1、 Zhaozhou Meng is a PhD candidate, Jianshun Zhang is a professor and director of Building Energy and Environmental Systems Laboratory, Department of Mechanical and Aerospace Engineering, Syracuse University, Syracuse, NY A Simplified and Scalable Heat-Flow Based Approach for Optimizing the Form, Mas
2、sing and Orientation for High Performance Building Design Zhaozhou Meng, PE Jianshun Zhang, PhD Student Member ASHRAE Fellow ASHRAE ABSTRACT This paper introduces a simplified and scalable heat-flow based approach for form, massing and orientation early stage design of high-performance buildings (HP
3、Bs). A reference building model (RBM) is first defined with pre-selected building materials and HVAC systems for the intended climate and site conditions. The energy performance of this RBM is estimated by whole building energy simulation. Heat fluxes from the enclosure are extracted from RBM simula
4、tion. A simplified and physics-based correlation model was developed to predict how these fluxes would be affected by the shape of the building geometry, window-wall ratio (WWR), and orientation of a proposed building design, which can significantly differ from the RBM. Based on building space heat
5、balance, the predicted heat fluxes were then used to predict the energy consumption of the proposed building. INTRODUCTION Buildings continue to become more complex and must meet stringent energy, indoor environmental quality (IEQ), and other project requirements which often have competing goals. HP
6、Bs design calls for integration, especially for early stage design that has fundamental impact on building performance. Decisions made during early design stage can significantly affect and limit later design choices (Zhaozhou Meng, 2014). For example: building form and massing design affect enclosu
7、re and environmental systems design. Although detailed whole building energy simulation can be used to inform designers to achieve better performance, it is generally too time consuming for early design stage in which fast feedbacks on design choices are needed while insufficient design details are
8、available for such simulation. Other methods using statistics or artificial intelligence techniques have been developed (Betul Bektas Ekici, 2011, Athanasios Tsanas, 2012). However, their applications are limited to design parameters that were selected to build the model. The objective of this study
9、 was to develop a simplified and scalable heat-flow based approach to support the early stage HPBs design integration and optimization. METHODS In this study, the whole building was categorized into multi-design factors (site and climate, form and massing, internal configuration, external enclosure,
10、 environmental systems, energy systems, water systems, material use and embodied energy, and system interdependencies) (Michael Pelken, 2013). While considering all design factors, it focused on the integration of important form and massing design (include: orientation, aspect ratio, window to wall
11、ratio and placements on different facades) for given enclosure (wall, window types) and environmental systems design. External enclosure of a building separates the outdoor environment from indoor spaces. It regulates the heat flows passing through it, for example: conducted heat flow through opaque
12、 walls and solar radiation through windows. In order to integrate form and massing design and provide fast performance feedback, it is very important to quickly quantify heat flows through building enclosure. This method originated from fundamental heat balance principles, as shown in Figure 1(a) an
13、d Equation (1). In order to maintain the indoor air temperature at setting point (left part of equation), the zone air energy loss/gain through building enclosure (Qloss/gain) and internal loads (Qint) via multiple heat transfer mechanisms (including: radiation, conduction, and convection) should be
14、 balanced by HVAC systems (QHVAC) which directly determines the energy consumption needed for the space conditioning. E n v el o p e C o n c o c t i o nS u r f a c e C o n v e c t i o nIn t er n a l Hea t G a i nS u r f a c e R a d i a t io nHV A C S y s t ema) b) Figure 1 (a) Building (Zone) energy
15、 balance and (b) hierarchical energy so is their calculations. Instead of directly calculating heat flows, this method predicts building energy performance using heat flow predicted from correlations against RBM. The hierarchical heat flow and energy prediction overview is shown in Figure 1 (b): hea
16、t fluxes through all enclosure components (roof, facades, and ground floor) were extracted from RBM and aggregated from zone up to whole building level. Then the total energy required to balance the gain/loss was obtained by correlations between the relative change of energy consumption from the ref
17、erence building and the change in design parameters. EnergyPlus was used to perform whole building simulation in this study to obtain the data for the correlation development (U.S. Department of Energy, 2015). For projects located at various locations, corresponding TMY3 (Typical Meteorological Year
