What is building energy simulation?

Hour-by-hour calculation of the energy needed for heating and cooling, accounting for heat flow through the envelope (walls, roof, floor, windows), ventilation, solar gain, internal gains and thermal mass. It's used at design stage to compare insulation and equipment and to check energy-code compliance. EnergyPlus is a representative program.

Can annual heating demand be estimated simply?

Yes, by the degree-day method: multiply the heat-loss coefficient H (sum of envelope U×A plus ventilation) by heating degree-days. For a house with 150 m² of walls and so on, older insulation gives H = 386 W/K and about 32,400 kWh a year at 3,500 degree-days. Solar gain, internal gains and thermal mass, however, need hourly calculation to capture properly.

What changes with better insulation?

The heat-loss coefficient falls 58% to 163 W/K, and to 122 W/K (68% lower) with heat-recovery ventilation. As insulation improves, ventilation's share of heat loss rises from 13% to 31%, so heat recovery becomes worthwhile. Internal gains of 400 W also lower the balance-point temperature by 2.5–3.3 K, shortening the heating season itself.

Building Energy Simulation

Category: Thermal Analysis – HVAC | Revised 2026-10-01

CAE visualization for building energy theory - technical simulation diagram

Theory: adding up the heat-loss paths

Overview

🙋

How much does better insulation cut a house's heating bill?

🎓

First add up every path by which heat leaves. For walls, roof, floor and windows, multiply heat transmittance (U-value) by area, then add the heat carried off by ventilation air. That's the heat-loss coefficient H, in watts lost per kelvin of indoor–outdoor difference. Annual heating demand is roughly H times heating degree-days, the indoor–outdoor difference accumulated over the heating season. In reality sun, appliances and occupants help heat the house and thermal mass smooths temperature swings, so precise answers need hour-by-hour simulation.

Heat-loss coefficient

$$ H = \sum_i U_i A_i + \rho c_p\, n\, V\,(1 - \eta_{HR}),\qquad \rho c_p \approx 0.34\ \mathrm{Wh/(m^3K)} $$

n is air changes per hour, V the interior volume and η_HR the heat-recovery efficiency.

Degree-days and peak load

$$ Q_{year} = \frac{24\,H\cdot HDD}{1000}\ \mathrm{[kWh]},\qquad P_{peak} = H\,(T_i - T_{o,design}),\qquad \Delta T_{bp} = \frac{Q_{gain}}{H} $$

HDD is heating degree-days (K·day) and ΔT_bp the drop in balance-point temperature due to internal gains Q_gain.

Coffee Break Trivia

Programs born of the oil crisis

Hourly building-energy programs were developed in earnest after the 1970s oil crises. In the US, DOE-2 and BLAST were created and merged into EnergyPlus, released in 2001. Japan developed air-conditioning load programs such as HASP in the same era. All use hourly weather data and step through conduction within walls and solar gains in sequence.

Worked examples

Example 1: house heat-loss coefficient (walls 150, roof 60, floor 60, windows 25 m²; 300 m³; 0.5 ACH)

Specification	U (wall, roof, floor, window)	H	Window share	Ventilation share
Older insulation	0.9 / 0.6 / 0.8 / 4.65	386 W/K	30%	13%
Improved insulation	0.3 / 0.15 / 0.3 / 1.6	163 W/K	25%	31%
Improved + 80% heat recovery	0.3 / 0.15 / 0.3 / 1.6	122 W/K	33%	8%

Example 2: annual and peak heating load (indoor 20 °C)

Specification	1,500 HDD	3,500 HDD	Peak (−2 °C)	Peak (−15 °C)
Older insulation	13,905 kWh	32,445 kWh	8.5 kW	13.5 kW
Improved insulation	5,868 kWh	13,692 kWh	3.6 kW	5.7 kW
Improved + heat recovery	4,399 kWh	10,265 kWh	2.7 kW	4.3 kW

With a heat pump of COP 3, electricity at 3,500 HDD is about 10,800, 4,600 and 3,400 kWh.

🙋

Better insulation makes ventilation's share bigger.

🎓

Ventilation loss (51 W/K) doesn't change with insulation, so it stands out as everything else shrinks — 13% in the older house, 31% after improvement. Then recovering exhaust heat with heat-recovery ventilation pays off. Cutting ventilation rates to reduce loss, though, harms indoor air quality and should be avoided. Internal gains matter too: with an average 400 W from appliances and people, the insulated house's balance point drops 2.5 K, shortening the heating season. Including this in the degree-day method requires lowering the base temperature. Solar gains behave similarly and matter more in well-insulated houses, so evaluate them with hourly simulation.

Simulation workflow

Hand-calculate the heat-loss coefficient from envelope areas, U-values and ventilation as a benchmark.
Prepare hourly weather data (temperature, solar) for the site.
Enter wall layers (conductivity, density, specific heat) to capture thermal mass.
Set internal gains from occupancy, lighting and equipment schedules.
Compare annual results with the hand degree-day estimate for gross discrepancies.

Coffee Break Trivia

“Heating costs double the estimate”

An insulation-upgraded house used nearly twice the simulated heating energy. The simulation assumed the design ventilation of 0.5 ACH, but leakage through gaps around window frames and pipe penetrations was large; an airtightness test showed effective air change above 1 per hour. In a well-insulated house ventilation is a large share of loss, so infiltration hit directly. Redoing the air sealing brought heating energy close to the calculation.

Common mistakes

Mistakes and fixes

Mistake	Effect	Fix
Ignoring infiltration	Heating load underestimated	Include airtightness
Ignoring thermal bridges	Heat loss underestimated	Add linear transmittances
Fixed degree-day base	Overestimate for tight houses	Correct for gains
Wrong-site weather	Loads far off	Use site data
Cutting ventilation to save energy	Air quality suffers	Heat-recovery ventilation

🙋

I'd like to learn related topics.

🎓

Rate this article

Thank you for your feedback!

Helpful

More details

Report error

Helpful

More details

Report error

Written by the NovaSolver Editorial Team
Built with engineers and AI (editorial policy) — Sitemap

About the Authors

Building Energy Simulation

Theory: adding up the heat-loss paths

Overview

Heat-loss coefficient

Degree-days and peak load

Programs born of the oil crisis

Worked examples

Example 1: house heat-loss coefficient (walls 150, roof 60, floor 60, windows 25 m²; 300 m³; 0.5 ACH)

Example 2: annual and peak heating load (indoor 20 °C)

Simulation workflow

“Heating costs double the estimate”

Common mistakes

Mistakes and fixes

Related Topics

Related fields