Thermal zoning for building HVAC design and energy simulation: A literature review

Energy & Buildings 203 (2019) 109429

Contents lists available at ScienceDirect

Energy & Buildings journal homepage: www.elsevier.com/locate/enbuild

Thermal zoning for building HVAC design and energy simulation: A literature review Minjae Shin a,∗, Jeff S. Haberl b a b

Department of Civil, Construction, and Environmental Engineering, College of Engineering, The University of Alabama, Tuscaloosa, AL 35487, USA Department of Architecture, College of Architecture, Texas A&M University, College Station, TX 77843, USA

a r t i c l e

i n f o

Article history: Received 26 April 2019 Revised 6 August 2019 Accepted 11 September 2019 Available online 11 September 2019 Keywords: Building HVAC design Building energy simulation Thermal zone Thermal zoning method Indoor temperature profile

a b s t r a c t Building energy simulation programs can be useful tools in evaluating building energy performance during a building’s lifecycle, both at the design and operation stages. In addition, simulating building energy usage has become a key strategy in designing high performance buildings that can better meet the needs of society without consuming excess resources. Therefore, it is important to provide accurate predictions of building energy performance in building design and construction projects. Although many previous studies have addressed the accuracy of building energy simulations, very few studies of this subject have mentioned the importance of Heating, Ventilation, and Air-Conditioning (HVAC) thermal zoning strategies to sustainable building design. This research provides a systematic literature review of building thermal zoning for building energy simulation. This work also reviews previous definitions of HVAC thermal zoning and its application in building energy simulation programs, including those appearing in earlier studies of the development of new thermal zoning methods for simulation modeling. The results indicate that future research is needed to develop a well-documented and accurate thermal zoning method capable of assisting designers with their building energy simulation needs. © 2019 Elsevier B.V. All rights reserved.

1. Introduction Currently, one of the largest consumer energy sectors worldwide is the buildings sector. In the United States, buildings account for approximately 40% of the source energy used nationwide, if one includes thermal waste from non-renewable electricity generation [1]. A recent study also showed that buildings and construction together account for 36% of global final energy use and 39% of energy-related carbon dioxide (CO2 ) emissions, when upstream power generation is included [2]. Moreover, HVAC-related energy use represents 50% of building energy consumption and 20% of total energy consumption in the US [3]. In the construction sector, it is conventional practice to provide a final product (i.e., the building) without full testing [4]. In general, when building construction is completed, ownership and operation are transferred to the owner, sometimes without feedback with regards to operational performance [5]. This can be a critical problem, since there are no perfect systems or products. Therefore, a quality improvement and enhancement process are needed at the product (i.e., building) design stage that offers constant feed-

∗

Corresponding author. E-mail address: [email protected] (M. Shin).

https://doi.org/10.1016/j.enbuild.2019.109429 0378-7788/© 2019 Elsevier B.V. All rights reserved.

back to product designers regarding whether or not their creation performs as expected. Consequently, in order to design and produce higher quality structures, feedback from building operational performance measurements is needed to predict the accuracy of the performance of buildings during the design stage. Hourly whole-building energy simulation modeling during the building design and construction process would offer opportunities to test the performance of a building envelope and HVAC system prior to completion of the building. Such building energy performance simulation programs would be useful tools for evaluating building energy performance during a building’s lifecycle, both at the design and operation stages [6]. Simulating the annual energy usage of buildings has become a key strategy in designing high-performance buildings that can best meet the needs of society. The automated exchange of data between the architect’s design software and energy consultant’s building energy simulation program is essential to the future of the building design process. Several leading Computer-Aided Design (CAD) vendors now offer Building Information Modeling (BIM) software that they claim is capable of accurately simulating building energy use, and even automatically generating HVAC thermal zones in a proposed design. For instance, one commercially available schematic design tool, Autodesk Revit [7], has a feature for

2

M. Shin and J.S. Haberl / Energy & Buildings 203 (2019) 109429

automatically dividing a building’s geometry into perimeter and core zones at each floor, based on ASHRAE 90.1–2016, Appendix G [8]. However, Autodesk has not released the details of their zoning algorithm or provided a discussion of how the procedure works in their software. Therefore, how can HVAC design engineers know if the program accurately represents their designs? Furthermore, the American Society of Heating, Refrigerating, and Air-Conditioning Engineers (ASHRAE) does not provide guidelines regarding the methods for the correct thermal zoning of producing a complete thermal model from printed floor plans, CAD models, or BIM applications, which includes instructions about thermal zoning. The literature shows that significant progress has been made towards the development of integrated design tools; this has primarily been motivated by large software providers and their development of new software and adaptation of existing software programs [9]. Unfortunately, these new tools still overly simplify when it comes to specifying the inputs for Building Energy Simulations (BES). ASHRAE Research Project 1468 [10–12] made several significant contributions regarding the interoperability of BIM and BES software. To accomplish this, the RP-1468 project conducted empirical sensitivity tests of multiple options for different envelope components of BES models. These tests provided examples of how to solve select difficult modeling problems, such as facades with curved surfaces and shaded windows. Based on the results of this project, a conceptual thermal BIM model was able to generate instructions for creating an input file for a BES from a BIM, with a high level of accuracy. However, this project did not provide specific advice about building thermal zoning issues, which is an important feature in today’s buildings and can have a significant impact on their thermal behavior and energy consumption. As has previously been mentioned, several of today’s BIM tools automatically produce default building thermal zoning in the required BES electronic formats. However, these same models do not provide detailed documentation regarding how their algorithm(s) work, nor any guidance on how to create and evaluate building thermal zones in a BES during the early stages of the design process. Rather, in some cases they rely solely on the user to select the thermal zones. 2. Methodology In this study, a comprehensive literature review was conducted regarding the selection of a building thermal zoning technique for HVAC design and building energy simulation, particularly focusing on features, methods, and procedures. Specifically, the objectives of this study were to: (1) introduce fundamental basic concepts, together with definitions related to thermal zoning obtained from the literature; (2) summarize the various thermal zoning methods and practical applications of HVAC system design and building energy analysis; (3) investigate the impact of thermal zoning strategies during the building energy modeling process; and (4) suggest future developments and their potential research significance related to the new thermal zoning method for building energy simulation. A literature review was carried out using the academic search engines Google Scholar, Mendeley, and Scopus, with “thermal zone,” “thermal zoning,” “HVAC design,” “building energy simulation,” and “indoor temperature profile” serving as main keywords. In addition to research articles, different types of literature such as textbooks, standards, and HVAC handbooks were also reviewed with respect to thermal zoning and related definitions, methods, and applications.

This work is organized into five sections. Section 1 provides an introduction and background regarding the importance of thermal zoning in building energy simulation. Section 2 describes the overall methodology used to conduct this study. Section 3 contains a literature review of previous research and information related to this study, including a review of building energy simulation tools, definition of building thermal zones, thermal zoning related to HVAC design and building energy simulation, and building thermal zoning methods for building energy simulation. Section 4 discusses several promising research directions for thermal zoning methods in building energy simulation. Finally, the conclusions drawn are described in Section 5. 3. Building thermal zoning The literature has called building thermal zones by a variety of different names, such as thermal zones [8], thermal blocks [13], and HVAC zones [13]. However, these different terms all indicate the same idea and concept. In ASHRAE Standard 90.1–2016 (ASHRAE 2016, p. 12), an HVAC zone is clearly defined as “a space or group of spaces within a building with heating and cooling requirements that are sufficiently similar so that desired conditions (e.g., temperature) can be maintained throughout using a single sensor (e.g., thermostat or temperature sensor).” In addition to ASHRAE Standard 90.1-2016, the Commercial Energy Services Network (COMNET) has also developed commercial building energy modeling guidelines and procedures (COMNET 2010, pp. 2-2) that provide the following definitions of thermal blocks and HVAC zones. In COMNET, an HVAC zone is “a physical space within the building that has its own thermostat and zonal HVAC system for maintaining thermal comfort” (COMNET 2010, pp. 22). HVAC zones are usually identified in the HVAC plans for a new building. An HVAC zone should not be split between different thermal blocks. However, a thermal block may include more than one HVAC zone. In COMNET, a thermal block is “a space or collection of spaces within a building having sufficiently similar space-conditioning requirements so that those conditions could be maintained with a single thermal controlling device such as a thermostat” (COMNET 2010, pp. 2-2). In addition, a thermal block is a heat transfer concept and not always a geometric notion; therefore, spaces need not be contiguous to be combined within a single thermal block. However, they are controlled by a single thermostat. In summary, a building thermal block or thermal zone is the portion of a building that is controlled and maintained by a single thermostat sensor that has its own setpoint and schedule. All HVAC zones in that thermal block should maintain the same temperature throughout the day. However, in reality, not all spaces in a thermal zone maintain the same temperature. The concept of thermal zoning can be separated in terms of HVAC design and building energy simulation, since both are required to analyze a building. Therefore, the importance of the distinction of thermal zoning in HVAC design versus thermal zoning in building energy simulation is discussed below. 3.1. Thermal zoning in HVAC design In addition to temperature, there are several other parameters that should be maintained in a single thermal zone, such as humidity, outside air ventilation, operating periods, and pressurization. The most common reason for variations in thermal zoning in HVAC design is variations in thermal load [14]. However, designing buildings with one thermal zone for each room is not practical for most HVAC systems and can be costly to design, sim-

