TP-02: Wrappers

Overview

The wrapper layer sits between the raw dwelling input data and the core HEM simulation engine. Its purpose is to enforce the standardised assumptions required by the Future Homes Standard (FHS), transforming a dwelling description into a fully configured simulation input. The wrapper does not perform energy calculations itself. It configures schedules, gains profiles, heating patterns, and system parameters so that the core engine can run under controlled, repeatable conditions.

Three assessment modes are defined:

1. Actual dwelling: the dwelling as designed, with standardised occupancy and usage patterns applied.
2. Notional dwelling: a reference building with the same geometry but prescribed fabric performance, systems, and controls, used to derive the Target Emission Rate (TER) and Target Primary Energy Rate (TPER).
3. Fabric Energy Efficiency (FEE): a variant that isolates the thermal performance of the building fabric by substituting idealised heating, cooling, and hot water systems.

All three modes share a common preprocessing pipeline that generates occupancy schedules, metabolic gains, lighting gains, appliance gains, DHW event schedules, heating and cooling control patterns, and ventilation settings. The notional and FEE modes then overlay further transformations to replace or constrain specific input parameters.

Inputs

Configuration methodology

Simulation time parameters

The wrapper fixes the simulation period and timestep for all assessment modes:

Occupancy calculation

The number of occupants $N$ is derived from the total floor area $A_{TFA}$ and the number of bedrooms $n_{bed}$ (capped at 5).

For a single bedroom dwelling, a sigmoid relationship applies:

$N=1+j\left(1-e^{k\cdot A_{TFA}^2}\right)$

where:

• $j=0.4373$
• $k=-0.001902$

For dwellings with two or more bedrooms, fixed values are used:

Metabolic gains

The total body surface area of occupants is calculated as:

$A_{body}=2.0001\times N^{0.8492}\text{[m²]}$

Half-hourly metabolic gain profiles (in W/m² of body surface area) are defined separately for weekdays and weekends. The absolute metabolic gain at each timestep is:

$Q_{met}(t)=A_{body}\times q_{met}(t)\text{[W]}$

where $q_{met}(t)$ is the profile value at timestep $t$. The resulting schedule repeats on a weekly cycle for 53 weeks.

Heating schedules

Heating setpoint temperatures are fixed:

The heating periods vary by zone type, day type, and control type:

Living room, weekday: 07:00 to 09:30 and 16:30 to 22:00.

Living room, weekend: 08:30 to 22:00.

Rest of dwelling, weekday (control type 2, separate temperature control): same as living room.

Rest of dwelling, weekday (control type 3, separate time and temperature control): 07:00 to 09:30 and 18:30 to 22:00.

Rest of dwelling, weekend: 08:30 to 22:00 (same as living room for all control types).

Outside the heating periods, the setpoint is null (heating off), unless a setback temperature is specified by the heating system.

Cooling schedules

Where a cooling system is present, the cooling setpoint is 24.0 °C. The schedule differs from heating:

Living room, weekday: 07:00 to 09:30 and 18:30 to 22:00.

Living room, weekend: 08:30 to 22:30.

Rest of dwelling (all days): 22:00 to 07:00 (night-time cooling only).

Water heating schedules

For storage-based hot water systems, the wrapper defines minimum and maximum temperature controls:

The two-hour hold at 60 °C provides time for the tank to reach and maintain the sterilisation temperature. Instantaneous systems (combi boilers, point-of-use heaters, HIUs) operate at 52 °C continuously with no scheduled control.

Hot water event scheduling

Daily volume estimate

The average daily hot water volume is derived from the number of occupants:

$V_{HW,daily}=0.70\times 60.3\times N^{0.71}\text{[litres]}$

The 0.70 factor removes an estimated 30% attributable to pipework losses present in the source monitoring data.

An electric shower correction factor is then applied. The original dataset under-represents showers because approximately 30% of monitored homes had an additional electric shower used at half the rate of the main shower. The correction factor is:

$F_{elec}=1-p_{shower,orig}+p_{shower,orig}\times (1+p_{IES}\times r_{IES})$

where:

• $p_{shower,orig}=0.60685$, the proportion of hot water volume due to showers in the original dataset
• $p_{IES}=0.3$, the proportion of homes with an additional electric shower
• $r_{IES}=0.5$, the relative usage rate of the electric shower

This gives $F_{elec}\approx 1.091$.

