Authors:
Olivia Guerra-Santin, Assistant Professor in the chair of Smart Architectural Technologies in the Department of the Built Environment at Eindhoven University of Technology (TU/e)
L. Xu
S. Boess
E. van Beek.
Although building performance monitoring has been investigated for more than a couple of decades, it has yet to become a regular practice. In the few cases where buildings are monitored after renovation activities, not all aspects related to building performance are considered, which hinders the usefulness of the results, and the possibilities to learn from these investigations and projects.
A literature review on Dutch renovation projects from which data was publicly available showed that monitoring campaigns usually include measuring energy consumption (and production), recording thermostat settings, and some investigation into the satisfaction of the users with the renovation. Thermal comfort and air quality can also be found in some projects. However, the actual needs, preferences, activities of the occupants, and their interaction with systems, which are known to affect occupants’ behaviour and therefore energy performance, are unfortunately often neglected.
Furthermore, a common problem when evaluating the energy performance of buildings is that the benchmark or expected performance is often too optimistic. More knowledge of the building conditions before the renovation could help to clarify deviations in the energy performance, but there is often no information on the actual performance of the building before the renovation.
This article aims to contribute with practical monitoring guidelines to be used by practitioners, based on several monitoring case studies in Europe. An ideal monitoring plan is presented, but alterations to it and shortcuts are discussed considering that monitoring activities can be time-consuming and expensive, and thus, not all buildings can be monitored in the same manner.
The following scheme shows the current aspects that are monitored in practice in the Netherlands, and the methods used for gathering data.
In the Netherlands, at least 70% of the building residents in social housing projects must accept to take part in the renovation process. For this reason, in the pre-renovation and renovation phases, the focus of investigation lies on the occupants' satisfaction with the renovation process, including aspects such as participation in the process, communication activities, and the clarity of information given to the occupants. Thus, the contact with residents starts many months before the renovation activities take place. This is, of course, a great opportunity to involve the users in the design/renovation process but could also help to facilitate the monitoring of the buildings before the renovation.
In the post-renovation phase, the quality of information and communication with the users is sometimes still investigated, but the focus shifts towards energy consumption, indoor environmental quality, and sometimes on the maintenance of the building interfaces (e.g., of the mechanical ventilation system). However, occupants' behaviour is seldom monitored. In the cases when the use of the systems is different from the ‘expected behaviour’ (e.g., higher or lower thermostat setting), the reasons behind the differences are never investigated. Furthermore, the occupant-building interaction is hard to understand without a reference for previous behaviour or building performance.
The common data collection methods are building monitoring, surveys among residents, interviews with residents and interviews with professionals. Building monitoring (through the use of meters and sensors) is used for quantifiable variables, such as indoor conditions and energy consumption. Surveys and interviews with residents are used to investigate the satisfaction of the residents with different aspects of their own homes, including information given about maintenance and operation. Interviews are also used to determine the satisfaction with the renovation process as a whole, while interviews with professionals have the same purpose but are conducted indirectly through the opinions of the stakeholders involved in the process.
The following guidelines are suggested based on the lessons learned from several European monitoring projects, such as the current Dutch RVO funded project IEBB-Integrated energy transition in existing buildings, the BTA/Climate Kic funded projects 2ndSkin and BOCS, and the EU funded Low Carbon Built Environment programme, among others.
These are classified into: phases of monitoring (before and after), aspects to be monitored, and methods of data collection. In this article, we focus on aspects to evaluate building performance, thus the evaluation of the renovation process itself is out of the scope of this paper.
An inventory of the important equipment and appliances in the homes to be renovated should be made, paying particular attention to devices consuming large amounts of energy, such as electric heaters, sunbeds, etc. A talk/interview with the residents can help to identify these devices. It is also important to ask how often they are utilised.
Energy consumption – ideally from energy meters for one year (one heating season and one cooling season), but monthly energy readings or energy bills are also useful. In cases when both gas and electricity are used, both meters/readings/bills should be recorded. In cases when a long monitoring campaign is possible, energy consumption for space heating and domestic hot water should be sought.
This information will help to identify households with a higher energy consumption baseline as a consequence, for example, the presence at home during the whole day, a larger household, the participation in hobbies or lifestyles that consume more energy.
After renovation, it is becoming normal practice to install smart meters in all homes (dwellings or flats), since it would be easier to obtain the data to assess the performance, and to help users track of their own consumption. Individual meters (per home) are recommended to be able to determine the energy use per household for reasons related to performance evaluation but in some cases, also in relation to energy billing and contracts with housing associations. A common meter for a building block of flats is sometimes seen, especially in rental homes when the energy bill is included in the rent. However, this practice hinders both the understanding of the building performance and the ability to pinpoint and solve problems when they arise.