M. Shin and J.S. Haberl / Energy & Buildings 203 (2019) 109429

ulate, and construct. Therefore, HVAC engineers must determine whether an individual indoor space or groups of adjacent indoor spaces in a building have similar thermal loads prior to the selection of an HVAC system [15]. In this section, the relevant literature on thermal zoning in HVAC design is described, including: Gay and Fawcett [16], Bovay [17], Kreider [18], Bachman [19], McDowall [14], Hamrick [20], and Grondzik and Kwok [15]. The term “zoning” or “thermal zone” can be found in the earliest HVAC design literature. For example, Gay and Fawcett [16] considered thermal zone control strategies for both small and large commercial buildings. They argued that buildings should be divided into thermal zones for greater accuracy of HVAC control, since various indoor environmental conditions can exist in a building, including: (1) exposure to prevailing winds and solar radiation and the degree of shelter from surroundings; (2) varying rates of occupancy, depending on the activities and schedules of occupants; and (3) various methods of construction in different sections, producing unequal heating requirements. Bovay [17] introduced a thermal zoning procedure for the HVAC design of large commercial buildings. This was needed because HVAC systems must be capable of satisfying ever-changing loads, every day of the year. In this procedure, each space in an exposure has its own individual requirements. It was also recommended that a load profile be developed for any desired zone or space as a function of the outdoor temperature. The procedure provides a visual picture of the load requirements that must be met by an HVAC system across a wide range of conditions. Unfortunately, the procedure stated that exterior zones would widely fluctuate in terms of cooling requirements, due to variable lighting and human occupancy, the diurnal variations of outdoor dry-bulb temperature, sun and cloud cover, and shading from adjacent buildings imposing a significant and continually changing load on exterior zones. Bovay’s procedure also stated that interior zones are usually isolated from the variable loads caused by outdoor weather, except for outdoor air supplied for ventilation, which is treated by specialized equipment that is not in the space itself. Therefore, the primary loads include heat from lights, business machines, and people. The interior zone loads experience very little variation; quite often, these loads are constant. Kreider [18] also included information about thermal zoning. He argued that even when the entire building is kept at the same temperature, a multi-zone analysis is necessary if the spatial distribution of heat gains in different zones is non-uniform. For example, such would be the case for exterior perimeter zones if the facades of the building face different orientations (i.e., north, east, south, and west). As another example, consider a well-mixed, single-zone building with large exterior windows on the north and south sides, during a sunny winter day when the solar heat gain just balances the total heat loss. In such a case, neither heating nor cooling would be required for the entire building, according to a one-zone building energy simulation analysis. However, in most buildings, moving the excess heat from an area near a south-facing window to an area with a north-facing window for only a few hours at the end of the day is impractical. In addition, Kreider stated that the basic criterion for zoning is the ability to control indoor comfort conditions. Therefore, in choosing the zones for a multi-zone analysis, the designer should try to match the heating and cooling supply to the distribution of heat gains and losses. The most common and important division is between the interior and perimeter zones, because the interior of a building is not exposed to the changing environment. In addition, different facades of the perimeter should be considered separately for cooling load calculations, since solar heat gain can vary by orientation, depending on the time of day.

3

Bachman [19] also offered several pieces of advice regarding thermal zoning, suggesting five characteristics that would allow it to satisfy the thermal loads at varying times of peak gain in all rooms in a zone, to any degree of uniformity: (1) Similar solar exposure and orientation: East- and westfacing rooms will have vastly different schedules of thermal needs, just as rooms with large window areas will have different needs from rooms with smaller windows. (2) Similar envelope exposure: Perimeter rooms with exposure to the outdoor environment through the exterior envelope will have different heating and cooling needs than rooms in the core of the building, which will always need cooling in all but the most extreme cold climates because they have no direct means of heat loss to the exterior of the building. (3) Similar occupancy type and density: Zones such as libraries and auditoriums should be grouped in different thermal zones because they have significantly different 24-hour profiles for internal loads. Likewise, adjacent private offices and large classrooms should not share a thermostat, since these two zones can have very different internal loads and ventilation requirements. (4) Similar schedules: For example, weekday classrooms in a church school would not be in the same thermal zone as the Sunday congregational assembly space. In a similar fashion, offices with weekend use where cooling systems may be activated for the comfort of a few workers should not be zoned together with large lobby spaces where there are no workers on the weekends. (5) Shared incremental capacity: Where multiple, modular HVAC systems are used, it is common engineering practice to select small package units and distribute them as needed across the different thermal zones of the building. Retail buildings typically use modular rooftop HVAC units of about eight to 10 tons in cooling capacity and divide the retail floor area into thermal zones of an appropriate size. McDowall [14] also covered thermal zoning design considerations in terms of thermal variations, as follows: (1) Solar gain: Solar gain through windows can create a significant difference in cooling loads or the need for heating at varying times of the day, according to window orientation, season, and the prevailing weather conditions. (2) Wall or roof heat gains and losses: Thermal zones directly under the roof in a multi-floor building will experience more heat gain in the summer and heat loss in the winter than similar thermal zones directly below the upper floor. (3) Occupancy: The use of multiple zones and the importance of maintaining good temperature control will influence how critical thermal zoning is. (4) Equipment and associated heat loads: Equipment that gives off significant heat may require a separate thermal zone in order to maintain a reasonable temperature for the occupants. For example, a row of private offices may work well as a single thermal zone, but the addition of a significant number of computers in one of those offices might make it very warm compared to the other offices. Therefore, that office might then require a separate thermal zone. Hamrick [20] also demonstrated the impact of solar loading on building operations. According to Hamrick, solar loading is an important aspect of the operation of a building. As the sun travels across the sky during the day, the amount of solar energy that is absorbed on each of the exterior thermal zones varies. In addition, the south-facing side of a building will see different amounts of

4

M. Shin and J.S. Haberl / Energy & Buildings 203 (2019) 109429

solar energy at different times of the year because of the seasonal variation of the sun’s path across the sky. A typical floor in a multistory building in midsummer was used as an example to demonstrate how only interior thermal zones would use a constant volume of air for a constant occupancy load, regardless of varying solar loads. Finally, Grondzik and Kwok [15] also provided advice about thermal zoning. In their book they offered the following three influential factors on thermal zoning as follows: (1) Function: This is particularly important because of the variations in internal heat gain due to different thermal zone functions. Zone functions may also influence the layout of thermal zoning in a building. (2) Schedule: This is closely related to zone function. If one activity has operating hours different from those of the remainder of the building, a separate mechanical system should be provided for thermal zone(s) with different activities. Otherwise, large equipment required to serve the whole building at peak loads will be too large for moderate loads and inefficient at providing heating, dehumidifying, or cooling in a single thermal zone. (3) Orientation: The authors stated that the degree of exposure to daylight, direct sun, and wind are also important to thermal zoning. In the book, a single floor in a square, multi-story office building was used as an example. The authors argued that on cold, sunny, windy days, the perimeter spaces with direct sun through the windows would gain more heat than they lost and, thus, would require cooling. This cooling might be accomplished by opening the windows in the zone when the outdoor conditions permit. However, too much cold air from the open windows might cause discomfort to occupants near those windows. In addition, perimeter spaces on the same floor but without any direct sun might suffer simultaneous heat loss due to heat dissipating through the walls, glass, infiltration system, and any lack of electric lighting (i.e., internal load). Therefore, the authors argued that under certain conditions, such spaces would require heat in one part of the building from a mechanical system at the same time that an adjacent perimeter zone on the same floor would require cooling. Therefore, these spaces might also need cooling from a mechanical system, even when it was cold outside (i.e., when outside air temperatures were well below the thermostat setpoint temperature). In summary, the above literature regarding thermal zoning in HVAC system design provides select criteria for the division of thermal zones, including: (a) solar gain, (b) orientation, (c) occupancy, (d) schedule, and (e) space function. The concept of building thermal zones in the HVAC design process can be clearly described using these factors. However, all of the criteria reviewed only discussed qualitative attributes of thermal zoning in HVAC system design. None of the previous literature provided a general purpose or quantitative method for the selection of thermal zones. 3.2. Thermal zoning in whole-building energy simulation Whole-building energy simulation programs (e.g., DOE-2.1e, DOE-2.2/eQUEST, EnergyPlus, TRNSYS, etc.) are widely used to evaluate building energy performance during the early design phase. During the building energy modeling process, a large number of input parameters must be determined by the software’s user. One important input for producing accurate and reliable results from the analysis is thermal zoning. However, although thermal zoning strategies based on standard practices do exist, there is currently no well-documented, standardized method of thermal zoning for

