4 Modelling building integrated heating and cooling
This chapter describes the simulation programs used in this thesis. A thorough description of the numerical models used in the thesis can be found in Appendix A. Therefore, this chapter only outlines the implemented models, the possibilities of the use of the models as well as the implications of using them for the analyses in the thesis.
4.1 General introduction to simulation programs
Finally, one more program is mentioned. This is called the International Building Physics Toolbox (IBPT), which has been developed in cooperation with Chalmers University of Technology and the Section of Building Physics and Services at the Technical University of Denmark. The program is a common platform for modelling building physics problems using a programming language with a graphical user interface. IBPT is also a modular program that has been developed to be a simple way to creating complex models through an extensive reuse of programming code. The purpose of this has been to demonstrate a different approach to creating simulation programs using open source code and downloading through the
internet. In exchange for the use, newly developed modules can be returned to the “editors” of the webpage and included in future versions of the program.
4.1.2 Programming elements
As stated above, the programs used in this work are modular. The modules and their
implementation are briefly mentioned in this section, with a detailed description in Appendix A including the system of equations used in the models.
Dynamic simulation models of building integrated heating and cooling systems should include:
– Dynamical implementation of temperature and heat flows in the constructions. This means that the model must include heat storage in order to include the effect of the time lag.
– Multidimensional models are required in order to correctly include the temperature distribution.
– Detailed radiation exchange between internal surfaces
– Inclusion of distribution of solar radiation on internal surfaces – Inclusion of pipes if hydronic systems are modelled
– Control systems
– Ground volume, preferably in two or three dimensions (only applicable for floor heating, not for thermo active components)
Based on this list of requirements, FHSim and TASim have been designed to consist of the elements, which have been described in Table 4.1.
Table 4.1 Modules in the simulation programs FHSim and TASim
Module Description Building integrated
heating and cooling systems
The models of building integrated heating and cooling systems used in the analyses are based on different level of detail and different
implementation of the heating/cooling coil.
Generally two different modelling types have been used:
- Finite control volume (FCV) method, and
- Thermal network based on a resistance/capacitance (RC) model.
The implementation of the FCV models is either one- or two-dimensional with “many” nodal points, whereas the RC models are models with only a few nodes, where the heat capacity of the construction is lumped.
Module Description
The models are in all cases dynamic.
The boundary conditions towards the room reflect the requirement of including radiant heat transfer, since this is the main heat transfer from the construction. Therefore, the boundary conditions treat convection and radiation separately.
The different models used are described in detail in Chapter 5 for floor heating systems and Chapter 6 for thermo active components.
Building elements (i.e. walls, ceilings, inner walls, floors, etc.)
The models of the building elements excluding building integrated heating and cooling systems are one-dimensional implementations of the FCV method.
The boundary conditions are combined with both radiation (long and short wave) and convection.
The following should be noticed for the models.
- The ceiling can face either outdoor conditions or another room.
- The inner walls are assumed to have an adiabatic boundary condition in the middle of the wall.
- The walls are facing outside conditions.
Windows including
solar gains The window model is split into a transparent and an opaque part corresponding to glazing and frame.
In the glazing model the transmission and absorption of the solar radiation is based on the incident angle. The heat transfer is modelled by a finite control volume model with four nodal points; one in each glass layer and one on each of the external surfaces. Here, the heat balance is influenced by the absorbed solar radiation in each of the glass layers. A dynamic lumped model is used.
The frame is also a dynamic lumped model with four nodal points, two internally in the frame and two surface nodes.
Solar shading is possible by introducing a reduction factor of the solar gain, which can be controlled by the heat flux on the external surface.
Room model The room model is a single node air volume with convective heat exchange with the surrounding surfaces. The air node also has heat gains from the convective parts of the internal sources and incoming solar gains. The solar gains are assumed to have a convective part coming from the part of the radiation which is absorbed by surfaces with low thermal capacity like for instance furniture which is therefore directly transferred to the room air by convection.
