Build a Blower-Door Infiltration Delta Reporter — implementation

## Build
Process field blower-door ACH50 test data against EnergyPlus ZoneInfiltration elements. Compute the variance between empirical natural leakage and baseline epJSON assumptions to highlight necessary model adjustments.

## User
Building Commissioning and Energy Modeler

## Workflow
1. Upload empirical blower door test results including structural dimensions.
2. Extract baseline ZoneInfiltration:DesignFlowRate estimates from project epJSON.
3. Compute actual natural infiltration flow rates using empirical formulas.
4. Output a delta variance report prioritizing zones that require calibration.

## Interface
Calibration workbench presenting blower door curve charts, current versus calculated infiltration rates, and delta reports.

## Data
Utilize open-source DOE commercial reference building EnergyPlus epJSON files from the public NREL repository.
Starter examples below are synthetic. They are not downloaded records.

## Source
[EnergyPlus](https://raw.githubusercontent.com/NREL/EnergyPlus/develop/README.md)
Extracts ZoneInfiltration:DesignFlowRate baseline data from the EnergyPlus epJSON schema for comparison against field tests.

## Tools
dspy, scipy, pydantic.
DSPy facilitates systematic logic structuring and schema-aligned reporting generation for epJSON infiltration comparisons.
Hosted reasoning model capable of technical building physics reasoning and strict JSON generation.

## Deliverables
Submit runnable source, pinned dependencies, setup commands, a working interface, example outputs, and a recorded end-to-end demo. Include tests and an action log.

## Acceptance
- Converts ACH50 empirical figures to operational design flow rate (m3/s/m2) using natural infiltration heuristics.
- Calculates variance delta percentages accurately against baseline epJSON flow parameters.
- Produces valid structured JSON variance reports without directly mutating the source epJSON files.

## Starter Examples
```json
[
  {
    "input": {
      "building_zone": "Main_Office_Floor1",
      "zone_volume_m3": 3600,
      "envelope_area_m2": 1000,
      "measured_ach50": 5,
      "current_epjson_flow_per_exterior_area": 0.0002
    },
    "expected_output": {
      "building_zone": "Main_Office_Floor1",
      "measured_ach50": 5,
      "calculated_natural_leakage_m3_s_m2": 0.00025,
      "calibration_delta_pct": 25,
      "status": "CALIBRATION_REQUIRED",
      "note": "simplified synthetic app record"
    }
  },
  {
    "input": {
      "building_zone": "Retail_Entry_Vestibule",
      "zone_volume_m3": 1440,
      "envelope_area_m2": 400,
      "measured_ach50": 2,
      "current_epjson_flow_per_exterior_area": 0.0001
    },
    "expected_output": {
      "building_zone": "Retail_Entry_Vestibule",
      "measured_ach50": 2,
      "calculated_natural_leakage_m3_s_m2": 0.0001,
      "calibration_delta_pct": 0,
      "status": "ACCEPTABLE_VARIANCE",
      "note": "simplified synthetic app record"
    }
  }
]
```

## Safety
Data processing focuses on diagnostic reporting; model adjustment and subsequent simulation runs require manual validation by engineers.
Use mock external writes. Require human review before any real deployment.