Event generation

Annual DHW events are generated stochastically using a decile-banding approach. The daily volume determines which of ten consumption deciles the dwelling falls into. Each decile has associated probability distributions for event counts, volumes, and timings by day of week and hour of day.

For each hour of each day of the year, the number of events of each type (shower, bath, other) is drawn from a Poisson distribution. A banding correction factor scales the Poisson rate parameter:

$F_{band}=\frac{V_{HW,daily}}{V_{cal,decile}}$

where $V_{cal,decile}$ is the calibration volume for the assigned decile. A fixed random seed (37) ensures reproducibility.

Event times are uniformly distributed within each hour. An overlap check prevents simultaneous shower and bath events by reassigning event times when collisions are detected.

Event duration adjustment

Event durations are adjusted by monthly behavioural factors from SAP 10.2 Table J5:

A volume calibration factor $F_{HW}$ normalises the generated event volumes to match the target annual volume:

$F_{HW}=\frac{365\times V_{HW,daily}}{V_{ref,annual}}$

where $V_{ref,annual}$ is the total volume from the uncalibrated event list.

For "other" (non-shower, non-bath) draw-offs, monthly factors from SAP 10.2 Table J2 apply instead, and a Part G bonus of 0.95 is applied when the dwelling meets the 125 l/day water use target.

Bath volume calculation

The fill volume of a bath accounts for occupant body displacement:

$V_{fill}=\left(V_{disp}+V_{std,fill}\right)\times \frac{S_{bath}}{S_{std}}-V_{disp}$

where:

• $V_{std,fill}=73$ litres, the standard fill volume
• $S_{std}=180$ litres, the standard bath size
• $S_{bath}$ is the actual bath size
• $V_{disp}$ is the average occupant displacement volume, estimated from average body mass

The average body mass is calculated from the adult/child split:

$N_{adults}=\frac{2.0001\times N^{0.8492}-1.07451\times N}{1.888074-1.07451}$

$N_{children}=N-N_{adults}$

$\bar{w}=\frac{78.6\times N_{adults}+33.01\times N_{children}}{N}$

Displacement volume equals body mass in kg (assuming density equal to water).

Event allocation

Events are distributed across the available outlets in the dwelling. If multiple showers are present, events cycle through them in sequence. The same applies to baths and other outlets. If no shower is present, shower events are allocated to baths, and vice versa. If neither shower nor bath is present, both event types are allocated as "other" draw-offs using a standard 180-litre tub at 8.0 l/min.

Mixed hot water temperatures are fixed at 41 °C for showers, baths, and other draw-offs.

Lighting gains

Annual lighting energy demand is calculated from the total floor area and occupancy:

$L=1139\times (A_{TFA}\times N{)}^{0.39}\text{[lumens]}$

The overall lighting efficacy is the area-weighted average of zone efficacies:

${\eta }_{light}=\frac{{\sum }_z{\eta }_z\times A_z}{{\sum }_z{A}_z}\text{[lm/W]}$

Annual energy consumption is then:

$E_{light}=\frac{L}{{\eta }_{light}}\text{[kWh/year]}$

Half-hourly demand profiles are defined per month (12 profiles of 48 values each). Each profile value is scaled by the daily energy consumption and a daylight correction factor that accounts for the glazing area, orientation, and solar conditions at each timestep.

A gains fraction of 0.85 is applied to exclude energy used for external lighting (security lights, garages, sheds).

Where the installed lighting capacity is insufficient to deliver the required lumens, a top-up lighting demand is added at an assumed efficacy of 21.3 lm/W.

Appliance gains

Appliance energy demand and internal gains are generated from a combination of usage frequency, energy per cycle, and time-of-day propensity profiles. The wrapper models the following appliance categories:

For event-based appliances, individual use events are generated stochastically using Poisson-distributed event counts at each timestep, shaped by the propensity profile. Event durations include random variation drawn from a normal distribution, with a convergence-adjusted scaling factor $F_{appliance}$ to ensure the total annual demand matches the target.