all types of buildings that could be analyzed by a whole-building energy simulation program. One example of a standard thermal zoning strategy can be found in Appendix G of ASHRAE Standard 90.1-2016 [8], which provides general thermal zoning guidelines for generic building HVAC systems. In Appendix G, the basic rules for thermal zoning for a whole-building energy simulation include the following: (a) Separate interior and perimeter spaces: Assign separate thermal blocks to interior spaces located more than 15 feet from an exterior wall and to perimeter spaces within 15 feet of the exterior. (b) Separate orientations with significant amounts of glazing into separate zones, including: Glazed exterior walls, which should be assigned to different perimeter thermal blocks for each major orientation (i.e., north, east, south, and west). Glazed orientations within 45° of each other may be combined. Spaces with two or more glazed orientations, such as corner offices, should be separated from glazed thermal zones having different orientations. (c) Separate top, bottom, and middle floors: Spaces exposed to ambient conditions such as the top floor or an overhanging floor, and spaces in contact with the ground such as the ground floor, should be separately zoned from thermal zones exposed to ambient conditions, such as intermediate floors in a multi-story building. Fig. 1 shows examples of thermal zoning layouts that follow the thermal zoning guidelines of ASHRAE Standard 90.1-2016. The International Building Performance Simulation Association (IBPSA) also provides a few simple criteria regarding thermal zoning for use when a building energy simulation model is created [21]. These include: usage, temperature control, solar gain, perimeter or interior location, HVAC distribution system type, and separate interior and perimeter thermal zones. (a) Usage: Any rooms that are combined into a single thermal zone should have similar internal loads (i.e., people, lights, and equipment) and usage schedules. For example, it would not be appropriate to put a high-density variable-occupancy conference room in the same zone as a regular moderatedensity office space that has constant occupancy. (b) Temperature control: Any rooms that are combined into a single thermal zone should have the same heating and cooling setpoints and the same thermostat schedules. Since a thermal zone is controlled by one thermostat, it is imperative that all rooms in that thermal zone have the same temperature setpoints. (c) Solar gain: Any rooms combined into a single thermal zone should have similar solar gains. Shading should also be considered when determining thermal zoning according to solar exposure. At a minimum, for perimeter zones with glazed openings, there should be at least one thermal zone for each compass direction. For additional accuracy, include a thermal zone for any fenestration that varies by 45° or more. Unglazed exterior zones can be combined if the other criteria are satisfied. (d) Perimeter or interior location: Perimeter areas should be thermally zoned as separate from the interior spaces, with the depth of perimeter zoning typically 3.7 m (12 ft) to 4.6 m (15 ft) from the exterior wall. This is important as the heating and cooling requirements can vary greatly; perimeter thermal zones can require winter heating, while in the same building, core thermal zones with no exterior exposure can require year-round cooling. (e) HVAC distribution system type: In a building energy model, one cannot combine thermal zones if they are served by

M. Shin and J.S. Haberl / Energy & Buildings 203 (2019) 109429

5

Fig. 1. Examples of thermal zoning layouts with different building floor plan shapes.

different HVAC distribution system types (i.e., a radiant floor versus a fan coil unit). Since the entire thermal zone would be assigned to one HVAC system type, one can only combine thermal zones served by the same type of HVAC system. (f) Separate interior and perimeter thermal zones: Assign separate thermal blocks to interior thermal zones located more than 4.6 m (15 ft) from an exterior wall, and to perimeter thermal zones within 4.6 m (15 ft) of the exterior. The Chartered Institution of Building Services Engineers (CIBSE) Applications Manual AM11 [22] also provides guidance on thermal zoning for building energy simulation purposes. The manual states that the primary purpose of thermal zoning is to avoid undue complexity. Therefore, thermal zones should be grouped together into one thermal zone if: (a) they are likely to perform similarly without environmental controls; (b) they have similar heating and cooling equipment and thermostat setpoints; (c) the internal gains from occupants, lighting, and equipment are similar; and (d) the solar gains are similar. In addition, thermal zones should be split into more than one thermal zone if: (a) variations in environmental conditions within the thermal zone are of importance, (b) there is likely to be temperature stratification, and (c) solar or internal gains differ significantly throughout the space, and mixing of the air is limited. Finally, the CIBSE manual states that the boundary between such thermal zones should not be assigned a thermal mass nor impede radiative heat transfer. Also, the geometry should be simplified, but care should be taken to conserve area, volume, and orientation. In summary, the standards and guidelines related to building energy simulation modeling commonly state that the interior space of a simulated building needs to be separated into core and perimeter thermal zones and divided by orientation. In addition, the spaces that have similar internal loads, occupancies, and schedules can be aggregated into one thermal zone in the simulation model. 3.3. Previous studies about thermal zoning for building energy simulation In the past four decades of building energy simulation program development, only a few studies have focused on the im-

pact of thermal zoning strategies during the building energy modeling process. This section examines the concept of thermal zoning as described in these previous studies, as well as the published thermal zoning procedures for building energy simulation. In 1971, during the time when building energy simulation programs were first being developed, the US Post Office sponsored the development of a computer simulation program to analyze the energy consumption of the post office facilities across the US [23]; this is commonly referred to as the Post Office Program. This building energy simulation program is considered the first general purpose public domain program for wholebuilding simulation; it calculated hourly cooling and heating loads and HVAC system and plant energy use, and was capable of performing an hourly economic analysis [23,24]. Lokmanhekim [25] described the basic structure of the program as consisting of four main subprograms, which ran in sequence: (1) load calculation, (2) thermal load plot, (3) systems simulation, and (4) economic analysis. Lokmanhekim also provided a definition for space and thermal zoning for use when performing thermal zoning in the simulation program. In this procedure, it was recommended that the thermal zoning for a simulation utilize one-day interior temperature time-series plots for spaces; these could be compared to determine compatible groupings of spaces into thermal zones. However, this study did not provide a detailed procedure for selecting thermal zones for an hourly building energy simulation. In addition, a simple, qualitative, thermal zoning selection statement was only mentioned once, and was not demonstrated and/or verified through testing via case study simulations. Heidell and Taylor [26] were some of the first to examine how well a DOE-2 simulation was calibrated for a large office building, by matching the actual end-use energy consumption of the building to that which was calculated by the simulation. In their study, measured data were compared with the simulation results, including end-use energy consumption and heating and cooling loads by thermal zone. The results were presented for monthly energy use and peak demand. The authors recommended that the simulation model have the same thermal zones as the actual building’s HVAC thermal zones. The study also showed that the building schedules were important variables in a well-calibrated simulation model.

6

M. Shin and J.S. Haberl / Energy & Buildings 203 (2019) 109429

Fig. 2. Models with different numbers of thermal zones: (a) one-zone, (b) five-zone, and (c) 18-zone. Figures were resketched from Hinchey [28].

