is there such a thing as a users guide for DOEs and especially for the post processing opportunities?

I am using the software a few days and I'm wondering what all the post processing monitors show and how the numbers are calculated. Can you help me to understand these thing a bit better?

Cheers Max

]]>

As I try to do Hyperstudy setup for parametrized model,

after the nominal run. The out file is created and I can see the stress results there but as I want to create an output response,

The responses can not be read by hyperstudy results reader from the out file in approaches>m1> run01 only text extraction appears.

What can be the problem?

Thank you

]]>

is it possible to define and save a result (graph) in the output response window as a vector with all the given points instead of just evaluate one result for a specific point.

For example: I have the given curve as shown in the picture. Now i want to save this curve or graph as a vector or in a similar format. With the tutorials I can only read out one specific value of the curve at one given point. But my aspiration is it to save the whole curve, to display it or use it for a vectorial calculation.

Do you have any advise how i can do this with hyperstudy? Is it possible to do this in the define output response window?

Thanks in advance and cheers

]]>

I am interested in setting up a linked multi-model run in hyperstudy, where the response of the first model could violate a constraint and thus signify no need to run the second model. Is there a way to do this? If I were to wrap my second model solver execution (e.g. radioss) in a templex or compose script, is there a way to access the variables within hyperstudy (i.e. output response from first model)? This would allow me to apply logic on whether to run the second model.

Thanks,

Jacob

]]>

I ran into a problem recently. Whenever trying to define a discrete variable in the negative range ... e.g. -10 to -1 with stepSize = 1 (which is: -10,-9,-8,-7,-6,-5,-4,-3,-2,-1), I get an exception with the data type Real. Is this a bug?

Best regards

Merula

]]>

Can someone please help me???

I uploaded the data that hyperstudy gives error '(For more information see)

]]>When I follow the path step by step at the output response step, the problems is the output file is not recognizable by HyperWorks. What is the problem? because I need to define volume/mass as output response.

]]>Thank you.

Vinit

]]>

I am a new user of Hyperstudy and I want to do a CFD shape optimization coupling HyperMorph and AcuSolve solver through Hyperstudy.

In my model I have only one shape variable. I get a nominal solution for the velocity and pressure field from Hyperstudy for my initial geometry (mesh). My question is, can I use this solution as an **initialization** for the following analyses that Hyperstudy performs in order to get the optimum one? I suppose that in this way a more robust and fast method for the alternative design analysis could be implemented, since solving every possible flow field from scratch (with no guessing about the initial conditions) is avoided. Is there an automatic way to instruct Hyperstudy to use the previously acquainted solution as initial conditions for the next analysis? Should it be down through scripting ?

Have I got it right or there is something that I am missing..?

Thanks in advance!

Cheers,

Alex

]]>

I urgently need your help. I have to do an optimisation of a B-Pillar which i have already simulated in Radioss according to 3 point Bending.

This was my first time with hyperstudy so i guess i did many thing wrong by defining output responces and contraints i guess.

The aim is to hold the intrusion level same and minimize the mass of the structure.

I would be really appriciated if someone could help me.

i ll post some screenshots here because somehow i cant save the hyperstudy file. Actually i can save but cant open again. It gives error.

I just added the size parameters. Shape parameters i will add them later when i understand how hyperstudy works exactly

After nominal run i tried to evalute the study

But i get this error :

Full folder is in the attachments. I need urgent help to finish master thesis.

Thanks in advance

]]>

I would be really appriciated if someone can answer my question. I dont know if it is possible in Hyperworks to do implicit parametrisation.

I am currently working on a cross section shape optimization of a b Pillar.

This b pillar consists of 3 metal sheet. Outer inner and the reinforcement. Reinforcement should go up and down in the y direction

Outer and inner pillar should be parametrised in x direction. I have already made required morphing and saved them as shape. This is called as an explicit shape parametrisation while the geometrical conflict and overlapping can only be prevented by having a smaller margin of movement. In order words when i move the reinforcement more than 12mm outside pillar and reinforcement will cause an overlapping. Is there any way to control this?

When the outer pillars are wider there is whole bunch of space that reinforcement in y direction to move. But if move it than there will be overlapping when it is narrower. Such as in the picture belows. Hope i could explain my problem.

]]>

Regards

Ruben

]]>I am studying in crash box optimation, I set up the model as following figure, this model is deformed until reach incompressible condition (To observe total crushing performance), as a result the force goes up considerably after 120mm.

However, I need to consider the data between 0 - 120 mm (78% of total displacement), but I can not find the ** Expression** for cut the data in HyperStudy.