Ideally, meters could be installed for each of the different final energy uses (i.e., space heating, water heating, electrical appliances, etc.), but this practice is rare in residential projects. Metering (and access to the data) of locally produced energy in projects with photovoltaic panels is also relevant for the performance assessment.
The information on the total energy consumption (and production) can show what the actual energy performance of the building is. More detailed information on different end-uses in energy consumption can help to understand cases of underperformance, allowing for troubleshooting and maintenance during the occupancy phase of the building. On the other hand, monitoring the performance of all individual dwellings within a renovation project, instead of looking into the neighbourhood/building block as a whole, can help to determine the extent of variability of energy consumption, and remove their levelling effect on the final project performance.
A wide selection of plug meters is available in the market for monitoring before renovation. In cases, especially in pre-renovation monitoring, there might be a lack of resources to install meters in the dwellings, or perhaps these dwellings have collective meters. In these cases, looking into individual energy readings or bills can help if there metering is done per dwelling.
For collective metering dwellings, it might not be possible to determine the exact energy consumption per dwelling, and so, an inventory of existing appliances and electronics, knowledge on the use of heating and cooling systems and devices (temperature level but also for how long they are turned on), and window opening practices can help to determine the households that consume more energy, or, in an ideal scenario, use the information to perform building simulations to calculate the energy consumption before the renovation.
After renovation, permanent energy meters are recommended, especially if the information can be made available to the residents in ways that are understandable, for example in terms of energy costs, or by comparing their consumption with the previous situation (before the renovation) or with other similar households.
Measurements of at least temperature, relative humidity and CO2 should be taken. In cold climate countries these should include at least some cold winter days and in warm climate countries some warmer summer days. In view of climate change, the measurements should ideally include both extremes. CO2 level should be measured at least at one point in the dwellings, for example, the most problematic room; this could be a room where all the household members gather, or a small bedroom with two occupants. These measurements will help to identify the danger of overheating in summer, overall CO2 levels, and humidity problems.
During the post-renovation phase, temperature, relative humidity, and CO2 should be monitored. CO2 level should be measured at least in bedrooms, but in houses with large households, measurements can also be relevant in common areas (e.g., living room). In addition to assessing the indoor environmental conditions of the renovated home, these measurements will help to identify the danger of overheating in summer, the CO2 levels at night in bedrooms, which are particularly important in airtight homes, and issues with air dryness in the winter.
Commercial sensors can be used to accurately measure indoor parameters. CO2 sensors are often more expensive and thus, fewer sensors are usually used. In case of budget restrictions, we suggest using sensors with integrated CO2 in the most problematic areas (bedrooms and living room) and using temperature and Relative Humidity (RH) sensors for other areas. These last, more simple sensors have the advantage that they can work on batteries, and thus are less intrusive to the household members.
Sensors can be used in this phase to determine how often the residents’ open windows, the thermostat setting and setback, the use of individual radiators in different bedrooms, and their presence at home. However, the use of sensors is not always possible due to their cost, the time required for monitoring and data analysis, and due to their intrusiveness. Thus, in some cases, a questionnaire survey and diaries can help us to obtain the same information from the users.
This information can serve as a baseline for behaviour after renovation, allowing us to determine how habits, understanding of systems and interface instructions and the interfaces themselves could be affecting the performance of the building. Furthermore, this information can be used as input in building simulation to predict more accurately the energy performance of the dwelling after renovation.
Just like in the pre-renovation phase, either sensors or questionnaire surveys/diaries can be to determine the behaviour of people at home. However, in this phase, the focus also lies on investigating how the users interact with the new systems and interfaces installed, how aligned is this behaviour to the instructions/expectations of the designers and installers, and how well the residents understand the use of these new systems and interfaces. Thus, also information related to the instructions received by the residents during the handover of their homes, and written manuals or guidelines they have received, should be collected, and analysed.
Large deviations between the expected (assumed) and actual occupants’ behaviour can help to explain large energy consumption differences and can also help identifying problems with individual systems and interfaces (due to faults in the equipment or lack of commissioning). Unforeseen occupants’ behaviour, especially if it is also different to the pre-renovation behaviour (and therefore, not a habit), may indicate that the occupants are trying different ways to achieve their required level of comfort when the systems cannot provide it (for example using extra heating systems, or opening windows at night after the renovation).