18、 3) weather data file is used to provide hourly values of solar radiation and meteorological elements for a 1-year period (S. Wilcox, 2008). Reference Building Model The RBM is a simplified version of proposed design which is easy to build, yet can represent the early stage performance of proposed d
19、esign. It is defined as a single zone building with rectangular footprint, 0 orientation, and 1.5 aspect ratio as shown in Figure 2. Fenestrations are evenly distributed on all facades with equal WWR at 33%. It also shared the same total floor area, enclosure materials and assemblies, and HVAC syste
20、ms with proposed design. Standards and guidelines like ASHRAE 55, 62.1, 90.1, NREL reference building were used to develop the RBM (Chen, 2013). Zone settings of RBM should be the same as proposed design. Figure 2 A RBM building example RBM will be automatically generated in software (Chen, 2013). R
21、eference model accommodates the influence of climate conditions on the dynamical enclosure heat transfer. It is pre-simulated which can establish the corresponding basis of heat fluxes (convective heat on interior surface) of all interior surfaces and energy consumptions for a period of time that is
22、 user defined. Heat Flow Prediction Heat flow passing through building enclosure is very complex. It depends on both inside and outside space conditions as well as the assembly thermal properties (thickness, conductivity, specific heat etc.). Figure 3 is an example showing various heat transfer proc
23、esses that affect the energy flow through a typical wall assembly and window system (U.S. Department of Energy, 2015). They include: outside temperature, wind speed and direction, and surface condition impacted convections; direct, reflected, and diffused sunlight absorbed on surfaces; longwave radi
24、ation received from the adjacent environment etc. The inside surface involves additional received longwave radiation from internal sources (people, equipment and lightings). Heat flow passing through windows can be even more complex, involving: solar radiation transmitted through windows (beam and d
25、iffused), absorbed by windows themselves and beam covered interior surfaces, reflected and redistributed on inside surfaces etc. W a l l I n d o o r S p a c eO u t d o o r E n v i r o n me n tSh o r t w a v e r a d ia t io n fr o m s o la r a n d in t e r n a l s o u r c e sL o n g w a v e r a d ia
26、t i o n e x c h a n g e s w it h o t h e r s u r fa c e s in z o n eL o n g w a v e r a d ia t i o n fr o m in t e r n a l s o u r c e sC o n v e c t iv e h e a t e x c h a n g e w it h z o n e a irC o n d u c t io n fr o m o u t s id eL o n g w a v e r a d ia t io n fr o m t h e e n v ir o n m e n
27、tSh o r t w a v e r a d ia t io n , in c lu d in g d ir e c t , r e fle c t e d , a n d d if f u s e d s u n li g h tC o n v e c t iv e e x c h a n g e w it h o u t d o o r a irW a l l I n d o o r S p a c eO u t d o o r E n v i r o n me n tS o la r R a d ia t io nF r o m W in d o w s I n it ia l T r
28、 a n s m it t e d Di f f u s e S o la r Su r fa c e R e c e iv e d B e a m So la rSu r fa c e R e fl e c t e d B e a m S o la rFigure 3 Heat transfer of wall (left) and window (right) In order to facilitate fast estimation, instead of directly calculating heat flows, the following method was develop
29、ed to predict the heat flows using extracted heat fluxes from presimulated RBM and correlation functions to capture effects of building orientation: a) To extract the heat fluxes of each building enclosure component, the developed method systematically decomposed the whole building enclosure accordi
30、ng to their heat transfer characteristics: i) roof is always facing the sky and fully exposed to solar radiation which is hardly affected by building rotation; ii) faades (walls, windows) are heavily influenced by their directions due to different amount of radiation received and transmitted; iii) g
31、round floor has relatively stable outside boundary condition (underground temperature); but its inside surface can be affected by window size and placement on different facades which introduce solar radiation with varied transmitted intensities. So the developed method classified building enclosure
32、surfaces as: roof, wall/window facing four directions (S, E, N, and W), and ground floor. b) The orientation impact on enclosure heat transfer is captured by heat flux coefficient Cori x for each type of enclosure. It is defined as the heat flux ratio, shown in Equation (2), between rotated and orig
33、inal RBM (0 degree). Rotated RBM has the same orientation as proposed design. (2) Where x is the orientation degree(s), qori 0 is the heat flux of RBM. For example, if a proposed design oriented at 20 degrees (counter-clockwise from North), the same orientation RBM will be simulated and compared wit