According to the figure below, that show you the Force and Displacement collected from T01 file, how can I trim the data range from 0 - 120mm using some **Functions.**

Below this, there are output responses that already create but need to select the appropiate range first.

If you need further information please do not hersitate to let me know.

Thank you for your help

Best regards,

Nut

]]>

I'm doing optimization using hyperstudy and want to obtain kriging interpolations. I cannot find the respond surface option in 2017.2. edition . it was available in version 12 (see figure )

]]>I trt to do spotweld optimization( location) on crash box using hypermorph and hyperstudy . İ change the location of spotweld domains . And ı see its location change in structure. But when ı run the optimization in hypersrudy it give me failure after 2. Run. İs it possible do spotweld location optimization using hypermorph and hyperstudy?

]]>

I want to optimize the location of the trigger in the crashbox. When I changed the trigger position with the handles, the mesh changes. does the mesh automatically correct in each solution (50 iteration) or does it stay the same during the solution?

]]>

I am working on optimization the vehicle crash box, I have some question about Hyperstudy / Hypermorph coupling for shape optimization.

As first figure below, I created 2 Shapes (yellow and green) by translating the handles in -X,+X direction to increase the taper angle.

In HyperStudy, I defined the model type as HyperMesh, while I am importing input variables from HyperMesh, they ask me to identify the lower and upper bound once (as second figure below).

In this study, I would like to have the design variables including** crash box thickness, shape1 and shape 2.**

**Question**

1. I seperated the shape into 2 because I have no idear to handle with them, can there cause any problem further?

2. If I want to keep the handle distance (+10,-10 in x direction) which is defined in the HyperMesh, Should I set up the input variables as following?

__ Variable 1 (dv_1)__ Lower= 0, Nominal = 0, Upper = 10

__ Variable 2 (dv_2)__ Lower= 0, Nominal = 0, Upper = -10

Are there correct?

Thank you for your calification.

Best regards,

Wisawanart

]]>

I am setting up the analysis of the crashbox for optimization.

I have some questions regarding the Hyperstudy,

1. As figure below, can I set the position of those handles (red points) as a design variables because I need to find out the optimal taper angle. When I looked into the .tpl file, there cannot be adjusted them while shell thickness can be specified the lower upper bound directly.

If impossible, could you suggest me the similar ways to carry out this task.

2. When I __Evaluate Tasks__ for both **nominal run** and **optimization**. I think, the software is trying to solve the results but unfortunately when I look into the CPU load, it just 9% as same as the memory, completely different from usual situations. Normally, I define number of CPU cores as 24 (-nt 24) in RADIOSS.

There have something wrong? or how I can define the core number in Hyperstudy? it is very slow now even a small problem.

thank you for your help,

best regards

Wisawanart

]]>i want to do that crossbar can slide through frame. I defined morph volume. But i don't know how can i connect desvar and morph volume?

Is it right way? i have seen just size/shape configuration, so it's not my way.

i had tryed to save volume and looked for txt file. it doesn't consist elements, so i can't use it in hyperstudy.

could anyone explain me how does it works?=)

Thanks.

]]>
There the presenter is using an equation to solve the optimization problem (ref. In video time of eqn is 41:49) i.e. { sum((v_1 + 0.01*v_2)^{2} } where v_1 is Actual Toe & v_2 is Wheel vertical displacement.

Could you please explain how did he decide to optimize suspension geometry based upon this equation?

Ref video:

]]>

I'm trying to use HyperStudy to configure postures of an idustrial robot and subsequently calculate the eigenfrequencies of the structure. I have a HyperMesh model with all the robot axis divided into components and tcl scripting files which can rotate the components when I run it in HyperMesh.

My problem that I face now is how do I call a TCL-script that rotates the components and therefore also the mesh with respect to an input parameter and subsequently Optistruct proceeds with its calcuations? Or is there a more convenient way to solve this problem?

Best regards

]]>

I am a new user in HyperStudy,

I faced with the error in creating process of output response, the f(x) is shown as the figure below.

Previously, I tried to fill in ** ((dv_2)−(5*(dv_1)^2)/(4*3.142^2)+5*dv_1/3.142−6)^2+10*(1−1/8*3.142)*cos(dv_1)+10; **(when X1=dv_1, X2=dv_2)

but it is not working.

Please suggest me how to put such kind of equation in Hyperstudy.

Best regards

]]>

I imported model from Hypermesh. For output response in setup, one response is normal stress of one part of my model. How can i choose a specific component's stress value as my output response?

Thank you

Alan

]]>