However, the authors did not explain how the actual HVAC thermal zones were used to create thermostatic zones in the simulation model. Goldberg [27] evaluated five building energy simulation programs in terms of experimental, long-term, and transient energy usage data for two residential houses. His-two-step validation methodology utilized a constrained optimization approach involving a parametric variation of individual building envelope features. The results illustrated the viability of the parametric variation methodology and the importance of earth-contact heat transfer modeling in heating-dominated climates. In the study, the computation time for an annual simulation period was reduced from 7.25 h for 10 zones to 2.78 h for four zones, a 62% decrease. However, the study did not explain how the actual thermal zones in a building should be created for the simulation model. Hinchey [28] tested how sensitive DOE-2 results were to assumptions about the number of thermal zones in a large commercial office building. In her study, different thermal zoning assumptions were related to one another by comparing the graphical output from the simulation program to the measured data. One-zone, five-zone, and 18-zone building energy models (see Fig. 2) were all developed for the same building. The results showed that a difference of only 3.5% of total energy consumption was found between the different models. However, although a typical traditional thermal zoning approach (i.e., the core and perimeter method) was applied in this study, the researcher did not consider other ways of zoning a building. In addition, only one specific building and HVAC system type was tested, and only one thermostat setting was considered for all zones. Therefore, the results from the study might not apply to other building types with varying thermostat temperatures. In addition, it is unclear if variations in the exterior window areas were considered. Samuels et al. [29] studied problems with current Australian energy-efficient design guides for application to thermal zoning in solar-efficient designs with north-oriented living rooms (i.e., towards the equator, similar to southern exposure). Their survey showed that occupants preferred to have winter sunlight and daylight penetration into their bedrooms as well as their living rooms. However, this study only suggested alternative thermal zoning principles, and did not provide a thermal zoning selection method or numerical procedure. In addition, the object of this study was residential buildings, many of which included passive house designs. Therefore, the results might not apply to other building types. Pan et al. [30] developed a method for calibrating a computer simulation on the basis of guidelines published in previous literature. Their model calibration was conducted by comparing simulation output against measured energy use. In their study, the simulation model was divided into one internal zone and four perimeter zones facing north, south, east, and west, with a depth of 4.2 m (13.8 ft) from the external wall. However, only a typical and traditional thermal zoning approach (i.e., core and perimeter method) was applied in this study; the authors did not consider other ways of zoning a building.

Musau and Steemers [31] investigated energy use in laboratory buildings, showing that interior space planning and/or the ways the space is used may have an effect. The results showed the percentage variations of peak winter loads when interior physical definition, activity organization, and orientation were within a range of 40%. This was true for everything except for the effect of open versus closed floor plans, which resulted in a variation of 73% in the total peak winter load. The summer load variations were within 50% of one another among open, mixed, and closed layouts, and 84% among different closed-plan layouts. In addition, the study showed that the most significant factor regarding energy use was whether the laboratory plan was open or closed. However, the research considered only three common interior laboratory space layouts. In addition, the base-case model did not include variations in exterior window effects such as window size, number of windows, window type, etc. Finally, the study did not provide a detailed method or procedure for how to zone spaces in the simulation. In a separate study, Musau and Steemers [32] investigated the impacts on energy use of the different ways in which office spaces can be organized and used. The analysis indicated that the variations in combined thermal and lighting loads were 19% and 51% of the base-case loads during the peak winter and summer periods in the UK, respectively. The study demonstrated that space planning and utilization can have significant impacts on energy use and are important in assessing energy performance. However, this work considered only five typical office space interior layouts. In addition, the base-case model did not include an evaluation of exterior window size, placement, properties, etc. Finally, it did not provide any details about a method or procedure for zoning spaces in the simulation. Tian and Love [33] analyzed a multi-floor radiant slab cooling system in an institutional building using a calibrated simulation model with measured building energy use and meteorological data. This study found that core zones had smaller cooling load fluctuations and peak cooling loads per unit floor area than did perimeter zones. However, the study used only a traditional thermal zoning approach (i.e., core and perimeter method). In addition, the research did not utilize a parametric analysis, only a one-time simulation. Finally, the results did not show how changes in zoning might have impacted the annual energy use in the simulation. Smith et al. [34] presented a method for automatically generating a building energy model from an architect’s basic massing model. In their study, a usability test was conducted with architects and students, revealing that most users could obtain a simulation of a simple model in less than two hours. Fig. 3 shows the output of the automatic thermal zoning model from the basic building form without floors. However, only the typical and traditional thermal zoning approach (i.e., core and perimeter method) was applied in this study; the authors did not consider other ways of zoning the building. In addition, the research did not provide a method or procedure for how to zone spaces in the simulation. Finally, the results did not demonstrate the impact of different zoning strategies on annual energy use in the simulation.

M. Shin and J.S. Haberl / Energy & Buildings 203 (2019) 109429

Fig. 3. Automatic thermal zoning in the simulation model: (a) basic building form without floors, and (b) building energy model with core offset and perimeter zones. Resketched from Smith et al. [34].

Fig. 4. Comparison of thermal zoning methods: (a) traditional core and perimeter zoning strategy, and (b) zone-typing method. Resketched from Raftery [35].

Raftery [35] developed a new thermal zoning method (i.e., zone-typing method) that defined the various types of thermal zones in a model, based on four major criteria: (1) the function of the space, (2) position of the zone relative to the exterior, (3) available measured data, and (4) method used to condition the zone. Fig. 4 shows a comparison of the traditional core and perimeter thermal zoning method and the new method applied to the same floor plan. Raftery’s zone-typing method yielded a more detailed thermal zoning plan than did the traditional core and perimeter thermal zoning process. The method increased run time from 0.7 h to 3.6 h, an increase of 370%. However, the zone-typing method did not have a numerical selection procedure that would lend itself to automation. In addition, this procedure relied on a user’s subjec-

7

tive use of a simulation program, and the thermal zoning method was not verified through any test or case study. Finally, the results did not show the impacts of different zoning strategies applied to the same building on the annual energy use in the simulation. O’Brien et al. [36] conducted sensitivity analyses that: (1) quantified the impact of thermal zoning and inter-zonal airflow on building performance, (2) optimized a south-facing glazing area, and (3) optimized thermal comfort for passive solar houses. This study showed the relationship between thermal zoning and interzonal airflow rate. The results demonstrated that passive solar buildings, in particular, can benefit from increased air circulation with a forced air system because such a system allows solar gains to be redistributed throughout the building, thus reducing zone overheating from direct gain and total energy consumption. In addition, the study showed that with increased air circulation, heating and cooling energy was reduced by a total of 16%, while the magnitude of overheating was reduced by 55%. However, this study tested only one specific building type (i.e., a passive solar house) and was based on actual thermostat zones. In addition, the thermal zoning method used did not have a numerical procedure that would lend itself to automation. Finally, this research did not verify the performance (i.e., thermal comfort) of the variations that were studied, and did not provide a method or procedure for how to zone spaces in the simulation. Smith [37] tested various thermal zoning configurations, based on Appendix G of ASHRAE Standard 90.1-2007; the goal was to determine the impact on building energy use. The results showed that one- and two-zone models underestimated the energy use as compared to a five-zone model. This work also suggested that a better default for the perimeter offset in a core and perimeter zoning configuration would be closer to 4.9 m (16 ft) rather than 3.0 m (10 ft) and 6.1 m (20 ft) perimeter offsets. This study demonstrated the importance of using reasonable zoning assumptions in conceptual models, but did not provide a generalized method or procedure for how to zone a building in the simulation. Georgescu et al. [38] analyzed a detailed building energy model using an optimization method called the Koopman operator, which is an infinite-dimensional linear operator that captures non-linear, finite-dimensional dynamics without linearization, in order to identify and develop zoning approximations based on observations of zone temperature. The purpose of Georgescu’s approximation was to reduce the complexity of the model while minimally impacting model accuracy. In this study, the original detailed model contained 191 zones. The number of zones was reduced to 32 using the developed method, with only a 3.3% error in annual heating and cooling load prediction. The research also included guidelines for maintaining model accuracy, including: (1) when merging zones, the thermal mass of unmodeled walls should be captured; (2) zones containing exterior surfaces should not be merged with zones that do not contain exterior surfaces; (3) perimeter zones that are merged should have similar surface orientations and window areas; and (4) zones with small volumes and surface areas can be merged with much larger adjacent zones with little loss of accuracy. However, a detailed model of all of the zones must first be created to establish a baseline model for comparison. Unfortunately, creating a detailed model of all zones takes a significant amount of time, and often implies uncertainty (i.e., when the number of parameters in the model increases). Moreover, this study did not provide a generalized, step-by-step procedure for how to establish the thermal zones in the building simulation model. Bleil De Souza and Alsaadani [39] examined how recommended settings for internal gains and ventilation rates together with the use of different zoning strategies, can produce significant variations in the predicted energy demands of office buildings. Their results showed that the thermal behavior of each zone was mainly influ-

