«Prepared by BRE on behalf of the Department of Energy and Climate Change December 2013 BRE report number 286733a The EFUS has been undertaken by BRE ...»
Report 4: Main heating systems
Prepared by BRE on behalf of the
Department of Energy and Climate Change
BRE report number 286733a
The EFUS has been undertaken by BRE on behalf of the Department of Energy and Climate Change (DECC).
Report editors and lead authors: Jack Hulme, Adele Beaumont and Claire Summers.
Project directed by: John Riley and Jack Hulme.
Data manager: Mike Kay.
Supporting authors and analysts: Mike Kay, Busola Siyanbola, Tad Nowak, Peter Iles, Andrew Gemmell, John Hart, John Henderson, Afi Adjei, Lorna Hamilton, Caroline Buchanan, Helen Garrett, Charlotte Turner, Sharon Monahan, Janet Utley, Sara Coward, Vicky Yan & Matt Custard.
Additional thanks to the wider team of reviewers and contributors at BRE, DECC and elsewhere, including GfK NOP Social Research, Gemini Data Loggers, Consumer Futures, G4S, Eon, British Gas, and for the input of the Project Steering Group and Peer Reviewers.
Executive Summary The main aim of the 2011 Energy Follow-Up-Survey (EFUS) is to collect new data on patterns of household and dwelling energy use in order to update the current modelling assumptions about how energy is used in the home.
This report presents analysis of:
1. the data collected during the household interview on primary (main) space heating systems and usage
2. the results of the household heating patterns as determined from analysis of the EFUS temperature data
3. results from the collection of metered fuel consumption from the EFUS.
Analysis is based on the interview sample weighted to the national level, using a weighting factor specific to the interview sample. The results presented in this report are therefore representative of the English housing stock, with a population of 21.9 million households. The results of this analysis will be used to inform energy efficiency policy and to inform and update the assumptions in the BRE Domestic Energy Model (BREDEM) and the UK Standard Assessment Procedure (SAP).
The main conclusions resulting from the analysis are summarised below.
Heating Season The majority of householders report that they start heating their home on a regular daily § basis in October and finish sometime in March or April. The average (mean) length of the heating season is reported to be 5.6 months. These householder reported findings are supported by analysis of the temperature data.
Approximately 2% of households (the equivalent of 0.5 million households) report that they § heat their homes daily throughout the year. A similar proportion report that they do not heat their homes at any time of the year.
Daily/Weekly heating patterns for households heating in a regular manner The majority of households (16.0 million, 73%) state that they heat their homes in a regular § manner, that is they turn their heating on and off at set times of the day, although this pattern can change for different days of the week and at weekends.
Centrally heated homes
Almost 10% of centrally heated properties do not have a timer with which to control their § heating system. A further 23% have a timer but do not use it to control the system.
Nearly 35% of households using a timer to control a central heating system in a regular § manner report that they manually override the heating system to turn it off when it is timed to be on at least once every week or more.
The majority of regularly heated centrally heated households (70%) report that that they § have their heating come on twice per day. A further 21% have their heating on once per day, accounting for 91% of all regularly centrally heated households.
From the analysis of the November 2011, December 2011 and January 2012 temperature § data it can be seen that the majority of households (76%) maintain the same number of heating periods over these winter months.
60% of households with a central heating system controlled by a timer to give a regular § heating pattern report that they switch on their heating for an additional period of ‘boost’ heating at least once a week. This boost period is typically 1‐2 hours per day.
The average number of hours that the heating is on for a centrally heated household whose § home is regularly heated on a daily basis (excluding any boost heating) is 7.5 hours. Those that have their heating on once per day, typically have it on for 14.5 hours whereas those that have it on twice per day have it on for approximately half that time, typically for 2 hours in the first period and 5 hours in the second period.
Analysis of the temperature data to derive timings of heating patterns results in a good § agreement of the householder reported data. Using January 2012 as representative of a typical winter month, the temperature data shows that, for those households that heat once per day, the heating is typically on for just over 14 hours whereas those that heat twice per day typically have it on for approximately 2 hours in the first period and 6 hours in the second period during weekdays.
For centrally heated households, the median total number of hours of heating (including § boost heating) is 8.7 hours according to the householder reported data. This compares to
9.4 hours according to the temperature data. The use of secondary heating seems likely to bring these two estimates closer. Further analysis to combine the main and secondary heating hours reported by the householder may help in understanding the differences.
Analysis of the temperature data indicates that the average (median) daily hours of heating § increases by 1 hour between November 2011 and December 2011, and remains the same as December for January 2012.
The most common heating pattern, comprising between 32% (temperature logger data) and § 39% (interview data) of centrally heated households heating their homes in a regular manner, is one in which heating is on twice daily, first at a ‘wakeup time’ for 4 hours and then at ‘home-time’ for 4-10 hours.
Results from the temperature logger data also corroborates the conclusion drawn from the § analysis of the interview data in that although there is a shift in timings of heating at a weekend compared to a weekday for approximately 25% of the population, for the stock as a whole, the number of hours of heating at weekends remains approximately the same as for weekdays.
Initial bivariate comparisons indicate that factors such as dwelling type, region, tenure, age § of occupants and whether occupants are in during the day are likely to be predictor variables for the total number of heating hours. This could be explored further using a multivariate analysis to provide additional insights. It is of particular interest to note that households that are in during the day on weekdays report heating their homes for a median of 9.4 hours per day, compared to households that are out during the day on weekdays reporting heating of 8.0 hours per day. The median number of hours of heating reported by households that are in during the day is lower than often assumed values in many energy modelling applications. It is important to recognise, however, that many energy modelling applications are defined to aspirational, desirable or other standardised levels rather than attempting to modelling actual usage.
