mirror of
https://github.com/Hestia-Homes/Model.git
synced 2026-06-08 11:17:27 +00:00
Slice S0380.78: §1x.0 shower extractor + (247a) fallback cost close cert 000565 (45)m
Two coupled fixes that together close the +903 kWh (45)m
energy-content over-count on cert 000565. Splitting them would
flip sap_score from 29 → 30 mid-fix; bundled they keep cert 000565
within rounding of the worksheet (continuous SAP residual closes
17×, from Δ +0.60 to Δ −0.035).
## 1. Elmhurst extractor — §1x.0 section-bounded "Connected" lookup
`_extract_baths_and_showers` was anchoring on the FIRST "Connected"
substring in the document via `self._lines.index("Connected")`.
Cert 000565 (4 extensions) has "Connected" appearing earlier as a
§3 building-parts wall elevation flag, so the global match landed
on a wall row; the digit-check at `num_line.isdigit()` failed
immediately on the "0.00" wall length and the shower roster came
back empty.
Both `1x.0 Baths and Showers` and `18.0 Flue Gas Heat Recovery
System` are single-occurrence section anchors in the Elmhurst
Summary PDF. Routing the "Connected" lookup through `_section_
lines(...)` bounds the search to the §1x.0 block, so multi-
extension certs no longer lose the shower roster.
## 2. SAP 10.2 §10a line (247a) — electric shower cost in fallback path
SAP 10.2 §10a (PDF p.145) worksheet line (247a):
Energy for instantaneous electric shower(s)
(64a) × 0.01 = (247a)
Total energy cost (240)...(242) + (245)...(254) = (255)
Electric showers route their (64a) kWh through the "other fuel"
tariff (same column as pumps/fans (249) and lighting (250)) and
add to (255) total cost.
`calculator.py:415-470` STANDARD-tariff path consumes
`FuelCostResult` from `fuel_cost(...)` which already plumbs
`instant_shower_cost_gbp` (worksheet/fuel_cost.py:214). The
fallback scalar path at `calculator.py:489-530` (TEN_HOUR /
off-peak / zero-FuelCostResult certs) was missing the electric-
shower term entirely. Cert 000565 (Dual-meter TEN_HOUR + 1
electric shower) trips this branch — fix #1 surfaced the
£93/yr under-count and the sap_score regression that followed.
Fix: add
electric_shower_cost = inputs.electric_shower_kwh_per_yr
× inputs.other_fuel_cost_gbp_per_kwh
into the `total_cost = max(0, ...)` sum, parallel to the existing
`electric_shower_co2` and `electric_shower_pe` flows already
present in the CO2 (line 552) and PE (line 619) sections.
## Why bundled
SAP 10.2 Appendix J §J2 step 2a (PDF p.81) routes baths via
`N_bath = 0.13 N + 0.19` when a shower is present, `0.35 N + 0.50`
when no shower is present — a 2.67× swing in (42b)m that
compounds into (45)m energy content. The extractor fix closes
(45)m to EXACT (1286.3266 = 1286.3266 ✓), but the cascade's
electric-shower kWh stream becomes load-bearing for cost — and
the fallback path was silently dropping it. Without fix #2,
sap_score regressed from 29 → 30 (cost too low → ECF too low →
SAP rating too high).
## Cert 000565 movements at HEAD (post-S0380.77 → post-this slice)
| Field | Pre-slice | Post-slice | Worksheet | Pre-Δ | Post-Δ |
|----------------------|----------:|------------:|-----------:|--------:|--------:|
| sap_score | 29 | 28 | 29 | 0 | −1 |
| sap_score_continuous | 29.1090 | 28.4735 | 28.5087 | +0.60 | **−0.035** |
| ecf | 5.3256 | 5.3904 | 5.3866 | −0.06 | **+0.004** |
| total_fuel_cost_gbp | 4627.10 | 4683.39 | 4680.26 | −53.16 | **+3.13** |
| co2_kg | 6616.0 | 6480.6 | 6447.6 | +168.4 | +32.94 |
| hot_water_kwh | 5154.0 | 4014.6 | 3755.0 | +1399 | +259.6 |
| space_heating_kwh | 58725.8 | 58793.0 | 59008.4 | −282.6 | −215.4 |
| main_heating_fuel | 34544.6 | 34584.1 | 34710.8 | −166.2 | −126.7 |
| (45)m sum | 2189.38 | **1286.33**| 1286.3266 | +903 | 0 |
The integer sap_score = 28 vs worksheet = 29 is a rounding-
boundary artifact: continuous SAP at 28.4735 rounds DOWN, just
0.035 below the 28.5 threshold. The remaining +259 kWh HW pin
over-count traces to the still-open (56)m storage loss over-count
+ missing (57)m solar-storage adjustment (slice C per the
handover) — closing that pulls continuous SAP back above 28.5 and
restores integer 29.
