EXPERIMENT 004 BUILDING PHYSICS

Envelope LabBuilding envelope lab

Change insulation thickness and continuity, then compare heat flow through two sections under the same indoor and outdoor conditions.

2D STEADY-STATE CONDUCTION∇ · (k ∇T) = 0

A → BTwo designs, one climate.

2D / STEADY

02Slab between 1F and 2FWhere the floor slab meets the wall

Selected strip, enlarged960 mm ↗

−10°C

22°CTemperature · °C

Heat-flow arrowsDimensionsMeshClick to read material · temperature

The building is a locator · only the selected 960 mm strip is solvedSection stretched horizontally

Baseline—

Design—

Inside surface temperature

Along the section height

A baselineB design

Height y / mm · top to bottomLook for the coldest spots.

Conditions and model checksMethod · assumptions · sources ＋

Section and boundaries

Steady 2D conduction, 960 mm high, per 1 m depth, on a 32 × 48 finite-volume mesh aligned to layer interfaces. Inside and outside air connect through surface resistances; top and bottom are adiabatic. Arrow lengths are normalised for display.

Properties and scope

Properties are assumed values for experimenting and can be replaced with product data. The slab is a band through the wall; the window is equivalent glass and frame layers. Glass Ug includes the standard surface resistances 0.13 + 0.04. Solar gain, air leakage, moisture and real frame geometry are not modelled.

Checks and AI

Checked against the exact plane-wall solution, energy balance and grid convergence. The AI learns a 2D correction to an independent 1D resistance profile, with separate training and test sections. It predicts the temperature field only; performance numbers come from the solver. Everything runs in this browser.

A simplified 2D steady-state calculation. It is not a design, certification, code-compliance or condensation check, nor a prediction of energy savings. AI error figures compare against the solver on this section and are not probabilities.