Non-centrally heated homes
For non-centrally heated households, the average (median) number of hours of heating on a § weekday is 13.0 hours according to the householder reported data and 12.5 hours according to the temperature data. The average (median) number of hours of heating on a weekend day 13.0 hours according to both datasets.
Households heating in a non-regular manner 27% of households (5.9 million) report that they do not heat their homes in a regular way, § that is to say that they either do not use the heating regularly on a weekly basis or, if they do, they do not use it at regular times on a daily basis.
Initial bivariate comparisons suggest that factors determining whether a household will heat § their home in a non-regular manner are likely to be the dwelling characteristics of type, floor area and heating system type and fuel, along with the household characteristics of tenure, household size, household income and under-occupancy status.
Extent of main heating Around 65% of households (14.3 million households) have one or more rooms that are not § heated by the main heating system. Of these, 82% have one or more rooms with no main heating, and 40% have one or more rooms with the main heating turned off.
The majority (68%) of ‘other’ rooms (cellars, attics, outbuildings etc. that are habitable and § with a power supply from the home) are not heated by the main heating system.
Conservatories, separate WCs, bedrooms, hallways and kitchens are more likely not to be heated by the main heating system than living rooms, dining rooms, studies and bathrooms.
Initial bivariate analyses suggest that the dwelling characteristics of dwelling age and type, § the type of main heating and number of insulation measures are likely to be underlying factors determining whether a household has one or more rooms not heated by the main heating system.
The only significant difference in the likelihood of a certain household characteristic group § having one or more rooms not heated by the main heating system is seen for fuel poverty status; households that are calculated to be fuel poor are more likely to have one or more rooms not heated by the main heating system compared to those households that are not fuel poor.
Achieved temperatures The average temperature to which the thermostat is set is reported to be 20°C.
§ Using the temperature data it can be concluded that the average temperature achieved for § the living room (zone 1) falls within the range of 19.7-20.4, with the average being 20.2°C, and the average temperature achieved for zone 2 falls within the range of 18.7-19.4, with the average being 19.1°C.
Using the temperature data, the temperatures achieved after a significant period of heating § are higher among older households, and for households living in dwellings with at least some level of insulation present. This latter finding provides some evidence of occupant ‘takeback’ of energy savings following insulation.
Metered fuel consumption 50% of households with mains gas central heating systems use between 10,000 and 20,000 § kWh of gas per year. The median consumption is significantly lower for households heating in a non-regular manner compared to those heating in a regular way.
50% of households using electrical heating systems (storage heaters and room heaters) use § between 4,000 and 9,500 kWh of electricity per year.
The analysis has highlighted a number of areas when the SAP and typical BREDEM energy modelling assumptions differ from those reported by households and seen in the temperature data. Future development of these methodologies should consider whether these assumptions should be revised in light of these findings. It should, however, be recognised that assumptions used in energy models are often set at an aspirational standard (for health or warmth for example), or a standardised value for comparison, rather than attempting to model actual usage.
Although there are likely to be some inherent uncertainties in the temperature data due to the difficulty in determining the heating patterns accurately and also limitations in the ‘snapshot’ of responses given by householders in the interview survey, if the results from the householder reported data and the temperature loggers data are taken in combination the primary areas where
SAP assumptions differ from those reported by households are:
The average (mean) length of the heating season, as derived from the householders’ § interview responses is 5.6 months. There is no statistically significant difference between this and the mean heating season length derived from the temperature data, which is 5.7 months.
The results from both the household interview survey and the temperature data provide § evidence that suggests that the 8 month heating season currently used in SAP (October to May) may be an overestimate of at least 1 month, possibly 2 months, compared to actual heating seasons, although this could be influenced by the milder than usual spring temperatures recorded in 2011.
Currently, SAP 2009 implements a heating pattern of 9 hours for weekdays and 16 hours for § weekends in the living room for all heating system types. The results presented in this report suggest that the weekend hours are being overestimated in SAP and that the weekend hours of heating should be the same as the weekday hours.
Analysis undertaken in this report suggests that the 9 hours of heating currently used in SAP § remains a reasonable approximation for centrally heated dwellings. The results from this analysis do suggest that SAP may underestimate the number of hours of heating in noncentrally heated dwellings.
Households that are in during the day on weekdays report heating their homes for a longer § period of time (median 9.4 hours per day) than households that are not in during the day on weekdays (median 8 hours per day). This difference is lower than the typically assumed standards in many energy modelling applications which attempt to account for household occupancy.
Currently, SAP 2009 implements a demand temperature of 21°C in zone 1 and 18-21°C in § zone 2. The results from the temperature data generally support these modelling assumptions, although the zone 1 temperature used in SAP is approximately 0.8°C higher than the average living room achieved temperature from the EFUS temperature data.
The temperature data shows that for those households heating twice a day, the first period § of heating is typically for a short interval and the time that the heating is on for in many households is not sufficient to bring the room temperatures to the required temperatures.
This finding differs to the current SAP methodology which assumes the demand temperature is met during the shorter heating period.
Table of Contents 1 Introduction
2.1 EFUS Interview Survey
2.1.1 Data collection and processing
2.1.2 Data Quality
2.2 EFUS Temperature logger data
2.2.1 Data collection and processing
2.2.2 Data quality
2.3 Weighting Factors
2.4 Calculating confidence intervals