equinor / equinor/everest-tutorials

Improvement suggestions for multi-objective optimization tutorial

Open
#73 1 comment 0 reactions 0 assignees View on GitHub
documentation enhancement
Dominant language
Assembly
Stars
3
Forks
2
Avg merge
1h 23m
Merged PRs (30d)
1

Description

**Description**
Multi-objective optimization tutorial currently lacks several relevant information for being useful or fully understandable

**Proposal**
Below is a list of suggested improvements:
- [ ] Include link / path to where in the repository to find the configuration file to perform this multi-objective experiment
- [x] If the idea is to avoid repetition with respect to single objective well rate optimization experiment, include link to that tutorial doc to increase coherence across tutorials. If not, state explicitly somewhere that this is optimization of time-varying water injection rate targets
- [x] Add snippet from configuration file to show how to define multiple objectives in `objective_functions` section
- [x] Clearly indicate in the text what are the weights assumed for NPV and FWPT
- [x] Correct formula `f(x) = w_1 * npv(x) - w_2 * fwpt(x)`: it should be `f(x) = w_1 * npv(x) - w_2 * fwpt(x)` and stated that `fwpt` objective is to be minimized (therefore, multiplier -1 in `fm_extract_summary_data`)
- [x] Add snippet from configuration file to show usage of `fm_npv` and `fm_extract_summary_data` to generate objective function files
- [x] Improve objective function plots to display evolution of npv and fwpt objectives over iterations (instead of total combined objective only)
- [x] Unclear why there are two sets of results / plots at the end. Is there one multi-objective optimization experiment or two? The text only describes one experiment

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.