A great variety of data collection methods can be used to capture the behaviours of people at home. Some of these are described in other sections, since energy meters, electrical plug meters, and indoor environment sensors can be used to ascertain the behaviour of people. For example, high CO2 levels and the use of energy measured by energy/plug meters can give an indication of the presence at home. Or an RH sensor in the bathroom can indicate when showers are being taken. Other, more specialised sensors can be used to monitor behaviour, for example, window switch sensors or surface temperature in radiators. Other sensors such as movement sensors, lighting sensors and noise sensors can be used to detect behaviour, however, more time-consuming data cleaning and analysis in necessary.
Furthermore, diaries, questionnaire surveys and walkthroughs can be used also to gather data related to behaviour. Diaries consist of asking residents to fill in formats (printed or digital) stating the (energy-related) activities done at home. Walkthroughs are interviews that are carried done at users’ home that focuses on how they experience different aspects of their home. They usually require the users to think about specific scenarios related to their daily activities.
These instruments are more time-consuming for the residents, but they can generate very good data for the analysis too. The selection of instruments requires consideration of the number of households to be included in the activities since the interviews (and walkthroughs) will take more time than a questionnaire survey.
Users’ needs and preferences may affect occupant behaviour and therefore energy and indoor performance. The set of factors may change over time, but interventions targeting ‘behavioural change’ or the interfaces used might not have an effect in the short term. These needs and preferences come from household’s lifestyle (e.g., full employment vs. homemakers vs. homeworking, etc.), background (e.g. having lived in a warmer/cooler country), health condition (older people, people with less mobility), etc. These household characteristics do not easily or rapidly change and therefore, the affected behaviour might also not change.
Questionnaire surveys and talks/interviews with the residents can be used to determine how much these factors might affect behaviour and building performance. Furthermore, these factors will highlight the main differences among households and also could be used to determine the most suitable renovation concepts for the specific households.
To further identify and understand the above-stated deviations in behaviour, we also need to know whether the behaviours are indeed driven by faults in the systems or lack of feedback from the building interfaces, or whether they are caused by specific needs and preferences of people. It is important to establish this difference since issues with the systems and interfaces can be rectified, but the users’ needs and preferences cannot be changed, and thus, a different solution must be sought.
The data collected during the pre-renovation phase can be used also during the post-renovation phase to carry out this analysis, except in cases where a new household has moved into the building.
To gather information on needs, preferences, and thermal comfort, we can use the same instruments selected for the data collection on occupants' behaviour. Usual questions are related to household size and main occupations (e.g., full-time employee, full-time students, part-time employee, homemaker, etc.), general ages of household members (the presence of children and older people have been found to affect energy consumption), general health condition of household members, etc.
There are a few methods to determine the thermal comfort of building occupants. Indoor and outdoor measure parameters data plus a few other assumptions can be used to calculate thermal comfort through the adaptive comfort model or the predicted mean vote. However, if personal thermal comfort is to be investigated, asking the users to rate their own comfort is necessary. For this, we can use a thermal perception scale. Residents can be asked to rate their thermal comfort in general during a specific season, or at specific times during the day so that these ratings can be coupled with measured indoor parameters. This information can give more insight into the preferences for the comfort of the residents of the buildings.
Given that some of this information is considered personal and sensitive, special care must be taken in carrying out the data collection, management and storage within the privacy and ethical regulations. According to the current data protection rules, the basic requirements are that participants must give explicit consent for their data to be collected, analysed, and stored; most be informed of the purpose of the study and how their data will be used, managed, and stored, participants have the right for erasure and the right to withdraw their participation at any time, without having to explain. To ensure the safety of the participants ' data, the data must be pseudonymised as soon as possible, and actual names and contact information (including address) must be kept on an encrypted file accessed only by a limited number of people. Furthermore, any written report should make sure that the individual participants cannot be identified from the data, for example from the apartment or household description.
Monitoring can be expensive, time-consuming and sometimes unpredictable. Devices fail, participants withdraw or data gets lost. Preparing a monitoring campaign also takes time, especially in recruiting the participants for the study. In practice, it might be impossible to monitor all projects and all renovated homes. However, the more that we learn from these projects, we might need to monitor less in the future. Furthermore, with the help of Artificial Intelligence, we might be able to monitor more effectively in the near future.
Keywords: energy performance of buildings; industrialised renovation; building performance monitoring; social housing; renovation projects; renovation process; data collection methods.