8

M. Shin and J.S. Haberl / Energy & Buildings 203 (2019) 109429

enced by the relationships among different combinations of floor area, window area, and internal gain. In addition, they illustrated a method for zoning a building to predict ranges of heating and cooling demands, working with extremes in terms of window-tofloor area ratio and internal gain settings. However, they concluded that more simulations and tests were necessary to establish a set of criteria regarding how best to set up zoning, considering various combinations of floor area, window area, and internal gain. Finally, since only one climate and HVAC system were applied in this study, the authors suggested repeating the experiment for different systems in different climates. Jones et al. [40] described a series of five automated steps for translating geometric data from an un-zoned CAD model into a multi-zone building energy model. The study showed that if a full simulation is run, the building could be zoned by analyzing interior temperature profiles (i.e., Koopman operator). However, the research fell short of developing a generalized method of thermal zoning. In addition, only the common traditional thermal zoning approach (i.e., core and perimeter method) was applied to the case-study building. Dogan et al. [41] presented an algorithm to produce automated, multi-zone building energy models for urban and schematic designs. Their algorithm used a robust straight skeleton algorithm [42] with an arbitrary building massing, subdivided into core and perimeter thermal zones. A straight skeleton algorithm is a method of representing a polygon via a topological skeleton similar to the medial axis,1 but it differs in that the skeleton is composed of straight-line segments, while the medial axis of a polygon may involve parabolic curves. This can be computed by simulating the shrinking process by which it is defined. Their proposed algorithm was tested with various floor plans of varying levels of complexity. However, this research only demonstrated how to subdivide the floor volumes into thermal zones based on the traditional core and perimeter method. This study fell short of developing a procedure that automatically converted massing models into building energy models. Dogan et al. [43] introduced a general algorithm for automatically converting arbitrary building massing models into multi-zone, multi-floor building energy models. In this new method, the different layouts changed the annual heating and cooling loads by up to 21%. The test results showed that when using this new algorithm for small models, the geometry was computed within milliseconds. This was only slightly longer for larger models. For example, 184 zones required 15.5 s to compute. The EnergyPlus simulation time ranged from 20 s to five minutes for the largest model. This study basically followed the zoning method provided by Appendix G of ASHRAE Standard 90.1-2013, which is the core and perimeter method. However, this research did not consider a method for grouping spaces into a common zone beyond the grouping provided by the core and perimeter zoning process. Yi [44] developed an interface for suggesting optimized thermal zone layouts in order to facilitate a thermal zoning-based space arrangement. To accomplish this, four major performance criteria were adopted for evaluation, including: Energy Use Intensity (EUI), Predicted Mean Vote (PMV), daylight level, and room shading. This program allowed for the regrouping of thermal zones according to spatial functions. In this method, it was necessary to take local external conditions into account within the simulation, as well as consider extra subdivisions for the perimeter space(s). The results showed that indoor thermal conditions and the occupancy schedule had significant impacts on the final layout. However, the thermal zoning method used in this study was entirely based on spa1 The medial axis of an object is the set of all points having more than one closest point on the object’s boundary. Originally referred to as the topological skeleton, it was introduced by Blum as a tool for biological shape recognition.

tial function. Therefore, the research did not consider any thermal parameters for grouping spaces into zones. This work also relied on a user’s subjective judgement to select zones. Furthermore, the proposed method did not consider HVAC system type for thermal zoning. Georgescu and Mezic´ [45] introduced a systematic approach for creating zoning approximations in institutional buildings. In a fashion similar to that of their previous study [38] which utilized the Koopman operator, the time-series output produced by the building simulation was decomposed into spatial modes that captured the thermal behavior of a building at different time-scales. The study also provided guidelines for helping to maintain model accuracy: (1) one space use classification should be the same throughout the thermal zone, (2) all rooms in a thermal zone that are adjacent to glazed exterior walls should face the same orientation or their orientations should vary by less than 45°, and (3) separate zones should be assumed for interior and perimeter rooms. However, in this study, a detailed building model first needed to be created so that a simplified zone model could be produced. Heo et al. [46] investigated the impact of model simplification on thermal zoning, and internal load schedules for building energy simulations on the accuracy of model outputs. In order to examine the influence of simplification on thermal zoning, DesignBuilder and EnergyPlus software were used to apply four different thermal zoning layouts to a model of a two-story residential house. The baseline was the most detailed simulation model, where each space was assigned a thermal zone. The simulation outputs (i.e., annual electricity and heating demand) from the baseline model were compared with the outputs from models with three different thermal zoning layouts: (1) two adjacent spaces combined into a thermal zone, (2) all of the spaces on the same floor combined into a thermal zone, and (3) the whole building modeled as a thermal zone. The results showed that in general, when the number of thermal zones was decreased, the annual heating demand was underestimated. For example, the annual heating demand of the single-zone model for the entire building decreased by 24% compared to the outputs from the baseline model. Conversely, the annual equipment electricity consumption was overestimated by 21% in the models with small numbers of thermal zones. This study underscored the significance of thermal zoning strategies on model output accuracy and energy savings. Dogan and Reinhart [47] proposed an urban building energy modeling algorithm called “Shoeboxer,” which is an automated procedure for simplifying multi-zone urban building energy models using a basic core-perimeter thermal zoning strategy. Their case studies showed that the Showboxer method provided faster and more reliable results compared to core perimeter multi-zone whole-building energy models used for urban building energy modeling. The researchers compared the annual energy use intensities (EUI) of the Showboxer and multi-zone models, showing that the results were within a ± 15% margin of error. Chen and Hong [48] used EnergyPlus software to evaluate the impacts of three geometric (i.e., thermal zoning) modeling methods: (1) OneZone, which is one zone per floor, (2) AutoZone, which is a core-perimeter method, and (3) ProtoType, which is a detailed simulation method, on annual heating and cooling loads, autosized equipment capacities, and the annual energy use of 940 office and retail buildings in three climate conditions. The results showed that the thermal loads and equipment capacities decreased when the OneZone method was used, as compared to when the AutoZone method was employed. This included a 15.2% lower fan capacity, 11.1% lower cooling capacity, 11.0% lower heating capacity, 16.9% less heating load, and 7.5% less cooling load. Source energy use differences ranged from −7.6% to 5.1%. In addition, when the AutoZone method was compared to the ProtoType method, significant differences were found in the source energy use, ranging from

M. Shin and J.S. Haberl / Energy & Buildings 203 (2019) 109429

−12.1% to 19.0%, and even larger ranges of difference were found for thermal loads and equipment capacities. Sarkar [49] proposed a new thermal zoning method that can be used for HVAC system design, especially when determining the supply conditions and sizing of dedicated outdoor air systems (DOAS) with local recirculating units, based on design space cooling loads. A simulation case study was performed using IES-VE software and a prototype office building. The results showed that a DOAS with the new thermal zoning method, as compared to an un-zoned case (i.e., the entire building as a single thermal zone), consistently maintained the design temperature and RH limits for more than 98% of the occupied hours and without increasing the overall HVAC energy and chiller loads beyond 4% and 3%, respectively. Zhu et al. [50] developed a new method entitled Building Blocks Energy Estimation (BBEE) for estimating the energy performance of a group of buildings at the urban district level. The method provides a thermal zoning procedure that begins by dividing the building mass into core and perimeter spaces, based on the thermal zoning method provided by Appendix G of ASHRAE Standard 90.1–2010. Then, the zones are further divided into typical zones, the size of which is defined in this method as 4.5 m × 4.5 m × 4.0 m (width × depth × height). The zones are combined according to their orientation and window-to-wall ratio. Users can look up corresponding energy intensities from a preexisting database, so energy performance modeling and simulation are not required. A case study was conducted using the proposed BBEE method, and the results (i.e., annual heating demand) compared with those obtained from building energy simulation models via DeST software. The comparison showed that the difference in annual heating demand between the two methods was less than 6%. Despite a number of studies have examined the impact of thermal zoning strategies on building energy simulation models, only few studies attempted to develop a new thermal zoning method rather than using a traditional core and perimeter zoning method. Furthermore, these new proposed thermal zoning methods tend to consider only qualitative aspects for thermal zoning for both HVAC design and building energy simulation. However, the results from the most studies indicated that there is a significant variation in the simulation model output and accuracy based on the selection of thermal zoning method for building energy simulations. 4. Discussion This literature review has presented: (1) a definition of building thermal zones, (2) thermal zoning methods used in the actual HVAC design process, (3) thermal zoning methods used in building energy modeling, and (4) a review of previous studies about thermal zoning strategies for building energy simulation. The findings of the literature review are summarized below. (1) The previous literature has used several similar terms to describe building thermal zones, such as thermal zones, thermal blocks, HVAC zones, etc. However, all of these different terms indicate the same idea or concept, which is that a thermal zone is a portion of a building whose HVAC system is controlled by a single sensor (i.e., a thermostat). (2) Several previous studies that covered HVAC design and control were reviewed to investigate thermal zoning procedures and methods used in the HVAC design process. These studies provided a consensus regarding the criteria that should be used to divide thermal zones, including: (a) solar gain, (b) orientation, (c) occupancy, (d) schedule, and (e) space function. The concept of thermal zoning for the HVAC design process can be clearly described using these factors. How-

