Jump to content

Nahid3007

Members
  • Content Count

    13
  • Joined

  • Last visited

1 Follower

About Nahid3007

  • Rank
    Beginner

Profile Information

  • Gender
    Male
  • Country
    Germany
  • Are you University user?
    Yes

Recent Profile Visitors

1249 profile views
  1. Hi, so do you have any updates whether OptiStruct uses the Optimality criteria method as mentioned in 2017 User’s Guide? Thank you!
  2. Hi, Yes I have already noticed that. But I am still confused whether OptiStuct generally uses the Optimality criteria method or not as mentioned in 2017 User’s Guide (see attached figure)? Regards Nahid
  3. Hello community, I got a little bit confused while reading the OptiStruct 2017 User’s Guide. There I found that OptiStruct selects automatically the best optimization algorithm. Five items have been listed there (page 10-11): - Optimality criteria method - Convex approximation method - Method of feasible directions - Sequential quadratic programming - Advanced approximations . My question is: Does OptiStruct still selects the Optimality criteria method automatically? And for which type of optimization? There is no information about it in the User’s Guide or Reference Guide. I want to optimize the topology of the 2 dimensional MMB-Beam (see atteched figure). My objective is to minimize compliance subjected to a volume fraction constraint. I found that for my type of problem I shoud use 'OPTMETH=DUAL' since I have a large number of design variable and only one constraint. So I cannot choose the Optimality criteria method in topology optimization? Many thanks in advance Nahid
  4. Hello Community, how can I set up a smooth contact for contact TYPE7. Thank you in advance.
  5. Hello Community, how can I prevent my implicit simulation to diverge (see attached figures of the engine OUT file)? There is a sentence in the RADIOSS v13 User Guide: "For the hyper-elastic type laws (42, 62, 69, and 82), special treatments have been done to avoid the divergence." I have a problem similar to “Example 42: Rubber Ring” with my own model (also hyper-elastic material - MAT42). The settings in the engine file for my model are almost the same to Ex.: 42.
  6. Hello community, I have a problem in RADIOSS with an explicit simulation of a hydro bush where I try to find the basic stiffness of the bush. I ran the simulation with the original model and the simulation seems to never terminate because the time step drops to a very small value and the run time is increasing. So I decided to simplify the problem I am concerning with and try to solve the problem (more or less by try and error at the moment). In FIGURE_1 (attached) I have run three simulations with my simplified models. The simulation in the figure on the left hand side is really close to the problem I have with the hydro bush (more or less). It is a rubber to rubber material impact. So I have used MAT42 (hyperelastic material model) and I have used the contact TYPE 7. The cylindrical material is being pushed with a concentrated force to the wall for the impact. The total run time is 1 second and I have used the DT/NODA/CST/0 to increase DT. The first simulation in the figure on the left hand side was killed because it has reached the limit for its total mass error and the energy error. The second simulation in the figure in the middle was slightly chanced but with no changes in the engine file. Here, the run terminates (without issues) but the value for the energy error was still high and as you can see there are some distorted elements. In the last simulation in the figure on the right hand side I made the wall thicker and still did not change the commands in the engine file. Here, the run terminates properly and the energy error is also good. I am not allowed to change the model properties but this is the result I want to achieve with the simulation one and two. How can I improve simulation one and two? Could it be a problem in the contact definition or some mistakes in the property definition? I will attach the starter and engine file as well as the engine OUT file of simulation one and two. I would be really great if I can get some useful hints on how to improve the problem, because I am really new to RADIOSS. Thank you very much in advance.
  7. OK ... it also works with POLAD1 if I enter the right magnitude^^. It has to be 2500 instead of 10 for POLAD1. Then I will get the exact same solution.
  8. Aaaaaaahhhhhhhh, finally I solved this problem . I do not know why but the PLOAD1 is wrong somehow (this is really bitter ). I changed the distributed line load q into nodal forces (see attached figure). Now I get my expected results. The FE results match perfectly with my analytical results. Nevertheless, thank you for the efforts tinh.
  9. First of all, thank you very much for the approach tinh. I tried to use this. Since I only want to release the dof around the z-axis, I have pinned the nodes as depicted in the attached figure. I set the orientation to y-axis. But I do not get my expected results. Did I understand something incorrectly? Thanks in advance, again. Nahid
  10. Hello community, I have to do a beam analysis (see figure_1) but I don't know how to create a pin-joint in HyperMesh. I have tried to use the joints panel in the 1D main menu page. Then I have selected the revolution joints type, because this joint type will constraint 5 dofs and will leave one rotational dof open. This is exactly what I want. I have selected the two terminal nodes [Grids 5 and 6 (1000;0;0)] and switched first orientation vector to z-axis (figure_2). But it did not work because when solving the problem there is an error message: *** WARNING # 312 In static load case 1 the compliance is negative or large 2.545087E+11 Optimization/buckling analysis cannot be performed due to possible rigid body mode What did I do wrong? Or is this not the right way to create such a pin-joint for this type of analysis? Thanks in advance Nahid
  11. Thank you very very much, tinh and Rahul! Both approaches worked very well. Best regards Nahid
  12. Hello community, I have a question concerning to apply loads and constraints on 3D surfaces. I have created a solid model from a 2D surface. If I do a 3D multi solid mesh and apply the constraints and the pressure onto the two surfaces (see figure 1), I will get a result. And I can see that every node on the surface is constrained and all elements on the surface are applied with a pressure force. If I create a 2D mesh first and drag the elements along the thickness of the model for a 3D mesh and then apply the constraints and pressure onto the surfaces, I will get no results. And further I can see that not all nodes on the surface are constrained and not all elements on the top surface are applied with a pressure force although I have selected the two surfaces (see figure 2). Why can't I apply the constraints and pressure on surfaces if I create a 3D mesh from a 2D mesh? And why is it possible if I mesh the model with the multi solid subpanel? Thanks in advance.
×
×
  • Create New...