About DesignEngLab/RBTO

[xiaomifengya]

I'm sorry to bother you again. When I run your code "RBTO" with Matlab, Matlab displays "unable to resolve the name codes.reliability.iform". I wonder if it's convenient for you to share the complete code. I really need your help. Thank you!

[pbtrung]

Hi. Sorry for the late reply. Have you been able to run the code? It requires importing the CODES toolbox as a prerequisite.
http://codes.arizona.edu/toolbox/help/html/CODES.html

[xiaomifengya]

Thank you for your reply. I read your paper carefully some time ago, and have found this toolbox. The program has been able to run successfully. Now I would like to ask you two questions and hope to get your help. 1) If this toolbox is used, does it not need to calculate the sensitivity of the objective function and constraint function to random variables or random parameters in the program? 2) If I change the number "n" of KL expansion and the degree "p" of polynomial, how to select the number of collocation points? What are the arrangement rules of collocation points? Because I haven't found the relevant literature, I hope to get your help. Looking forward to your reply, thank you for your help!

…

------------------ 原始邮件 ------------------ 发件人: "DesignEngrLab/RBTO" ***@***.***>; 发送时间: 2022年6月7日(星期二) 晚上10:47 ***@***.***>; ***@***.******@***.***>; 主题: Re: [DesignEngrLab/RBTO] About DesignEngLab/RBTO (Issue #2) Hi. Sorry for the late reply. Have you been able to run the code? It requires you to import the CODES toolbox. http://codes.arizona.edu/toolbox/help/html/CODES.html — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: ***@***.***>

[pbtrung]

Sorry for the late reply again. I am being snowed under with the current work. I will more actively respond from now on.

1. Please look at the API of the toolbox and the function I used in my code, so you will know what is needed.
2. The number of collocation points is fixed for the third-degree, 2-dimensional polynomial used in my paper. Of course, a higher-order polynomial is entirely possible but requires significantly more work. We can discuss it further if you would like to do so.

I sent you an invite to a private repo so that we can chat privately. Please see the issue.

[kikidoyoulo]

I'm sorry to bother you .I'm very interested in RBTO, but I want to ask you a question. how can I change the code if I want to use volume as the constraint and minimize compliance like 99 lines? I tried a lot of methods but failed

[pbtrung]

Hi,

Thanks for your interest.
However, my specific problem does not use volume constraint, so adding such a constraint will be a different problem.
I have to see the formulation of your problem to comment any further.

[kikidoyoulo]

Thank you for your reply. Which paper did your Monte Carlo simulation program refer to?

[kikidoyoulo]

Thank you for your reply. How to display the final validation configuration by MCS? Because I want to use MCS to verify the correctness of the algorithm configuration.

…

------------------ 原始邮件 ------------------ 发件人: "DesignEngrLab/RBTO" ***@***.***>; 发送时间: 2022年10月28日(星期五) 上午10:47 ***@***.***>; ***@***.******@***.***>; 主题: Re: [DesignEngrLab/RBTO] About DesignEngLab/RBTO (Issue #2) Hi, Thanks for your interest. However, my specific problem does not use volume constraint, so adding such a constraint will be a different problem. I have to see the formulation of your problem to comment any further. — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: ***@***.***>

[pbtrung]

It's just standard MCS: sample the input random variables, run the simulation with the sample inputs to produce many outputs, and then compute the statistics of the output.
For the validation, just calculate the failure probability = number of failures / number of samples.

[xiaomifengya]

Hello, I would like to bother you again. When I use your code to apply it to compliant mechanisms, such as the reverser, the maximum mutual strain energy is taken as the objective function, the stress is taken as the limit state function, and the volume is also taken as the constraint. When I use the elastic modulus range in your article, the objective function changes irregularly with the increase of the beta value. I see that the objective function (volume) changes regularly with the increase of beta in your article. So I would like to ask what problems should be paid attention to when applying your code to the compliant mechanism, and how to solve this problem. I look forward to your reply. Thank you!

…

------------------ 原始邮件 ------------------ 发件人: "DesignEngrLab/RBTO" ***@***.***>; 发送时间: 2022年8月5日(星期五) 凌晨2:16 ***@***.***>; ***@***.******@***.***>; 主题: Re: [DesignEngrLab/RBTO] About DesignEngLab/RBTO (Issue #2) Sorry for the late reply again. I am being snowed under with the current work. I will more actively respond from now on. Please look at the API of the toolbox and the function I used in my code, so you will know what is needed. The number of collocation points is fixed for the third-degree, 2-dimensional polynomial used in my paper. Of course, a higher-order polynomial is entirely possible but requires significantly more work. We can discuss it further if you would like to do so. — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: ***@***.***>

[ASUNA125456456]

Sorry to disturb you, I want to know how to learn RBTO.

[pbtrung]

@xiaomifengya Heys, sorry for the late reply. Have you figured it out yet? What do you mean by irregular/regular change? Maybe a graph (objective vs. beta) would be helpful for diagnosis.
My initial guess is that stress is sensitive to material distribution; if stress is too big in certain regions (i.e, less material), it might make the objective function show all kinds of weird behaviors.

@ASUNA125456456 Are you serious? If yes, please read the paper first and the references therein.

Best.

[ASUNA125456456]

Thank you for your reply, I am getting started in this direction, but do not know how to learn, I have learned some knowledge of reliability and topology optimization, I do not know how to learn more in-depth RBTO. Thank you very much for your reply.