Question: The execution logic of the package

[Jungle-13]

Thank you ,Sir! I want to know that in what order should this whole set of codes be run, or what is the logic of this set of codes?

[rozsasarpi]

Running the code

Running the calibration: run main.m. This will also save the results in a mat file and create some plots, see the bottom of main.m. The idea is that once you defined your problem you will need to interact (run) only this single file (main.m).

Logic of the set of codes

Note that it is particular to the cases considered in the related paper but of course you can adapt it to your needs.

A brief sketch of the logic.

• main.m
  • main file
  • running the calibration (CALIBRATION section)
  • contains some general settings, e.g. objective function to be used, target reliability, actions to be considered, etc. (see the OPTIONS & MODEL SPECIFICATION section)
• all the relevant inputs and settings that define a calibration problem are collected in the Model struct.
• all the relevant results from calibration are collected in the Results struct.
• the above two are passed around in most high level functions.
• model_spec: definition/specification of the models for the calibration, e.g. distribution function of the random variables, weights to be used for design scenarios (based on load ratios).
• model_check: a basic check if the inputs and settings are consistent in Model (should follow model_spec).
• calibrate
  • performs the optimization
  • in the objective function of the optimization (obj_fun) it loops over all leading actions, limit states, and load ratios: for each case
  • a partial factor based design (pf_design) is performed in a way that the utilization is 100%, i.e. the design value of the action equals the design value of the resistance
  • preparation for the reliability analysis (prob_model) the partial factor based design is connected to the "full" probabilistic representation of the variables, i.e. going from standardized representative values to distributions.
  • reliability analysis (form_wrapper) using FORM from FERUM.
• post-processing: plot_reli_vs_loadratio, plot_multisolu_gamma, etc.

I hope that this helps. If you have further questions please let me know.

[Jungle-13]

Dear Sir. According to your prompt, I successfully ran the result. However, a new problem has appeared (haha), as follows:

Index in position 2 is invalid. Array indices must be positive integers or logical values.

Error in select_Results (line 30)
     [~, solu_idx] = min(sum(bsxfun(@minus, PF(:,pf_idx), pf).^2,2));

Error in plot_multisolu_gamma (line 30)
Results2 = select_Results(Model, Results, pf_idx, pf);

Error in main (line 90)
plot_multisolu_gamma(Model, Results)

[rozsasarpi]

Please try if changing this line:

https://github.com/rozsasarpi/Code-calibration-Belarus/blob/c35fa2b7d3dd5dce987c512d97060f6b33fd4da3/Matlab/plot_multisolu_gamma.m#L11

to this

pf_idx = 1;

solves it.

[Jungle-13]

Many thanks for you！

[rozsasarpi]

I fixed the index, see the above linked merge. Please let me know if you have further questions or the issue is solved.

[Jungle-13]

Yeah,Sir. Another problem as follows when I got the result: partial_f =

1.4940    0.8814    1.3046    0.9876    1.1832

---

Unrecognized function or variable 'plot_guides'.

Error in plot_multisolu_gamma (line 141) plot_guides(gca, Line)

Error in main (line 90) plot_multisolu_gamma(Model, Results)

---

[rozsasarpi]

It as another custom dependency that you need to add to your path: https://github.com/rozsasarpi/Plotting-Matlab/blob/1fc56ca127390009995924d1ac7590f415a43959/misc/plot_guides.m

I expect that a few more functions will be needed from the linked repository so it is probably the most convenient to add the entire content of the linked repo to your path.

[Jungle-13]

Sorry I was delayed by something, just saw your reply . I have added the file to my setup path. Luckily, I have run perfectly with your literature (Calibrating Partial Factors – Methodology, Input Data and Case Study of Steel Structures), thank you very much for your Patient and detailed reply. My series of questions is over (haw-haw). I wish you good health and all the best. Best wishes!

[rozsasarpi]

I'm glad to hear that it works. If you have a question related to the paper and/or the code in the future feel free to open an issue.

[Jungle-13]

Yeah,Sir. I will open an issuel on the Github if I have any question related to the paper and/or the code in the future. Thank you from from the bottom of my heart.  Best wishes! Zhangong Wang

------------------ 原始邮件 ------------------ 发件人: "rozsasarpi/Code-calibration-Belarus" @.>; 发送时间: 2022年1月15日(星期六) 晚上11:20 @.>; @.**@.>; 主题: Re: [rozsasarpi/Code-calibration-Belarus] Question: The execution logic of the package (Issue #3)

I'm glad to hear that it works. If you have a question related to the paper and/or the code in the future feel free to open an issue.

— Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android. You are receiving this because you authored the thread.Message ID: @.***>