Further, the room model includes radiation exchange between internal surfaces based on view factors between the surfaces. The surfaces also receive solar radiation. The distribution of the solar gains to the
surfaces is distributed evenly on the surfaces with the same heat flow density (W/m²) on all surfaces. The ceiling surface does not receive
Module Description any solar radiation.
In the room model, the room air temperature is calculated based on the energy balance of the convective gains and losses. The room air is assumed to be at a steady-state condition in each time-step, thereby finding the equilibrium temperature for the room air temperature. This means that the heat storing capability of the room air and furniture is excluded from the model.
Ventilation A simple balanced mechanical system with or without heat recovery is used. The effectiveness of the heat recovery unit is based on the temperature efficiency.
Infiltration The infiltration rate is constant and equal to the exfiltration rate.
Venting The windows can be opened to avoid overheating by giving a larger air change rate with the surrounding temperature. A method is used with a variable air change rate to ensure that the room air temperature is kept below the maximum set point value. A maximum allowed air change rate due to venting is included in the model. Therefore, if this
maximum rate is exceeded or if the outdoor temperature is higher than room air temperature, the room air temperature will rise above the allowed maximum value.
Control systems and strategies
Generally an on/off type control system with a dead band is used. This is based on the fact that the thermally heavy constructions will not (or only to a small degree) benefit from more advanced controls.
Different control strategies of the supply temperature have been investigated.
Thermal comfort Thermal comfort including mean radiant temperature, operative temperature, PMV and PPD, and radiation asymmetry.
Weather data Weather data based on the Danish Design Reference Year – or for validation purposes; based on measured data.
The weather data based on the design reference year are hourly data for outdoor temperature, sky temperature, air velocity and direct and diffuse solar radiation.
In section 4.2 and 4.3 the programs FHSim and TASim are described with respect to the context in which they are used.
The following general limitations have been used in the programming, some of which have already been mentioned in Table 4.1:
- Constant material properties. This means that the influence of moisture and temperature on the material properties have been omitted in the analysis. This is mainly important for the ground model with floor heating. The effect of neglecting the coupling is discussed in Chapter 3.
- Only one room air node has been included for the air in the room model, which means that it is not possible to model the vertical temperature distribution in the room. This is
however expected to be small for a room with heated/cooled surfaces since there is less air movement due to convection.
- Equilibrium room air temperature calculated each time step. Therefore the storage capacity of the room air and furniture is neglected. This means that the room temperature will change instantly with solar gains and ventilated outside air and therefore the damping on the temperature fluctuations from the heat capacity is not included.
- Only one zone in the model. Therefore the influence on multiple rooms with individual heating circuits and set points cannot be included.
- Fairly simple inclusion of the air handling in the model with respect to infiltration, ventilation and venting.
In other words, the programs used in this work are modular with models for walls (including solar radiation), ceiling, floor, ventilation, room, and weather data. The single zone room model includes detailed calculation of radiation exchange between internal surfaces based on view factors, which is important when modelling floor heating, as the room is heated mainly by radiation. Walls, ceiling, floor and windows are modelled using a finite control volume method with an implicit solution scheme. Except for the two-dimensional floor construction, the models are one-dimensional. The ventilation system is a simple balanced system
optionally with heat recovery. As input, measured data or weather data from a design reference year can be used.
4.1.3 Implementation of programs
Both FHSim and TASim have been developed in Matlab Release 13 (Mathworks, 2002).
Matlab has the great advantage that it is simple to implement simulation models since many functions are predefined, especially concerning handling of matrices and plotting of results.
This is a major advantage over for instance C/C++, where neither matrix handling nor graphs are an integral part of the program. The simulation time is expected to be slightly longer than if a lower level programming language is used. However, the implementation time is much shorter using Matlab, which therefore makes it an ideal tool for research where the code is often changed.