The annual energy demand for "other devices" (consumer electronics) is:

$E_{other}=30.0\times (N\times A_{TFA}{)}^{0.49}\text{[kWh/year]}$

Cooking demand is estimated from a top-down model:

$E_{cook,elec}=448\times 0.8+(171+98\times N)\times 0.2\text{[kWh/year]}$

This total is distributed across oven, hobs, microwave, and kettle in proportion to each appliance's share of mean annual demand.

Evaporative and cold water losses

Evaporative losses (negative internal gains) are proportional to occupancy:

$Q_{evap}=-25\times N\text{[W]}$

Cold water losses are similarly:

$Q_{cw}=-20\times N\text{[W]}$

Both are modulated by half-hourly daily profiles that vary by day of week, repeating for 52 full weeks plus one additional day.

Minimum ventilation rate

The minimum whole-dwelling ventilation rate follows Part F clause 1.24a, taking the greater of method A and method B:

${\dot{V}}_{min,A}=0.3\times A_{TFA}\text{[l/s]}$

Method B uses a lookup by bedroom count:

The minimum air change rate is:

$n_{min}=\frac{\max ({\dot{V}}_{min,A},{\dot{V}}_{min,B})}{V_{total}}\times \frac{3600}{1000}\text{[h⁻¹]}$

where $V_{total}$ is the total dwelling volume in m³.

Window opening and treatment schedules

Window opening is triggered when the internal air temperature exceeds 22 °C. Openability is unrestricted unless noise nuisance or security risk flags are set, in which case windows may only be opened during waking hours (07:00 to 23:00).

Curtain schedules follow occupant behaviour: curtains are opened during waking hours when the sun is above the horizon, closed during waking hours after sunset, and left unadjusted during sleeping hours. Blinds respond to solar irradiance thresholds: closing above 300 W/m² and opening below 200 W/m² during waking hours, with automatic systems responding continuously.

Post-processing: emissions and primary energy

After the core simulation completes, the wrapper applies fuel-specific emissions factors (kgCO2e/kWh) and primary energy factors (kWh/kWh delivered) to the energy import, export, and generation results for each energy supply. Electricity uses time-varying half-hourly factors; other fuels use constant annual factors.

The regulated emissions are calculated as:

$E_{reg}=E_{import}+E_{export}+E_{generated}-E_{unregulated}$

where unregulated energy covers appliances and cooking. The final metrics, normalised by total floor area, are:

• DER (Dwelling Emission Rate) for the actual dwelling, in kgCO2/m².
• TER (Target Emission Rate) for the notional dwelling, in kgCO2/m².
• DPER (Dwelling Primary Energy Rate) for the actual dwelling, in kWh/m².
• TPER (Target Primary Energy Rate) for the notional dwelling, in kWh/m².

Notional dwelling configuration

The notional dwelling retains the geometry and zone structure of the actual dwelling but replaces fabric performance, systems, and controls with prescribed values.

Notional fabric values

Ground floors additionally receive a floor construction thermal resistance $r_f=6.12$ m²K/W and a wall-floor junction linear thermal transmittance $\psi =0.16$ W/(mK).

Thermal bridge psi-values are set to SAP 10.2 Table R2 values. Point thermal bridges are set to zero.

Notional glazing limit

Total glazing area is capped at a fraction of TFA:

$f_{glaz,max}=0.25-f_{rooflight,correction}$

where the rooflight correction is:

$f_{rooflight,correction}=\max \left(0,\frac{A_{rooflight}}{A_{TFA}}\times \frac{{\bar{U}}_{rooflight}-1.2}{1.2}\right)$

If the actual glazing area exceeds this limit, all window and rooflight dimensions are reduced by a linear factor $\sqrt{A_{max}/A_{actual}}$, and the freed area is redistributed to opaque elements of the same orientation and pitch.

Notional infiltration and ventilation

All passive openings (PDUs, cowls, combustion appliances) are removed.

Notional heating system

The notional dwelling uses an air-source heat pump with EN 14825 test data at design flow temperatures of 35 °C and 55 °C. Capacity at each test point is scaled from the dwelling's design heat load. The heat pump feeds a wet distribution system with radiators sized for a 45 °C design flow temperature.

Radiator sizing follows:

$P_{rad}=c_{rad}\times ({\theta }_{flow,design}-{\theta }_{setpoint}{)}^n$

where $c_{rad}=1.89/50^n$, $n=1.34$, and the number of radiators per zone is $\lceil Q_{design,zone}/P_{rad}\rceil$.