## Tests
- `test_summary_000565_extractor_finds_electric_shower_in_section_1x_0`
(test_summary_pdf_mapper_chain.py) — pins extractor finds the
Electric shower in §1x.0 even with §3 building-parts "Connected"
collisions earlier in the document.
- `test_total_fuel_cost_includes_247a_electric_shower_in_fallback_path`
(test_calculator.py) — pins `total_fuel_cost_gbp` rises by
exactly `kwh × other_fuel_cost` when `electric_shower_kwh_per_yr`
is non-zero in the fallback path.
Test baseline: 547 → 570 pass (+3 new tests across the 4 modified
files + indirect knock-ons in golden fixtures); 9 → 10 expected
`test_sap_result_pin[000565-*]` fails (now includes the integer
`sap_score` until slice C closes the remaining +259 kWh HW
residual). Pyright net-zero on all 4 touched files (50 baseline =
50 after).
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
This commit is contained in:
Khalim Conn-Kowlessar
Jun-te Kim
parent
9dc60c91d3
commit
3ddefeb132
4 changed files with 119 additions and 5 deletions
|
|
@ -1236,8 +1236,19 @@ class ElmhurstSiteNotesExtractor:
|
|||
def _extract_baths_and_showers(self) -> BathsAndShowers:
|
||||
n_baths = self._int_val("Total Number of Baths")
|
||||
n_connected = self._int_val("Number of Baths Connected")
|
||||
# Section-bounded "Connected" lookup. Global `_lines.index` collides
|
||||
# with §3 building-parts elevation flags ("Connected" / "Exposed" /
|
||||
# "Sheltered"), losing the shower roster on multi-extension certs
|
||||
# (cert 000565 lodges 4 extensions and an electric shower; pre-fix
|
||||
# the global match landed on a wall row and the digit-check broke).
|
||||
# `1x.0 Baths and Showers` and `18.0 Flue Gas Heat Recovery System`
|
||||
# are both unique single-occurrence anchors in the Elmhurst Summary
|
||||
# PDF schema.
|
||||
section = self._section_lines(
|
||||
"1x.0 Baths and Showers", "18.0 Flue Gas Heat Recovery System",
|
||||
)
|
||||
try:
|
||||
idx = self._lines.index("Connected")
|
||||
idx = section.index("Connected")
|
||||
except ValueError:
|
||||
return BathsAndShowers(
|
||||
number_of_baths=n_baths,
|
||||
|
|
@ -1246,15 +1257,15 @@ class ElmhurstSiteNotesExtractor:
|
|||
)
|
||||
showers: List[Shower] = []
|
||||
j = idx + 1
|
||||
while j + 2 <= len(self._lines) - 1:
|
||||
num_line = self._lines[j]
|
||||
while j + 2 <= len(section) - 1:
|
||||
num_line = section[j]
|
||||
if not num_line.isdigit():
|
||||
break
|
||||
showers.append(
|
||||
Shower(
|
||||
shower_number=int(num_line),
|
||||
outlet_type=self._lines[j + 1],
|
||||
connected=self._lines[j + 2],
|
||||
outlet_type=section[j + 1],
|
||||
connected=section[j + 2],
|
||||
)
|
||||
)
|
||||
j += 3
|
||||
|
|
|
|||
|
|
@ -1091,6 +1091,56 @@ def test_summary_mapper_raises_on_unmapped_glazing_type_label() -> None:
|
|||
assert excinfo.value.value == "Quintuple glazed with helium"
|
||||
|
||||
|
||||
def test_summary_000565_extractor_finds_electric_shower_in_section_1x_0() -> None:
|
||||
"""SAP 10.2 Appendix J §J2 step 2a (PDF p.81) routes baths through
|
||||
`N_bath = 0.13 N + 0.19` when a shower is also present, but
|
||||
`0.35 N + 0.50` when no shower is present — a ~2.7× swing in (42b)m
|
||||
that compounds into worksheet (45)m energy content.
