Modelling building energy processes - Heat and mass transfer by convection

Key concepts

Lecture structure

  • Heat & mass transfer by convection:
    • You can access more information under a variety of topics:

    Course material (initial):

    SIMULATION TECHNIQUES FOR VENTILATION AND AIR FLOW PREDICTION

    Liddament, Martin W. Air Infiltration and Ventilation Centre Science Park, Sovereign Court Sir William Lyons Road, Coventry, Great Britain, CV4 7EZ.

    Summary

    Calculation techniques and numerical models are essential tools for any design process. They provide the means by which the designer can transfer a conceptual plan or idea into the final product. A wide range of techniques of varying complexity is available for the calculation of ventilation, air and pollutant distribution in buildings. The purpose of this paper is to outline modelling concepts in relation to need, applications, model types, availability and data. References are given to sources of code, especially those that are available in the public domain for research purposes.

    Introduction

    Air flow related models are needed for a wide range of applications. These vary from estimating air change rate for basic heat loss analysis to evaluating the air flow performance and cost benefit of alternative ventilation systems. The choice of calculation technique varies according to application, the required level of accuracy, the availability of data and design need with no single method being universally appropriate. It is therefore essential to understand the purpose and range of applicability of each type of model.

    Techniques and Applications

    A summary of primary applications and model selection is presented in the flow chart below.

    In general, methods fall into three main categories, these being:

    Empirical and Simplified Models

    These are used to estimate air change rate. This approach is normally acceptable for the sizing of heating and cooling plant. Approximate estimates are also often acceptable for large scale studies on groups of buildings to gain a general insight into the adequacy of ventilation and overall energy impact. Calculations at this level are often very basic and are frequently based on estimates of building airtightness. A common technique is to infer the average air change rate as a proportion of building airtightness as measured at an artificially induced pressure of 50 Pa (Q50 leakage). Air change is given by: 

    The drawback of this approach is that it is unresponsive to the impact on air change of the climatic driving forces of wind and temperature. This has been overcome by a more theoretically based simplified approach that has been developed for both single zoned [2] and multi-zoned [3] structures. Q50 leakage data is converted to an equivalent leakage area based on a 4 Pa reference pressure. The user must provide information on the distribution of leakage between horizontal and vertical surfaces, surrounding terrain and shielding conditions, wind speed and inside outside temperature difference. The air flow rate due to infiltration is given by: 

    Mechanical extract or supply only ventilation is incorporated by addition in quadrature i.e.: 

    Balanced mechanical ventilation is incorporated by direct addition.

    Zonal Models

    Zonal models may be used to predict: These models take the form of a flow network in which zones or rooms of differing pressure are interconnected by a set of flow paths. This network is approximated by a series of equations which represent the flow characteristics of each opening and the forces driving the air flow process. Zonal models vary in complexity from single zone approximations to multi-zone methods, in which each room is represented by a separate zone. In either case a flow rate is determined through each opening such that the flow into and out of each zone is balanced. If there are `j' flow paths penetrating a zone the flow balance is described by: 

    Flow through the i'th flow path is given by: 

    it follows, therefore, that: 

    The user must supply all `C', `n', and `p' values (i.e. for paths l to j), the remaining unknown in each zone is the internal pressure, pi. Unfortunately this cannot be derived directly and, instead, must be evaluated by numerical `iteration'. Methods and suggested values are described in refer [4]. Depending on the level of detail required, each flow path may represented an individual component, such as a gap or crack around a door or window, or a combination of components such as an entire section of a building. Purpose provided openings, flues and mechanical ventilation systems may also be incorporated into the network by representing their flow characteristics as additional paths. It is essential for all flow openings to be represented. Calculations will rapidly depart from reality if any openings are ignored. Pollutant distribution is determined by specifying the emission characteristics in each zone [7]. The dilution and transfer of pollutant in the air stream is represented mass conservation. A simple single zone model with listing has been published (AIDA[4]). Public domain multi-zone code listings include AIRNET[6] and CONTAM93[7]. Other widely available Codes include BREEZE[8] and COMIS[9].