Weather compensation control (ecodesign class 2) is applied, and a setback temperature of 18 °C is used.

For dwellings connected to a heat network, the notional heating system is replaced by an HIU with a maximum power of 45 kW, daily standing loss of 0.8 kWh, and building-level distribution losses of 62 W.

Notional DHW outlets

For houses with more than one storey under notional option A (excluding FEE), an instantaneous WWHRS (System B) is added with 50% heat recovery efficiency and 0.98 utilisation factor.

Notional lighting

Lighting efficacy is set to 120 lm/W for all zones.

Notional solar PV

Under notional option A (excluding FEE), a solar PV system is added. The peak power is based on the total floor area. PV is not added for notional option B or FEE assessments.

Notional on-site generation

Any on-site generation, PV diverters, and electric batteries from the actual dwelling are removed in the notional configuration.

Fabric Energy Efficiency configuration

The FEE assessment isolates fabric thermal performance by substituting idealised systems that do not interact with the fabric calculation. The dwelling geometry and fabric properties remain as designed.

FEE systems

Hot water distribution pipework is set to effectively zero (0.01 mm internal diameter, zero length). WWHRS, on-site generation, PV diverters, and electric batteries are removed. The heating control type is set to separate temperature control (type 2).

FEE metric

The Fabric Energy Efficiency is calculated in post-processing:

$FEE=\frac{Q_{heat}-Q_{cool}}{A_{TFA}}\text{[kWh/(m²cdotpyear)]}$

where $Q_{heat}$ is the annual space heating demand and $Q_{cool}$ is the annual space cooling demand (negative by convention, so the subtraction adds the cooling demand to the heating demand).

Outputs

Assumptions

• The number of bedrooms is capped at 5 for the purpose of the occupancy calculation. Dwellings with more than 5 bedrooms use the 5-bedroom occupancy value.
• Occupancy profiles are the same for weekdays and weekends, derived from the FHS appliance propensity data.
• Metabolic gains use separate weekday and weekend profiles, scaled by total body surface area.
• The hot water daily volume estimate is reduced by 30% to account for pipework losses present in the source monitoring data, and then uplifted by approximately 9.1% to correct for missing electric shower usage in the original sample.
• DHW event generation uses a fixed random seed (37) for reproducibility across all assessment runs.
• Mixed hot water delivery temperature is 41 °C for all outlet types.
• Hot water storage temperature is 52 °C for normal operation, with a daily sterilisation hold at 60 °C from midnight to 02:00.
• Occupant waking hours are 07:00 to 23:00, consistent with Part O assumptions.
• The Part G water use target bonus (5% reduction in "other" draw-off duration) applies when the dwelling achieves not more than 125 litres per person per day.
• The FEE calculation uses idealised systems with effectively unlimited capacity, so that the result depends solely on the building fabric and ventilation characteristics.
• Lighting gains exclude 15% of energy attributed to external or non-conditioned-space lighting.
• Monthly behavioural adjustment factors for DHW events follow SAP 10.2 Tables J2 and J5.
• Notional thermal bridge values follow SAP 10.2 Table R2.
• Electricity emissions and primary energy factors are time-varying (half-hourly); all other fuel factors are constant.

Cross-references

• TP-01: Overview and Climate Data -- climate data and simulation time parameters
• TP-03: External Conditions -- external temperature and solar data used in daylight factor calculation and window treatment schedules
• TP-04: Space Heating Demand -- heating setpoints and schedules feed into the zone heat balance
• TP-06: Ventilation and Infiltration -- minimum air change rate, MEV configuration, and window opening schedules
• TP-08: Solar Gains and Shading -- daylight correction factors for lighting gains; window treatment and blind schedules
• TP-09: Hot Water Demand -- DHW event schedules and daily volume calculation
• TP-10: Pipework and Ductwork Losses -- hot water distribution pipework configuration
• TP-12: Heat Pumps -- notional heat pump specification and EN 14825 test data
• TP-16: Heat Emitters -- notional radiator sizing and wet distribution system
• TP-17: Controls -- heating control types, setpoint time controls, and weather compensation
• TP-18: PV and Battery -- notional solar PV addition and removal of on-site generation in FEE mode