9

ever, none of the previous literature has provided a set of detailed thermal zoning rules, but rather only simple guidelines and factors for consideration when engineers are designing a building’s thermal zones. Therefore, in most designs, it appears that the thermal zoning task during the HVAC system design process tends to rely on the engineer’s experience and intuition. In this respect, a standardized thermal zoning method for various HVAC types and sizes should be developed and considered as a future research direction. For example, how a thermal zoning strategy would affect the basic specifications of HVAC system design (such as system flow, design supply temperature, and space humidity considerations) should be addressed. (3) A number of building energy modeling standards and guidelines for commercial buildings, such as Appendix G of ASHRAE Standard 90.1-2016, the IBPSA Building Energy Modeling Book (BEMBook), and the CIBSE Application Manual AM11, were reviewed in terms of information about thermal zoning for building energy simulation. These standards and guidelines state that the interior space of a simulated building needs to be separated into core and perimeter thermal zones, and then divided by orientation. In addition, the spaces with similar internal loads, occupancies, and schedules should be aggregated into one zone in the simulation model. Unfortunately, however, none of the previous literature provided a comprehensive, detailed thermal zoning method or comprehensive rules for complete, automated thermal zoning for use in energy simulation modeling. (4) Although there have now been thousands of articles written about building energy simulation modeling that document new approaches to building energy efficiency, only a few have addressed how to thermally zone a building in a building energy simulation program other than through the conventional zoning process (i.e., the core and perimeter thermal zoning method). In addition, none of the previous literature has rigorously tested the core perimeter thermal zoning method to determine if it gives the most accurate thermal zoning results in building energy simulations. In recent years, as urban-scale building energy modeling has received increased attention, several studies have attempted to develop new methods for predicting the overall building energy use across neighborhoods. However, these new methods were developed based on the traditional core and perimeter strategy, and the results were not completely verified in terms of accuracy and reliability. Therefore, further improvement of these new thermal zoning methods is required. 5. Conclusions and recommendations This review gives a detailed overview of the various definitions for thermal zone, thermal zoning strategies, and methods for building energy simulation and HVAC system design. To perform this review, the relevant literature on thermal zoning approaches (e.g., textbooks, standards and guidelines, user manuals for wholebuilding energy simulation software, previous studies associated with new thermal zoning procedures other than the traditional core and perimeter thermal zoning method, etc.) were reviewed. Based on the results of this review, the following conclusions were drawn: (1) Although several different terms (i.e., thermal zone, block, and HVAC zone) have been used to describe zones in HVAC system design and building energy tion modeling, one consistent concept was that of a

thermal thermal simulathermal

10

M. Shin and J.S. Haberl / Energy & Buildings 203 (2019) 109429

zone being a space (or group of spaces) that maintains the same temperature throughout the day, with its own thermostat setpoint and schedule. An additional consistent concept was that of thermal zoning as a process of grouping spaces together that have similar space conditioning requirements. (2) With regards to thermal zoning methods for HVAC system design and building energy simulation, it was observed that most of the literature described the most prevalent thermal zoning method (i.e., the core-perimeter thermal zoning method) and the primary factors affecting building thermal zoning in the HVAC system design process. Therefore, it can be concluded that decisions regarding thermal zoning made for HVAC system design would rely heavily on the engineer’s intuition and experience, since there are no guidelines or standards for selecting thermal zoning methods for the various types of HVAC system. (3) Most of the previous studies on thermal zoning for building energy simulation indicated that the selection of a thermal zoning method significantly impacted the simulation results (i.e., the accuracy of the predictions) and simulation run time. However, it was observed that only a few of the previous studies developed new thermal zoning approaches for building energy simulation programs that went beyond the conventional core and perimeter thermal zoning strategy. In addition, the newly proposed thermal zoning methods were primarily focused on simulating at the urban district level, and merely divided building mass into core and perimeter zones and roughly estimated the energy use for a group of buildings. Based on this comprehensive literature review of thermal zoning strategies, several future tasks are recommended. First, there is a need to develop a well-documented, accurate thermal zoning method that can assist designers in their design of building energy simulations. Such a new thermal zoning method should have an automated feature that creates the building’s thermal zones and provides feedback about anticipated indoor comfort conditions. In addition, future work should also focus on identifying the building features most likely to have the greatest impact on the thermal zoning results produced by the new method. Finally, further validation should be undertaken of any new thermal zoning method, using experimental measurements from carefully selected buildings.

Declaration of Competing Interest We wish to confirm that there are no known conflicts of interest associated with this publication and there has been no significant financial support for this work that could have influenced its outcome. We confirm that the manuscript has been read and approved by all named authors and that there are no other persons who satisfied the criteria for authorship but are not listed. We further confirm that the order of authors listed in the manuscript has been approved by all of us. We confirm that we have given due consideration to the protection of intellectual property associated with this work and that there are no impediments to publication, including the timing of publication, with respect to intellectual property. In so doing we confirm that we have followed the regulations of our institutions concerning intellectual property. We understand that the Corresponding Author (i.e., Minjae Shin: [email protected]) is the sole contact for the Editorial process (including Editorial Manager and direct communications with the office). He is responsible for communicating with the other authors about progress, submissions of revisions and final approval of proofs. We confirm that we have provided a current, correct email address which is accessible by the Corresponding Author and which has been configured to accept email from [email protected]. Acknowledgement This study was funded by the American Society of Heating, Refrigerating and Air-conditioning Engineers (ASHRAE) Graduate Student Grant-In-Aid, which is a fund allocated to the Research Program for the Society’s Fiscal Year 2014–2015. We thank two anonymous reviewers for valuable comments and suggestions. Supplementary materials Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.enbuild.2019.109429. Appendix. Previous studies about thermal zoning for building energy simulation

Author

1971

Lokmanhekim, Description of the M. program and details of the load program Heidell, J.A. Comparison of Taylor, Z.T. empirically measured end-use metered data with DOE 2.1 simulations Goldberg, L.F. A comparative validation of the long term energy consumption predictions of five residential building energy simulation programs in a heating climate Hinchey, S.B. Influence of thermal zone assumptions on DOE-2 energy use estimations of a commercial buildings Thermal zoning in Samuels, R. solar efficient design: Ballinger, J.A. User experiences and Coldicutt, S. Williamson, T.J. designer preconceptions Pan, Y. Huang, Calibrated building Z. Wu, G. energy simulation and its application in a high-rise commercial building in Shanghai Musau, F. Space planning and Steemers, K. energy efficiency in laboratory buildings: The role of spatial, activity and temporal diversity Musau, F. Space planning and Steemers, K. energy efficiency in office buildings: The role of spatial and temporal diversity

1985

1985

1991

1993

2007

2007

2008

Title

Design or simulation

Building type

Simulation tool

Method included exterior windows

Zoning method

When applied during design process

What did the literature do?

Simulation

N/A

N/A

N/A

N/A

N/A

Simulation

Office

DOE-2

N/A

Core and perimeter

After interior spaces are assigned.

Simulation

Residential

3D Scribe, EEDO, HOTCAN, SERIRES, CALPAS3

N/A

N/A

After interior spaces are assigned.

