vibe19-pandas-fdd-rules

name	vibe19-pandas-fdd-rules
description	Use when implementing HVAC fault rules in pandas for App 19 with Open-FDD cookbook parity: confirm_fault, poll_seconds, operational gates, ECON free-cool (web weather), VAV fixed flow, CW wet-bulb. Triggers on: FDD, fault, rule, cookbook, confirm_fault, economizer, ECON-3, VAV-7, CW-OPT, OAT-METEO, pandas rule.

Vibe19 — Pandas FDD rules (Open-FDD parity)

Primary reference

Pandas FDD Cookbook

Also: docs/OPENFDD_PARITY.md · docs/OPERATIONAL_GATES.md

Standard rule pipeline

# 1. raw mask from cookbook
raw = ...  # boolean Series aligned to d.index

# 2. operational gate — AND with active_mask (prefer fan_status over fan_cmd)
#    If gate applied and zero active samples → SKIPPED_EQUIPMENT_OFF

# 3. confirm
confirmed = confirm_fault(raw & active, poll_seconds=p["poll_seconds"])

# 4. rollup hours / pct using **active** samples as denominator when gated

Statuses: PASS | FAULT | SKIPPED_MISSING_ROLES | SKIPPED_EQUIPMENT_OFF | NOT_APPLICABLE_EQUIPMENT_TYPE | ERROR

Weather / free cooling

• Weather CSV enriched in app/weather_psychrometrics.py → wx_oa_t, wx_oa_rh, wx_oa_dewpoint, wx_oa_wetbulb
• ECON-3 (runner.econ3_compute): free cool when web dry-bulb in band (default 35–72°F) AND dewpoint < 60°F (derive from RH if needed); fault when mech cooling with damper closed; optional SAT≈SP (“keeping up”)
• OAT-METEO: |oa_t − wx_oa_t| > oat_err (default 5°F, slider)
• CW-OPT-1: CW supply colder than wet-bulb + approach (Stull) — replaced WX-2

Notable rule enhancements

ID	Behavior
VAV-7	Under min flow or fixed high flow (low rolling std + high mean) or high min_flow_sp
ECON-3	Web OAT + dewpoint free-cool window
CW-OPT-1	Condenser water vs wet-bulb optimization

Reference modules

Module	Role
app/rules/cookbook_catalog.py	Canonical 50-rule definitions
app/rules/runner.py	Skip / gate / confirm / ECON-3
app/rules/base.py	confirm_fault, hours_true
app/rules/operational_gate.py	RULE_GATES
app/role_map.py	Role aliases + resolve_role()

After catalog edits: python scripts/generate_rule_configs.py then pytest.

Tests

Synthetic DataFrames in tests/ — no client CSV in git. Parity script: scripts/csv_parity_check.py --building-folder ….