optimization problem

[vacupenda]

I encountered some issues while using the strength ratio for blade parameter optimization. So, I want to first run the parameter optimization problem for a single cross-section. Below is a portion of my .yml file, and here's the issue: When I run ivabs analyze main.yml alone, it calculates the cross-section information I need, such as the parameters 'mu', 'ea', "gj", "eiyy", "eizz", and 'sr', as shown in Figure 1. However, when I add the study module, it prompts an error message as shown in Figure 2.

analysis:
  steps:
    - step: "cs analysis"
      type: "cs"
      analysis: "h"
      output:
        value: [
          'mu',
          'ea', "gj", "eiyy", "eizz",
        ]
        
    - step: "recovery"
      type: "cs"
      analysis: "fi"
      section_response:
        data_form: "list"
        cases:
          - displacement: [1.14e-05,1.92e-06,2.00e-3]  # U1, U2, U3
            load:
              - -6.960731065297227e-05  # F1
              - 0.00044668538601352393  # F2
              - 0.015870420038137607  # F3
              - 0.000229788237956122  # M1
              - 8.13244480333271e-06  # M2
              - -1.4022416078181857e-06  # M3
      output:
        file_name: "strength_ratio.yml"
        file_format: "yml"
        value: ['sr',]
    
study:
  method:
    format: "keyword"
    multidim_parameter_study:
      partitions: [2, 1]
    # soga:  # Single objective genetic algorithm
    #   max_function_evaluations: 10
    #   population_size: 5
    #   seed: 1027
    #   print_each_pop: true
  variables:
    data_form: "explicit"
    list:
      - name: "ang_spar"
        type: "continuous"
        bounds: [0, 10]
      - name: "ply_spar"
        type: "discrete"
        space_type: "range"
        bounds: [950, 953]
  # responses:
  #   data_form: "explicit"
  #   objective_functions:
  #     - descriptor: 'sr'
  #       sense: 'maximize'
  responses:
    data_form: "explicit"
    response_functions:
      - descriptor: 'mu'
      - descriptor: 'ea'
      - descriptor: 'sr'
      - descriptor: 'gj'
      - descriptor: 'eiyy'
      - descriptor: 'eizz'
  interface:
    fork:
      parameters_file: "input.in"
      results_file: "output.out"
      file_save: on
      work_directory:
        directory_tag: on
        directory_save: on
    required_files:
      - "design/*"

[unitasium]

As shown in the first image, sr returns an array containing strength ratios for all load cases and cross-sections, even if there is only one case and one cs. To be used as objectives or constraints, any quantity must return a single number. For your case, you can output sr_min in the recovery step.

[unitasium]

It seems that the "recovery" step is not carried out. However, this file works fine on my machine. Can you try rerun this file in another clean directory?

[vacupenda]

My computer still cannot recognize sr_min. I wonder if it would be convenient for you to compress and send me your executable file, so I can check if it runs on my computer.

[vacupenda]

I re-cleaned the file content, and this time it showed that it was performing the “recovery,” but the 'sr_min' still couldn't be recognized.
The attachment is the file I used for calculation. I'm not sure where the problem lies. Could you please help me check if it can run successfully on your computer when you have time? The command I used is “ivabs iter main.yml”.

test_srmin.zip

[vacupenda]

I'm not sure if this file can run on your system, but if you have the time, could you please help me identify the issue? It consistently fails to run on my computer, and even after I reinstalled IVABS on another machine, the outcome was the same—it cannot recognize sr_min.

[unitasium]

Yes. I will test it.