wattlab-esco-bins

name	wattlab-esco-bins
description	Use when working on WattLab's ESCO bin-method savings calculators or weather bins: wattlab/bench/esco.py, wattlab/weather/bins.py, WeatherBins, OperatingSchedule, scheduling/OAD/DCV/static-pressure/DAT/hydronic-reset/ economizer savings, MCWB enthalpy, golden spreadsheet tests. Triggers on: bin method, OAT bins, Weather Man, ESCO calculator, kW/ton, fan laws, enthalpy, therms, scheduling savings.

WattLab — ESCO bin-method calculators

Screening-grade ECM savings the way ESCO engineers actually compute them: 5°F OAT bins × 3 daily shifts, MCWB enthalpy, fan laws, kW/ton. Ported from real retrofit program workbooks (anonymized as School A / School B — never reintroduce client, district, contractor, or building names).

Files

Path	Role
wattlab/bench/esco.py	9 registered calculators (dict in → dict out)
wattlab/weather/bins.py	WeatherBins, BinRow, OperatingSchedule, psychrometrics, washington_dc_noaa()
wattlab/bench/registry.py	@register / get(name)
tests/test_esco_golden.py	Golden pins to source-workbook cell values
vibe20_agent_spec/docs/ESCO_CALCULATORS.md	Formula + anchor documentation

Quick use

from wattlab.bench.registry import get
from wattlab.bench import runner  # noqa: F401 — registers calculators
from wattlab.weather.bins import washington_dc_noaa

out = get("scheduling_fan_bins")({
    "fan_kw_total": 8.579,
    "existing_schedule": {"shifts": [2, 8, 3], "days_per_week": 5},
    "proposed_schedule": {"shifts": [1, 8, 1.5], "days_per_week": 5,
                          "override_allowance": 0.10},
    "bins": washington_dc_noaa(),
})
out["savings_kwh"]

bins accepts a WeatherBins, row dicts, or WeatherBins.from_hourly(df) from vibe19 weather_observed.csv (timestamp + OAT columns).

Calculator cheat sheet

id	Savings driver	Key inputs beyond bins/schedule
scheduling_fan_bins	removed fan hours	fan_kw_total
scheduling_cooling_bins	removed vent-cooling ton-hours	oa_cfm_total, kw_per_ton, supply_enthalpy
scheduling_heating_bins	removed vent-heating below balance	oa_cfm_total, balance_point_f, boiler_efficiency
oad_unoccupied_closed	OA damper shut when unoccupied	mode: cooling|heating + as above
dcv_bins	avoided OA CFM	baseline_oa_cfm, proposed_oa_cfm
static_pressure_reset	fan-law speed reduction	units[{motor_kw, avg_speed_fraction, annual_hours}], pressure_ratio
dat_reset_bins	raised supply enthalpy per bin	reset[{temp, proposed_supply_enthalpy, vav_fraction}]
hydronic_reset_bins	HW/CHW/CDW reset ladder	capacity_mbh, mode, on_point_f, design_temp_f
dewpoint_economizer	free-cooling bins	unit_cfm_total, return_enthalpy, discharge_enthalpy

Sheet conventions everywhere: 4.5·CFM·Δh (total), 1.08·CFM·ΔT (sensible), 12,000 Btu/ton-h, therms = kBtu/100.

Hard rules

1. Golden tests are physics — test_esco_golden.py values come straight from source-workbook cells (e.g. static reset RTU 7 → 2,092.198 kWh; CV fan scheduling → 3,243.17 kWh). Changing math means changing goldens with a documented workbook basis.
2. No client names — School A / School B only. Scrub before commit.
3. Everything flows through WeatherBins + OperatingSchedule — no hardcoded 8760s, no invented hours.
4. New calculator = @register + golden test + row in docs/ESCO_CALCULATORS.md + (if catalog-priced) a branch in studio.py::estimate_proxy_savings.
5. Run python -m pytest tests/test_esco_golden.py -q before claiming done.