34、h original RBM (0 degree). This heat flux coefficient Cori x allows quick calculation of each enclosure surfaces heat flow. Total heat flow of proposed design (Qpre, ori x) is calculated by summation of heat flow of each type of enclosure as defined in Equation (3) below. (3) Where Qpre, ori x is th
35、e predicted total heat flow of proposed design, qori 0 is extracted heat flux of RBM, Cori x is heat flux coefficient, and surface area of proposed design Apro. Because the surface area of each type of enclosure can vary between RBM and proposed design, Apro is used to accommodate the surface area d
36、ifferences and provide the scalability of this heat-flow based approach. Heating Energy Prediction As introduced in Figure 1(a) and Equation (1), HVAC systems supplied energy (QHVAC) is provided to balance the energy loss/gain through building enclosure (Qloss/gain) as well as the internal heat gain
37、s (Qint). Because RBM and proposed design operate at the same climate, internal heat gain conditions and space set point for thermal comfort conditions, as a first of approximation under steady state with negligible internal heat gains, it can be assumed that energy consumption by HVAC system is pro
38、portional to heat loss or gain from the enclosure, shown in Equation (4): (4) Where Epre and Eref are the energy consumptions of the proposed design (with x degree orientation) and RBM, respectively; Cflow is the heat flow ratio that can be calculated from Equation (5). (5) Where Qpre, ori x is the
39、predicted total heat flow of proposed design, and Qref is simulated total heat flow of RBM. The energy consumption of the proposed design Epre can then be estimated using Cflow and simulated Eref .using Equation (4). Case Study An office building was used as a case study to illustrate this developed
40、 method. It is a 5-story building with floor area 55,000 ft2 (5110 m2), located in Syracuse. It is featured with rotated building orientation, shallow plate, and large windows on the south faade in order to implement passive energy saving strategies. At early design stage, in order to utilize the pa
41、ssive strategies by integrating form and massing design with other systems, the architectural team proposed a design with rectangular footprint, 20 degree orientation (counter-clockwise), aspect ratio at 3, and 50% WWR on the south faade and 33% for the rest of the facades. The RBM was first automat
42、ically generated. It shared the same floor area, enclosure assemblies, and HVAC systems with proposed design. To accommodate the 20-degree impact of form rotation, a RBM with 20-degree rotation was simulated. Then the heat flows of enclosure surfaces of both reference models were extracted for 24 ho
43、urs (10-minite resolution), including 10 surfaces in total: roof, walls and windows on four facades (S, E, N, and W), and ground floor. Following Equations (2) to (5), heat flows and heating energy were predicted and corresponding results are shown in Figure 4 and 5, respectively. Figure 4 Predicted
44、 heat flow error of proposed design Figure 5 Predicted heating energy of proposed design (Actual vs Predicted) Heat flow prediction in Figure 4 shows all surfaces are following similar variation trends, error peaks appear before and after noon during daytime and keep very low during the night. Groun
45、d floor gives much higher prediction error than the rest of the enclosure surfaces (with peak value around 103%). Roof and walls are on the second tier that ranges from 25% to 72%. Comparing to all the opaque surfaces, windows give quite low errors. The spikes occurring around 22:30 are caused due t
46、o sudden setting point change-which the heat transfer dynamics cant be well captured by simulation software due to intrinsic drawbacks of the steady state model used. Heating energy prediction is quite well as shown in Figure 5. It captures the trends and overlaps with actual energy for most of the
47、time. However, due to the heat flow prediction errors that occurred before and after noon, it is over predicted about 17% to 20% during this short period of time. Overall, the averaged prediction error of heating energy is -0.2%. Dash line is showing the heating energy prediction with internal heat
48、gain considered; it slightly reduces the over prediction error around noon. DISCUSSION AND RESULT ANALYSIS In order to further improve prediction accuracy, both inside and outside enclosure temperatures were examined in depth because temperature difference is the driving force of heat transfer. Due
49、to the greater error of opaque surfaces than transparent surfaces, roof, walls (S, E, N, and W), ground floor were examined. The outside surface temperatures of walls are shown in Figure 6 (left). It indicated that orientation effect on the outside surface temperatures of all facades were well represented by rotating the RBM to the same angle of proposed design: outside surface temperatures of rotated RBM overlaps with proposed design throughout the day but differs from RBM (0 degree). Outside surface temperatures of roof and ground floor were the