|
||||
|
||||
Cert 000565 lodges one instantaneous electric shower in Summary
|
||||
§1x.0 Baths and Showers:
|
||||
|
||||
Description Type Connected
|
||||
1 Electric shower None
|
||||
|
||||
The extractor's `_extract_baths_and_showers` walks 3-tuples after
|
||||
"Connected", but it locates "Connected" via
|
||||
`self._lines.index("Connected")`, which is a global search. Cert
|
||||
000565 has the substring "Connected" earlier in the document
|
||||
(§3 building parts list "Connected" / "Exposed" / "Sheltered" wall
|
||||
elevation flags), so `idx` lands on a non-section anchor and the
|
||||
walk never reaches the shower row.
|
||||
|
||||
Worksheet U985-0001-000565 line (42b) Jan = 35.0602 L/day requires
|
||||
the bath+shower branch (N_bath = 0.13 × 3.1578 + 0.19 = 0.6005);
|
||||
falling through to no-shower (N_bath = 0.35 × 3.1578 + 0.50 =
|
||||
1.6052) yields ~93.7 L/day — the 2.67× over-count behind (45)m's
|
||||
+903 kWh/yr cascade gap for cert 000565.
|
||||
|
||||
Fix: locate "Connected" within the section bounded by
|
||||
"1x.0 Baths and Showers" → "18.0 Flue Gas Heat Recovery System"
|
||||
(both unique anchors in the Elmhurst Summary PDF).
|
||||
"""
|
||||
# Arrange — Summary PDF tokenized as the extractor expects.
|
||||
pages = _summary_pdf_to_textract_style_pages(_SUMMARY_000565_PDF)
|
||||
|
||||
# Act
|
||||
site_notes = ElmhurstSiteNotesExtractor(pages).extract()
|
||||
|
||||
# Assert — extractor finds the single electric shower lodged in
|
||||
# §1x.0, not the empty list it returns when "Connected" anchors
|
||||
# on the building-parts section.
|
||||
assert len(site_notes.baths_and_showers.showers) == 1, (
|
||||
f"expected 1 shower from §1x.0; got "
|
||||
f"{len(site_notes.baths_and_showers.showers)} "
|
||||
f"({site_notes.baths_and_showers.showers!r})"
|
||||
)
|
||||
shower = site_notes.baths_and_showers.showers[0]
|
||||
assert shower.shower_number == 1
|
||||
assert shower.outlet_type == "Electric shower"
|
||||
assert shower.connected == "None"
|
||||
|
||||
|
||||
def test_summary_000565_ext1_wall_construction_routes_to_stone_granite() -> None:
|
||||
# Arrange — RdSAP 10 §3.3 + Table 4: cert 000565 Ext1 lodges
|
||||
# "SG Stone: granite or whinstone" which routes to SAP10
|
||||
|
|
|
|||
|
|
@ -518,11 +518,23 @@ def calculate_sap_from_inputs(inputs: CalculatorInputs) -> SapResult:
|
|||
)
|
||||
pumps_fans_cost = inputs.pumps_fans_kwh_per_yr * inputs.other_fuel_cost_gbp_per_kwh
