eulumdat-ugr

UGR (Unified Glare Rating) calculation from EULUMDAT (.ldt) photometric files — part of the eulumdat-* ecosystem.

Produces the full 19 × 10 UGR table per CIE 117:1995 and CIE 190:2010, compatible with DIALux and Relux catalogue output.

Installation

pip install eulumdat-ugr

Quick start

from pyldt import LdtReader
from eulumdat_ugr import UgrCalculator

ldt = LdtReader.read("luminaire.ldt")
result = UgrCalculator.compute(ldt)

# NumPy array (19, 10)
print(result.values)

# Export to CSV
with open("ugr_table.csv", "w") as f:
    f.write(result.to_csv())

# Export to JSON (compact)
with open("ugr_table.json", "w") as f:
    f.write(result.to_json())

# Export to JSON (human-readable)
with open("ugr_table.json", "w") as f:
    f.write(result.to_json(indent=2))

Output format

result.values is a (19 × 10) NumPy array:

• 19 rows — standard room configurations from CIE 190:2010 (2H×2H to 12H×8H)
• Columns 0–4 — crosswise orientation, 5 reflectance combinations
• Columns 5–9 — endwise orientation, same 5 reflectance combinations

Reflectance combinations (col % 5):

Standards

• CIE 117:1995 — UGR formula, Guth position index (Table 4.1)
• CIE 190:2010 — tabular method, background luminance (E_WID), standard room geometry

Validation

Validated against DIALux and Relux reference tables on 11 luminaire samples (SHR = 0.25):

DIALux values are systematically 0.1–0.8 UGR below Relux for small rooms (k ≤ 2.0) — a known inter-software variation. Our implementation follows CIE 190:2010 and aligns with Relux.

Dependencies

• eulumdat-py >= 1.0.0
• eulumdat-luminance >= 1.2.0
• numpy >= 1.24
• scipy >= 1.10

Geometry

Fixed geometry per CIE 190:2010 §4.2:

• H = 2.0 m (luminaire height above observer eye level)
• Observer at mid-point of short wall, eye level 1.2 m
• SHR = 0.25 (standard catalogue spacing)