Evaluated five building energy simulation programs in terms of experimental, long-term, and transient energy usage data for two residential houses.

Simulation

Office

DOE-2

N/A

Core and perimeter

After interior spaces are assigned.

Tested how sensitive DOE-2 results were to assumptions about the number of thermal zones in a large commercial office building.

Design

Residential

N/A

Yes

Zoned by orientation

N/A

Studied problems with current Australian energy-efficient design guides for application to thermal zoning in solar-efficient designs with north-oriented living rooms.

Simulation

Office

DOE-2

N/A

Core and perimeter

After interior spaces are assigned.

Developed a method for calibrating a computer simulation on the basis of guidelines published in previous literature.

Design

Laboratory

TAS

No

N/A

N/A

Investigated energy use in laboratory buildings, showing that interior space planning and/or the ways the space is used may have an effect.

Design

Office

TAS

No

N/A

N/A

Investigated the impacts on energy use of the different ways in which office spaces can be organized and used.

Provided a definition for space and thermal zoning for use when performing thermal zoning in the simulation program. Examined how well a DOE-2 simulation was calibrated for a large office building, by matching the actual end-use energy consumption of the building to that which was calculated by the simulation.

M. Shin and J.S. Haberl / Energy & Buildings 203 (2019) 109429

Year

(continued on next page)

11

12

Year

Author

2009

Tian, Z. Love, J.A.

2011

2011

2012

2012

2012

2013

2014

Energy performance optimization of radiant slab cooling using building simulation and field measurements Smith, L. Automated energy Bernhardt, K. model creation for Jezyk, M. conceptual design Raftery, P. Calibrated whole building energy simulation: An evidence-based methodology O’Brien, W. Thermal zoning and Athienitis, A. interzonal airflow in Kesik, T the design and simulation of solar houses: a sensitivity analysis Smith, L. Beyond the shoebox: Thermal zoning approaches for complex building shapes Georgescu, M. Creating zoning Eisenhower, B. approximations to Mezic, I. building energy models using the Koopman Operator De Souza, C.B. Thermal zoning in Alsaadani, S. speculative office buildings: discussing the connections between space layout and inside temperature control Jones, N.L. Automated translation McCrone, C.J. and thermal zoning of Walter, B.J. digital building models Pratt, K.B. for energy analysis Greenberg, D.P. Automated multi-zone Dogan, T. Reinhart, C. building energy model Michalatos, P. generation for schematic design and urban massing studies

Design or simulation

Building type

Simulation tool

Method included exterior windows

Zoning method

When applied during design process

What did the literature do?

Design

Office

EnergyPlus

Yes

Core and perimeter

After interior spaces are assigned.

Analyzed a multi-floor radiant slab cooling system in an institutional building using a calibrated simulation model with measured building energy use and meteorological data.

Simulation

Office

Green Building Yes Studio

Core and perimeter

Before interior spaces are assigned.

Simulation

Office and EnergyPlus manufacturing facility

Yes

Zone-typing

After interior spaces are assigned.

Presented a method for automatically generating a building energy model from an architect’s basic massing model. Developed a new thermal zoning method (i.e., zone-typing method) that defined the various types of thermal zones in a model, based on four major criteria.

Simulation

Residential

EnergyPlus

Yes

Direct/indirect N/A solar gain zone

Conducted sensitivity analyses that: 1) quantified the impact of thermal zoning and inter-zonal airflow on building performance, 2) optimized a south-facing glazing area, and 3) optimized thermal comfort for passive solar houses.

Simulation

Office and healthcare facility

Green Building Yes Studio, eQuest

Core and perimeter

Before interior spaces are assigned.

Tested various thermal zoning configurations, based on Appendix G of ASHRAE Standard 90.1–2007; the goal was to determine the impact on building energy use.

Simulation

Office and research facility

EnergyPlus

N/A

Koopman operator

After interior spaces are assigned.

Analyzed a detailed building energy model using an optimization method called the Koopman operator, in order to identify and develop zoning approximations based on observations of zone temperature.

Simulation

Office

EnergyPlus

Yes

Single zone, five-zone, and office in use

After interior spaces are assigned.

Examined how recommended settings for internal gains and ventilation rates together with the use of different zoning strategies, can produce significant variations in the predicted energy demands of office buildings.

Simulation

N/A

N/A

No

N/A

Before interior spaces are assigned.

Described a series of five automated steps for translating geometric data from an un-zoned CAD model into a multi-zone building energy model.

Simulation

N/A

EnergyPlus, Rhino, Archisim

Yes

Core and perimeter

Before interior spaces are assigned.

Presented an algorithm to produce automated, multi-zone building energy models for urban and schematic designs.

(continued on next page)

M. Shin and J.S. Haberl / Energy & Buildings 203 (2019) 109429

2011

Title

Year

Author

2015

Dogan, T. Reinhart, C. Michalatos, P.

2015

2015

2017

2018

2018

2019

Autozoner: an algorithm for automatic thermal zoning of buildings with unknown interior space definitions Yi, H. User-driven automation for optimal thermal-zone layout during space programming phases Georgescu, M. Building energy Mezic, I. modeling: A systematic approach to zoning and model reduction using Koopman mode analysis Heo, Y. Ren, G. Investigating an Sunikka-Blank, adequate level of M. modeling for energy analysis of domestic buildings

Dogan, T. Reinhart, C.

Shoeboxer: An algorithm for abstracted rapid multi-zone urban building energy model generation and simulation Chen, Y. Hong, Impacts of building T. geometry modeling methods on the simulation results of urban building energy models Sarkar, M. Thermal zoning based on design cooling loads: methodology and simulation case study for a DOAS with local recirculating units Chen, Y. Hong, Building Blocks Energy T. Piette, M.A. Estimation (BBEE): A method for building energy estimation on district level

Design or simulation

Building type

Simulation tool

Method included exterior windows

Zoning method

When applied during design process

Simulation

Office

EnergyPlus, Rhino, Archisim

Yes

Core and perimeter

Before interior spaces are assigned.

Introduced a general algorithm for automatically converting arbitrary building massing models into multi-zone, multi-floor building energy models.

Simulation

Office

Ecotect

Yes

Simulated Annealing, Genetic Algorithm

After interior spaces are assigned.

Developed an interface for suggesting optimized thermal zone layouts in order to facilitate a thermal zoning-based space arrangement.

Simulation

Office and research facility

EnergyPlus

N/A

Koopman operator

After interior spaces are assigned.

Introduced a systematic approach for creating zoning approximations in institutional buildings.

Simulation

Residential

DesignBuilder, EnergyPlus

No

After interior spaces are assigned.

Investigated the effect of reducing the number of thermal zones in modeling a domestic building on prediction accuracy.

Simulation

Hotel, service, EnergyPlus commercial, residential, office and core/circulation.

Yes

A zone for a space, a zone for two combined spaces, a zone for a floor, a zone for entire building Core and perimeter

Before interior spaces are assigned.

Presented an algorithm that abstracts an arbitrarily shaped set of building volumes into a group of simplified ‘shoebox’ building energy models.

Simulation

EnergyPlus Office and retail buildings

No

Pixel-based autozoning algorithm

After interior spaces are assigned.

Evaluated the impacts of three thermal zoning methods (i.e., OneZone, AutoZone, Prototype) on the simulated building performance in the urban context.

Simulation

Office

No

Based on design space cooling load

After interior spaces are assigned.

Proposed a method to link thermal zoning and HVAC system design in buildings based on spatial cooling loads and exploring its implications on resulting zone.

Simulation

Office and N/A retail buildings

No

Building Blocks After interior spaces are assigned. Energy Estimation Method

IES-VE

What did the literature do?

M. Shin and J.S. Haberl / Energy & Buildings 203 (2019) 109429

2016

Title

Developed a new method for estimating energy performance of a group of office buildings located in the same district.