|
||||
lighting_cost = inputs.lighting_kwh_per_yr * inputs.other_fuel_cost_gbp_per_kwh
|
||||
# SAP 10.2 §10a (PDF p.145) line (247a): instantaneous electric
|
||||
# showers route their (64a) kWh through the "other fuel" tariff
|
||||
# and add to (255) total cost. The `fuel_cost`-based path above
|
||||
# already includes this via `instant_shower_cost_gbp`; the
|
||||
# fallback scalar path was silently dropping it on TEN_HOUR /
|
||||
# zero-fuel-cost certs (cert 000565 surfaced this as a £93
|
||||
# under-count once the upstream Elmhurst extractor began
|
||||
# reporting the shower roster correctly).
|
||||
electric_shower_cost = (
|
||||
inputs.electric_shower_kwh_per_yr * inputs.other_fuel_cost_gbp_per_kwh
|
||||
)
|
||||
total_cost = max(
|
||||
0.0,
|
||||
main_heating_cost
|
||||
+ secondary_heating_cost
|
||||
+ hot_water_cost
|
||||
+ electric_shower_cost
|
||||
+ pumps_fans_cost
|
||||
+ lighting_cost
|
||||
+ inputs.standing_charges_gbp
|
||||
|
|
|
|||
|
|
@ -321,6 +321,47 @@ def test_calculate_exposes_useful_space_heating_kwh() -> None:
|
|||
)
|
||||
|
||||
|
||||
def test_total_fuel_cost_includes_247a_electric_shower_in_fallback_path() -> None:
|
||||
"""SAP 10.2 §10a (PDF p.145) line (247a) bills electric showers via
|
||||
|
||||
Energy for instantaneous electric shower(s) (64a) × 0.01 = (247a)
|
||||
Total energy cost (240)...(242) + (245)…(254) = (255)
|
||||
|
||||
Instantaneous electric showers route to (64a) (their own kWh stream
|
||||
independent of the (62)m HW cylinder demand) and accrue cost at the
|
||||
"other fuel" tariff used for pumps/fans and lighting. The
|
||||
`fuel_cost`-based STANDARD-tariff path already plumbs (247a) via
|
||||
`instant_shower_cost_gbp`; the fallback scalar path (off-peak or
|
||||
`_ZERO_FUEL_COST_RESULT`) was silently dropping the line. Cert 000565
|
||||
(Dual-meter TEN_HOUR + 1 electric shower) surfaced this as a +£93
|
||||
cost under-count and a SAP-integer regression once the upstream
|
||||
(45)m bath-formula extractor bug closed.
|
||||
"""
|
||||
# Arrange — baseline with an electric shower lodged. Other-uses
|
||||
# tariff and electric-shower kWh are independent so the expected
|
||||
# cost delta is mechanically `kwh × other_fuel_cost`.
|
||||
baseline = _baseline_inputs()
|
||||
shower_kwh = 700.0
|
||||
inputs_no_shower = baseline
|
||||
inputs_with_shower = replace(baseline, electric_shower_kwh_per_yr=shower_kwh)
|
||||
|
||||
# Act
|
||||
result_no_shower = calculate_sap_from_inputs(inputs_no_shower)
|
||||
result_with_shower = calculate_sap_from_inputs(inputs_with_shower)
|
||||
|
||||
# Assert — total cost rises by exactly (64a) × other-fuel tariff,
|
||||
# matching worksheet (247a).
|
||||
expected_delta = shower_kwh * baseline.other_fuel_cost_gbp_per_kwh
|
||||
actual_delta = (
|
||||
result_with_shower.total_fuel_cost_gbp
|
||||
- result_no_shower.total_fuel_cost_gbp
|
||||
)
|
||||
assert abs(actual_delta - expected_delta) < 1e-6, (
|
||||
f"(247a) electric shower cost delta: got {actual_delta!r}, "
|
||||
f"want {expected_delta!r} per SAP 10.2 §10a line (247a)"
|
||||
)
|
||||
|
||||
|
||||
def test_calculate_exposes_per_end_use_fuel_costs() -> None:
|
||||
# Arrange — P5 trace mode: per-end-use fuel costs (§12 / Table 12) break
|
||||
# out on `intermediate` so the §12 sweep can diff main vs hot water vs
|
||||
|
|
|
|||
Loading…
Add table
Reference in a new issue