[FBumann]

Prompt: I dived in a bit and I think we should just change it to the correct one (absent here). (review of #894)

Prompt: A stacked PR should do those changes!!

Note

The following content was generated by AI.

What this changes

In examples/pypsa.yaml, 34 parameters are declared missing: absent instead of 0. These are the 18 *_set parameters and the 16 global-constraint weights. Stacked on #894, which makes a value the only other choice.

Which parameters, and why the model stays the same

• The 18 *_set parameters, a given schedule or nominal power, such as Generator_p_set and Store_e_nom_set.
  • Each is read only in its own constraint, under its own presence test (where: Generator_p_set AND …).
  • A bare number name in a where is false at a missing row under both readings. So absent builds the same rows that 0 did.
• The 16 global-constraint weights: *_primary_energy_weight, *_operational_limit_weight, *_volume_weight, *_expansion_cost_weight and *_tech_capacity_weight.
  • Each is a factor inside a sum. Under absent, a missing weight leaves the summand out, where 0 added zero times it.
  • The sum has the same value, and the model has fewer zero terms.
  • The assumptions that read the storage weights use them only as a presence test.
• The fragments in examples/pypsa/ and the gallery pages are regenerated with tools.pypsa_split and tools.gallery.

Why

Bryn asked in review of #894 whether a _set table and a global-constraint weight should be absent. Each one applies only where it has a row, so absent is what the data means.

Verification

Gates on the same tree as 23e715e, where this change was first made on #894's branch and then reverted there (2d7355f):

• pixi run lint passes.
• pixi run test passes: 3077 passed, 1 skipped.
• pixi run docs-build passes.
• compile-tex was not run, because the container cannot download the TeX bundle. CI runs it.

The uses. A script listed every use of each changed parameter in examples/pypsa.yaml. All 18 *_set uses are under their own presence test. All 16 weights are factors inside a sum, or a presence test in an assumption.

Not measured. No test in this repository solves a model. The claim that the optimum does not change rests on the reading rules above. specsolve's differential rungs are where it would be measured.

Not done

• The cycle weights (Line_cycle_weight, Transformer_cycle_weight, Transformer_phase_shift_weight) stay missing: 0. They are factors in the KVL angle sums, the same pattern, but the review did not name them.
• A global-constraint row that reads no component at all can now be absent where it was 0 <= limit. This differs only where the limit is negative. Before, that row was infeasible.

🤖 Generated with Claude Code

https://claude.ai/code/session_01D2S9sN1jLGGMCCwqwVyGD6

[read-the-docs-community]

Documentation build overview

📚 mathspec | 🛠️ Build #35288798 | 📁 Comparing b255c76 against latest (0f892c0)

  🔍 Preview build  

32 files changed · ± 32 modified

± Modified

• reference/changelog/index.html
• reference/glossary/index.html
• reference/language/index.html
• reference/notation/index.html
• reference/program/index.html
• reference/reading/index.html
• examples/pypsa/carrier/index.html
• examples/pypsa/cost/index.html
• examples/pypsa/generator/index.html
• examples/pypsa/generator_commitment/index.html
• and 22 more...

[FBumann]

Note

The following content was generated by AI.

Closed as a duplicate of #912, which is also stacked on #894. #912 declares the same 34 parameters missing: absent, and 27 more.

---

Generated by Claude Code