    Computational Fluid Dynamics (CFD)

    Often information is needed about the pattern of air flow and the distribution of air temperature and pollutants in a space when mixing is not uniform. In the past, design has sometimes been based on the measurement of air flow patterns made in test chambers. More recently, `computational fluid dynamics' have been applied. Specific applications include the simulation and prediction of: These methods approximate the enclosed space by a series of control volumes or elements. Typically, the space may be divided into 30000 or more elements. The system of discretisation can be non uniform, so that clusters of elements can be located at areas of greatest interest. Air flow, turbulence, energy propagation and contaminant spread are represented in each of the control volumes by a series of discretised transport equations. In structure, these equations are identical but each on represents a different physical parameter. The generalised form of the transport equation is given by: 

    Considerable computational effort is normally necessary to solve this series of equations with processing times sometimes taking many hours. Nowadays, software has been adapted to run on the latest generation of PC's and, therefore, CFD techniques have found their way into the design office.

    Some public domain listings are available. A 2-dimensional laminar demonstration flow code is published in FORTRAN by Shih [10]. He also explains how this code may be developed to include three dimensions, turbulence, buoyancy and other flow parameters. A comprehensive 3-dimensional flow code, including buoyancy and turbulence, has also been published by Kurabuchi et al (EXACT3 [11]). Recently, Baker et al [12] have undertaken much development work on CFD techniques in the United States as part of an ASHRAE study. Within the international Energy Agency, several studies have investigated the performance and application of CFD methods [13],[14]. Many commercial codes are available.

    Combined Thermal and Air Flow Modelling

    It is possible to combine simplified, zonal or CFD air flow models with thermal models to predict a complete picture of thermal behaviour in a building. This is described in further detail by Kendrick [15].

    Conclusions and Discussion

    Simplified models have limited applications but are useful if approximate estimates of air change are needed. Zonal models can yield extremely useful information about flow and transport behaviour through a building. However, they do not have the same degree of commercial development as CFD techniques. Possibly, this is because their applicability is limited to building physics alone. Acceptance depends on good user interfaces, good graphical output and an extensive `transparent' database containing default information. Current multi-zone methods are seeking to achieve these objectives. CFD techniques are well established for many engineering applications. While they produce good graphical output, accuracy depends on a real knowledge of the flow mechanisms and flow sources in a building.

    REFERENCES

    [1]Dubrul C
    Inhabitants behaviour with regard to ventilation
    AIVC, Technical Note 23, 1988
    [2]Sherman M H, Grimsrud D T
    Measurement of infiltration using fan pressurisation and weather data
    Proc, AIVC 1st Annual Conference 1980
    [3]Feustel H E, Sherman M H
    A simplified model for predicting air flow in multizone structures
    LBL Applied Science Division report, February 1987
    [4]Orme M, Liddament M W, Wilson A
    An analysis and data summary of the AIVC's numerical database
    Technical Note 44, 1994
    [5]Liddament M W
    AIDA - an air infiltration development algorithm
    Air Infiltration Review, Vol. 11, No 1 December 1989.
    [6]Walton G N
    AIRNET - a computer program for building airflow network modelling
    National Institute of Standards and Technology (US) Report NISTIR 89-4072, April 1989
    [7]Walton G N
    CONTAM93 User Manual
    National Institute of Standards and Technology, (US) Report NISTIR 5385, 1993.
    [8]Perera M D A E S, Walker, R R, Hathaway M B, Oglesby O D, Warren P R
    Natural ventilation in large and multicelled buildings: theory, measurement and prediction
    CEC Report EUR 10552 EN, 1986
    [9]Pelletret R, Feustel H E
    IEA Annex 23: Multizone air flow and pollutant transport modelling
    IEA ECB&CS
    [10]Shih T M
    Numerical heat transfer
    Hemisphere Publishing Corporation (1984)
    [11]Kurabuchi T, Fang J B, Grot R A
    A numerical method for calculating indoor airflows using a turbulence model
    NIST Report, R89-4211 (United States), 1990.
    [12]Baker A J, Williams P T, Kelso R M
    Numerical calculation of room air motion - part 1: Maths, Physics, and CFD Modelling
    ASHRAE Trans, Vol 100, Pt 1, 1994.
    [13]Moser A
    The message of Annex 20: air flow patterns within buildings
    Proc. AIVC 12th Conference, Air Movement and Ventilation Control within Buildings, 1991.
    [14]Moser A
    The IEA work on guidelines for the ventilation of large enclosures
    Proc BEPAC (UK) Conference, Building Environmental Performance - Facing the Future, 1994.
    [15]Kendrick J F
    An overview of combined modelling of heat transport and air movement
    Technical Note 40, AIVC, 1993.

    Further reading: