docs(pypsa): the pypsa page lists seven more refusals and eight data-prep steps, and a capital cost says where fom_cost and overnight_cost enter

[FabianHofmann]

Prompt: Continue PyPSA parity for energy-models/mathspec: implement step 4 of the plan in tracking issue #783. Step 4 covers the parity items the language can already state. Run it as an ultracode workflow with at most 4 agents, one PR per group.

Note

The following content was generated by AI.

The PyPSA page now has seven more Refusals rows and a Data prep section with eight steps. Each capital cost says how fom_cost and overnight_cost enter, and the transmission cost limit reads the cost without fom_cost. No math changes.

Stacks on #787, which stacks on #786. Part of #783 (step 4, group 4: docs only).

Method, gate output, PyPSA cites, what is not done

Method. I re-read each #783 claim on PyPSA master 02bdcbba before I wrote it down. Line numbers in the new text are on 02bdcbba.

fom_cost and overnight_cost. The seven *_capital_cost descriptions in examples/pypsa.yaml said "periodized as an annuity in data prep". That is wrong when overnight_cost is absent. They now give PyPSA's periodized_cost: overnight_cost as an annuity times nyears where it is given, capital_cost where it is not, plus fom_cost (components.py:1126-1147, costs.py:102-203, read by the objective at optimize.py:222, :406). Line_expansion_cost_weight and Link_expansion_cost_weight now say the limit reads the capital_cost property, without fom_cost (components.py:1151-1169, global_constraints.py:935). The rung 3 and rung 5 index rows say the same in short form. period_weight_objective now says that data prep feeds one without multi_investment_periods (optimize.py:205-207, :264-266).

Refusals, seven rows.

PyPSA	claim checked
consistency.py:231-283	a warning, ConsistencyError only under strict (_log_or_raise, :45-49). The file builds the same: min > max keeps the fixed shift, and an infinite bound leaves the shift free
constraints.py:1819-1820	{c}-p_nom exists only for extendable, active units (variables.py:354-360), so .sel raises KeyError
constraints.py:2772, :2781	Generator-p spans only active_assets (variables.py:44-46), so e_sum_min or e_sum_max on an inactive unit raises KeyError. The file builds 0 >= e_sum_min
constraints.py:1558-1563	Empty LHS with non-zero RHS where every term of a loaded bus is masked in a snapshot
components/descriptors.py:141-144	get_active_assets(period) raises for a period outside n.investment_periods. Reached from global_constraints.py:88 and the transmission limits
constraints.py:1146-1148, :1153-1158	with another committable fixed unit, a non-committable unit that came in off and has only a shut-down ramp gets p >= (shut - 1) * p_nom at the first snapshot. The file gives p >= -p_nom. This binds only for a unit that can run below zero
constraints.py:120-125	0 * inf is set to 0 for a fixed build of inf

Data prep, eight steps. These are typed branches (network/power_flow.py:441-564) and the carrier fill from bus0 (:780-782, :831-837). Then the KVL weights x_pu_eff and r_pu_eff (networks.py:1360-1365), and bus0 for branches, Link and Process in tech_capacity_expansion_limit (global_constraints.py:93). Then the period weight on a plain run, and storage in primary_energy: only cyclic_state_of_charge / e_cyclic exclude a unit (global_constraints.py:436, :490), and operational_limit also reads active (:636, :687). Last, no capacity variable for active = False (variables.py:354-360), and no row for an empty global constraint (global_constraints.py:98-99, :538-539, :730-731, :844-845, :939-940).

Gates.

• pixi run python -m tools.pypsa_split split examples/pypsa and pixi run python -m tools.gallery: fragments and pages regenerated, not edited.
• uv run --script examples/references/pypsa/reference.py --check: every rung solves to its record.
• pixi run lint: pass.
• pixi run ci: 2621 passed, docs build with no issues, 55 TeX documents compiled.
• New Data prep section: 20 sentences, median 19 words, 4 over 25.

Not run. reference.py without --check: no rung is new or changed, so there is nothing to record.

Not done.

• No rung and no math change. The shut-down-ramp row is recorded as "not refused". Extending {c}_came_in_running_unless_committable to read ramp_limit_shut_down is a math change, and it belongs to the PyPSA parity with master: open points, new master changes and the PyPSA bugs found #783 item "First-snapshot ramp from a start-up or shut-down limit alone".
• The cite corrections C1 and C2 are not applied. The tree has no wrong C2 cite. Rung 57 cites constraints.py:1708 and rung 58 cites :1872. Both are 1.3.0 lines, and they move to 1817 and 1981 on master. docs(pypsa): the pypsa line references point at pypsa master #806 re-anchors these cites, the networks.py:1354-1361 cites (C1), and the two sentences that say line numbers are 1.3.0. If I changed those lines here, they would conflict with docs(pypsa): the pypsa line references point at pypsa master #806.
• So until docs(pypsa): the pypsa line references point at pypsa master #806 merges, the Refusals intro still says "Line numbers are PyPSA 1.3.0", but the seven new rows cite 02bdcbba.
• Line_cycle_weight and the *_tech_capacity_weight descriptions are unchanged. Their steps are on the page. docs(pypsa): the pypsa line references point at pypsa master #806 edits the Line_cycle_weight line.

[read-the-docs-community]

Documentation build overview

📚 mathspec | 🛠️ Build #34876558 | 📁 Comparing a69450c against latest (34c68fd)

  🔍 Preview build  

16 files changed · ± 16 modified

± Modified

• changelog/index.html
• examples/pypsa_linearized_uc/index.html
• examples/pypsa/generator/index.html
• examples/pypsa/generator_commitment/index.html
• examples/pypsa/generator_maintenance/index.html
• examples/pypsa/line/index.html
• examples/pypsa/link/index.html
• examples/pypsa/link_commitment/index.html
• examples/pypsa/link_maintenance/index.html
• examples/pypsa/process/index.html
• and 6 more...