13

14

M. Shin and J.S. Haberl / Energy & Buildings 203 (2019) 109429

References [1] U.S. EIA., Annual Energy Outlook 2018, DOE, Washington, D.C.: U.S., 2018. [2] T. Abergel, B. Dean, J. Dulac, Towards a zero-emission, efficient, and resilient buildings and construction sector: Global Status Report 2017, United Nations, 2017. [3] L. Pérez-Lombard, J. Ortiz, C. Pout, A review on buildings energy consumption information, Energy Build. 40 (2008) 394–398. [4] V. Bazjanac, Model based cost and energy performance estimation during schematic design, in: 22nd Conference on Information Technology in Construction, Dresden, Germany, 2005, pp. 677–688. [5] B. Bordass, A. Leaman, P. Ruyssevelt, Assessing building performance in use 5: conclusions and implications, Build. Res. Inf. 29 (2001) 144–157. [6] T. Maile, M. Fischer, V. Bazjanac, Building Energy Performance Simulation Tools - a life-Cycle and Interoperable Perspective, Stanford University, Stanford, CA, 2007. [7] Autodesk, Revit, Autodesk, Inc., San Francisco, CA, 2016. [8] ASHRAE, ANSI/ASHRAE/IES 90.1-2016 Energy Standard For Buildings Except Low-Rise Residential Buildings, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Atlanta, GA, 2016. [9] R. Hetherington, R. Laney, S. Peake, Zone modeling and visualisation: keys to the design of low carbon buildings, in: 2012 16th International Conference on Information Visualisation, 2012, pp. 495–503. [10] M.J. Clayton, J. Haberl, W. Yan, ASHRAE RP-1468, Automating building Thermal Model creation: Interoperability from BIM to Energy Simulation Software, Texas A&M University, College Station, TX, 2012. [11] S. Kota, F.J.F. Stipo, J. WoonSeong, K. Jong Bum, J.L.B. Alcocer, M.J. Clayton, et al., Development of a reference building information model for thermal model compliance testing–part I: guidelines for generating thermal model input files, ASHRAE Trans. 122 (2016) 256–266. [12] F.J.F. Stipo, S. Kota, J. WoonSeong, K. Jong Bum, J.L.B. Alcocer, J.S. Haberl, et al., Development of a reference building information model for thermal model compliance testing–part II: test cases and analysis, ASHRAE Trans. (2019). [13] COMNET, Commercial Buildings Energy Modeling Guidelines and Procedures, Commercial Energy Services Network, 2010. [14] R. McDowall, Fundamentals of HVAC Systems, Academic Press, 2006. [15] W.T. Grondzik, A.G. Kwok, Mechanical and Electrical Equipment For Buildings, 12th ed., John Wiley & Sons, Inc., Hoboken, NJ, 2014. [16] C.M. Gay, C.D.V. Fawcett, Mechanical and Electrical Equipment For Buildings, 1st ed, John Wiley & Sons, Inc., 1935. [17] H.E. Bovay, Handbook of Mechanical and Electrical Systems For Buildings, McGraw-Hill, Inc., 1981. [18] J.F. Kreider, Handbook of Heating, Ventilation, and Air Conditioning, CRC Press, 2001. [19] L.R. Bachman, Integrated Buildings: The Systems Basis of Architecture, John Wiley & Sons, Inc., Hoboken, NJ, 2003. [20] G. Hamrick, Engineer’s HVAC Handbook: A Comprehensive Guide to HVAC Fundamentals, 2nd ed, Price Industries Limited, 2012. [21] IBPSA, Thermal Zoning Determination, IBPSA-USA, 2012. [22] CIBSE, Applications Manual AM11: Building Energy and Environmental Modeling, The Chartered Institution of Building Services Engineers, London, UK, 1998. [23] USPS, U.S. Post Office Energy Analysis Program, United States Postal Service, Washington, D.C., 1971. [24] J.S. Haberl, S. Cho, Literature Review of Uncertainty of Analysis Methods (DOE-2 Program), Energy Systems Laboratory, Texas Engineering Experiment Station, College Station, TX, 2004. [25] M. Lokmanhekim, Description of the program and details of the load program, in: USPS symposium: Computer program for analysis of energy utilization, Washington, D.C., 1971, pp. 29–79. [26] J.A. Heidell, Z.T. Taylor, Comparison of empirically measured end-use metered data with doe 2.1 simulations, in: 1st International Building Simulation Conference, Seattle, WA, 1985, pp. 290–295. [27] L.F. Goldberg, A comparative validation of the long term energy consumption predictions of five residential building energy simulation programs in a heating climate, in: Building Energy Simulation Conference, Seattle, WA, 1985, pp. 282–289.

[28] S.B. Hinchey, Influence of Thermal Zone Assumptions On DOE-2 Energy Use Estimations of a Commercial Building, Texas A&M University, College Station, TX, 1991. [29] R. Samuels, J.A. Ballinger, S. Coldicutt, T.J. Williamson, Thermal zoning in solar efficient design: user experiences and designer preconceptions, Archit. Sci. Rev. 36 (1993) 151–156. [30] Y. Pan, Z. Huang, G. Wu, Calibrated building energy simulation and its application in a high-rise commercial building in Shanghai, Energy Build. 39 (, 6 ) (2007) 651–657. [31] F. Musau, K. Steemers, Space planning and energy efficiency in laboratory buildings: the role of spatial, activity and temporal diversity, Archit. Sci. Rev. 50 (2007) 281–292. [32] F. Musau, K. Steemers, Space planning and energy efficiency in office buildings: the role of spatial and temporal diversity, Archit. Sci. Rev. 51 (2008) 133–145. [33] Z. Tian, J.A. Love, Energy performance optimization of radiant slab cooling using building simulation and field measurements, Energy Build. 41 (2009) 320–330. [34] L. Smith, K. Bernhardt, M. Jezyk, Automated energy model creation for conceptual design, in: 2011 Symposium on Simulation for Architecture and Urban Design: Society for Computer Simulation International, 2011, pp. 13–20. [35] P. Raftery, Calibrated Whole Building Energy simulation: An evidence-Based Methodology, National University of Ireland, 2011. [36] W. O’Brien, A. Athienitis, T. Kesik, Thermal zoning and interzonal airflow in the design and simulation of solar houses: a sensitivity analysis, J. Build. Perform. Simul. 4 (2011) 239–256. [37] L. Smith, Beyond the shoebox: thermal zoning approaches for complex building shapes, ASHRAE Trans. 118 (2012) 141–148. [38] M. Georgescu, B. Eisenhower, I. Mezi, Creating zoning approximations to building energy model using the Koopman operator, in: IBPSA-USA SimBuild, 2012, pp. 40–47. 2012-08-03 ed. [39] C. Bleil De Souza, S. Alsaadani, Thermal zoning in speculative office buildings: discussing the connections between space layout and inside temperature control, First Building Simulation and Optimization Conference, 2012. [40] N.L. Jones, C.J. McCrone, B.J. Walter, K.B. Pratt, D.P. Greenberg, Automated translation and thermal zoning of digital building models for energy analysis, 13th Conference of International Building Performance Simulation Associations, 2013. [41] T. Dogan, C. Reinhart, P. Michalatos, Automated multi-zone building energy model generation for schematic design and urban massing studies, IBPSA eSim Conference, 2014. [42] P. Felkel, S. Obdrzalek, Straight skeleton implementation, in: Spring Conference on Computer Graphics, Budmerice, Slovakia, 1998, pp. 210–218. [43] T. Dogan, C. Reinhart, P. Michalatos, Autozoner: an algorithm for automatic thermal zoning of buildings with unknown interior space definitions, J. Build. Perform. Simul. 9 (2015) 176–189. [44] H. Yi, User-driven automation for optimal thermal-zone layout during space programming phases, Archit. Sci. Rev. 59 (2015) 279–306. [45] M. Georgescu, I. Mezic´ , Building energy modeling: a systematic approach to zoning and model reduction using Koopman mode analysis, Energy Build. 86 (2015) 794–802. [46] Y. Heo, G. Ren, M. Sunikka-Blank, Investigating an adequate level of modeling for energy analysis of domestic buildings, 3rd Asia Conference of International Building Performance Simulation Association, 2016. [47] T. Dogan, C. Reinhart, Shoeboxer: an algorithm for abstracted rapid multi-zone urban building energy model generation and simulation, Energy Build. 140 (2017) 140–153. [48] Y. Chen, T. Hong, Impacts of building geometry modeling methods on the simulation results of urban building energy models, Appl. Energy 215 (2018) 717–735. [49] M. Sarkar, Thermal zoning based on design cooling loads: methodology and simulation case study for a DOAS with local recirculating units, Int. J. Air-Cond. Refrig. 26 (2018). [50] P. Zhu, D. Yan, H. Sun, J. An, Y. Huang, Building blocks energy estimation (BBEE): a method for building energy estimation on district level, Energy Build. 185 (2019) 137–147.