel to be 0.1m, hence I calculated the CoG from the waterline to be 0.03m. I hope this is correct?
So could you help me with understanding whats going wrong here?
Hydrodynamic Response: Now, for the response analysis, I turned on the beta options and included the connection stiffness matrix in the time domain analysis and I get the same error saying  Â HYDRODYNAMICALLY INTERACTING STRUCTURES HAVE MOVED TOO FAR FROM ANALYSIS POSITION MOVEMENT EXCEEDS 75% STOPPED AT TIME = 0.50. as shown in the figure below. So I am not sure whats going wrong. Because I have double checked everything and it seems to be okay.

November 1, 2024 at 9:14 amShuangxing DuAnsys Employee
(1) Please check your reference article on the defintion of the reference degrees of freedoms for the stiffness matrix. Is the stiffness with respect to the motions at the ends of the connection points or with respect to the COGs of connecting structures?
(2) Please check the sign of the coefficients of the offdiagonal 6x6 matrices. I expect the values to be negative.

November 1, 2024 at 9:29 amS.S.S.ShankaVasukiSubscriber
Hi,
I checked the above two points you mentioned in the reference article and it seems they are not specified the reference DoF for the stiffness matrix. All they have provided is this matrix. Also this is link to the reference article: https://www.mdpi.com/19961073/17/9/2059
Â
But one question that is very weird is when I dont specify any connection stiffness matrix and I run the diffraction analysis just with the mass and inertial properties of the 15 bodies, the RAOs are still lower than the reference values. So does this mean that the mass and inertial properties are wrong??

November 1, 2024 at 9:45 amShuangxing DuAnsys Employee
No comment. Please check your model.

November 1, 2024 at 9:51 amS.S.S.ShankaVasukiSubscriber
Â
The comment is post checking my model. If my model was working I dont think I would be asking a question on the forum :)
I'm just trying to understand whats wrong in my model.
Â

 You must be logged in to reply to this topic.
 RSS is disabled.
 Speed up simulation in HFSS
 ANSYS Mechanical won’t open and stays on Blank Gray Screen (Path environment va)
 ANSYS DISPLAY RESOLUTION PROBLEM
 ansys fluent error when opening it “unexpected license problem”
 Workbench license error
 Ansys 2024 R1 â€“ The installation completed with warnings/errors.
 Could not connect to a valid licensing server
 LSDyna Beam elements outputs
 Not find helic___[]_.tgz file

771

407

264

201

162
Â© 2024 Copyright ANSYS, Inc. All rights reserved.